Journal of Communication

Kruglanski, Arie, W.;Chen,, Xiaoyan;Pierro,, Antonio;Mannetti,, Lucia;Erb,, Hans-Peter;Spiegel,, Scott

doi: 10.1111/j.1460-2466.2006.00285.xpmid: N/A

Abstract This study describes a unimodel of persuasion offered as an alternative to the dual-process models (the elaboration likelihood model and the heuristic-systematic model) in which the term persuasion has been understood over the past two decades. The unimodel asserts that information previously referred to as (peripheral or heuristic) cues and that referred to as message or issue arguments are functionally equivalent in the persuasion process: Both serve as an evidence for conclusions that recipients may reach, and neither is necessarily privileged by recipients’ (low or high) processing resources. The unimodel highlights several parameters of persuasion that have been relatively neglected in prior formulations. Attention to those parameters serves to integrate previous persuasion models and to offer suggestions for tailoring health communications about cancer-related issues. How many fundamentally different ways are there to being persuaded? Two, according to views prevalent in social psychology today. One way is thorough and based on a meticulous consideration of message and issue information. It represents the “high road” to persuasion afforded only under ample processing resources. The other way is relatively superficial and based on information exogenous to the message or the issue. This is a “low road” resorted to when the recipient’s processing resources are limited. Two influential models, differing in conceptual details, feature such a dual-mode approach to persuasion, Petty and Cacioppo’s (1986) elaboration likelihood model (ELM) and Chaiken’s (e.g., Chaiken, Liberman, & Eagly, 1989) heuristic-systematic model (HSM). Over the past 2 decades, both models inspired considerable research generally viewed as supportive of their implications (for reviews, see Eagly & Chaiken, 1993; Petty & Cacioppo, 1986). In this study, we offer an alternative interpretation of this evidence and consider persuasion from a different conceptual perspective referred to as the parametric “unimodel” (Chun, Kruglanski, & Spiegel, 2002; Erb et al., 2003; Kruglanski, Dechesne, et al., in press; Kruglanski, Erb, Pierro, Mannetti, & Chun, in press; Kruglanski & Thompson, 1999a, 1999b; Pierro, Mannetti, Kruglanski, & Sleeth-Keppler, 2004). The dual-mode models of persuasion The ELM The ELM postulates two, qualitatively distinct, routes to persuasion taken in different circumstances. It further assumes a “continuum of elaboration likelihood running from total absence of thought about issue relevant information available in a persuasion situation to complete elaboration of all the relevant information” (Petty, 1994, p. 229, italics ours). Extensive elaboration of message or issue information represents persuasion via the central route, and reliance on cues irrelevant to the message or the issue represents persuasion via the peripheral route. According to ELM’s notion of “multiple roles of variables” in persuasion, the same variable could serve as a cue under some conditions and as an issue argument under other conditions. For instance, if an expert physician recommended a given diet, expertise would represent a “cue” based on the general notion that “experts are correct” and having little to do with the diet as such. However, if an expert physician recommended the medical school from which she graduated, expertise would represent an “argument” based on the notion that the school that produced such an expert must be good. Finally, the ELM assumes that processing via the central route may be biased (e.g., by cues) when elaboration likelihood is high (Petty, Wheeler, & Bizer, 1999, p. 159). The HSM The HSM depicts systematic processing as a comprehensive analysis of “all informational input … in forming judgments” (Chaiken et al., 1989, p. 212). Heuristic processing, in contrast, relies on “… simple inferential rules” (Chaiken et al., 1989, p. 213). According to the HSM, systematic processing “… both demands and consumes cognitive capacity, whereas heuristic processing makes few capacity demands …” (Eagly & Chaiken, 1993, p. 328). Accordingly, systematic processing requires greater degree of effort and hence of motivational involvement than heuristic processing. Like the ELM, the HSM too views the distinction between their respective dual modes as “not merely quantitative, … but qualitative” (Eagly & Chaiken, 1993, p. 213). Like the ELM too, the HSM assumes that when recipients are willing to invest effort in systematic processing, the effects of such processing may often override the effects of heuristic processing if the two processing modes have contrasting implications. Finally, heuristic processing is assumed capable of biasing systematic processing. The foregoing concludes our “thumb nail” sketch of some of the ELM and HSM major postulates. For more extensive reviews, see Petty and Cacioppo (1986) and Eagly and Chaiken (1993, chap. 7). Empirical evidence for the dual-process notions To a large extent, support for the dual-mode notions consisted of demonstrations that cues interacted with motivation and capacity factors in exactly the opposite way as message arguments. Thus, in a classic persuasion experiment, Petty, Cacioppo, and Goldman (1981) issue relevance (determining the recipients’ motivational involvement in the topic) produced the opposite persuasive effects in conjunction with source expertise than in conjunction with message argument quality: Whereas argument quality was the more important determinant of persuasion under high (vs. low) personal relevance, source expertise was the more important determinant under low (vs. high) relevance. These interactive results and a plethora of similar findings (for reviews, see Eagly & Chaiken, 1993; Petty & Cacioppo, 1986) seem consistent with the dual-mode predictions. But the question remains what is it exactly that critically differentiates cue (or “heuristic”) information from “message argument” information in a way that would explain the results of persuasion studies. The purpose of this study is to address this particular question and discuss the way that this matters when we apply persuasion theory to cancer communications to make them more effective. In the Petty et al. (1981) procedure, cue information about source expertise (a) was presented to participants prior to the message arguments and (b) was considerably briefer and less complex than message argument information. This raises the possibility that the cue information was much easier to process in this case than was the message argument information. In other words, the reason why message arguments in the Petty et al. research had greater persuasive impact under high (vs. low) issue involvement may have been that they were lengthier and more complex and appeared later in the informational sequence. All these might have reduced the likelihood that they would be thoroughly processed when participants’ processing motivation was low (in the low personal involvement condition). For a parallel reason, the brief, easily processed, cue information in the Petty et al. study may have enjoyed a persuasive advantage in the low personal relevance condition. The presentation of “cue/heuristic” information briefly and upfront and of message argument information subsequently and more lengthily is not unique to Petty et al.’s (1981) research. We conducted a systematic review of all the pertinent papers in major social and personality journals, manipulating both cue and message information, that we could identify. Using the psychINFO database, we entered as search terms the names of authors identified with the dual-mode theories (Petty, Cacioppo, Chaiken, Eagly, and others). We ended up with 19 articles, 10 studies conducted from the ELM perspective and 9 from the HSM perspective. Two independent observers, blind to our hypothesis, read the method sections of these articles and rated the order, length, and complexity of the cue information and of the message information. Interrater agreement was determined for each article. Both raters agreed 100% of the time regarding the relative order, length, and complexity of the information (figure available from author upon request); their judgments differed only in magnitude on a 5-point scale. In 100% of the cases, the cue information was judged as easier to process than the message information. In 17 out of the 19 cases (89%), the cue was judged as shorter than the message information; this includes five studies where the cue had no length as such (e.g., it constituted “head movements,” or “the audience’s response”), but its processing was judged to take less time than processing the message, and in 15 of the 19 cases (79%), the cue appeared before the message information. It appears then that in this research the statistical interactions between cue information and message arguments on the one hand and factors that moderate processing effort like issue involvement or distraction on the other hand may have arisen due to the information’s length/complexity and ordinal position. At a minimum, we discovered a pervasive confounding of cue versus message arguments information with ordinal position and length. Yet, in principle, heuristic–cue information (e.g., information about the source) and message arguments need not systematically differ as far as the effort involved in their processing is concerned. Message arguments may be presented in a clear or succinct form requiring little effort to decipher or in an oblique form replete with irrelevant detail that may require considerable effort to digest. Similarly, persuasively relevant information external to the message can be presented briefly and succinctly or in a lengthy and complex manner. In fact, the notion that cues need not be simple and straightforward but could be elaborate and complex was explicitly emphasized by Petty and Cacioppo (1986, p. 130). If prior persuasion research has confounded the cue versus message arguments categories with informational length and complexity, if message arguments need not be lengthier or more complex than cues, and if previously reported differences between cue and message argument effects may have derived from their length/complexity differences in prior research—controlling for those differences should eliminate the differential cue versus arguments effects as well. Subsequently, we shall present empirical evidence for this assertion, but first, it may be well to articulate an alternative theoretical framework of persuasion that guided our research. Because this framework highlights the functional equivalence of cues/heuristics and message arguments in persuasion, it has become known as the “unimodel” (e.g., Erb et al., 2003; Kruglanski & Thompson, 1999a, 1999b; Kruglanski, Dechesne, et al., in press; Kruglanski, Erb, et al., in press). The unimodel The unimodel is based on Kruglanski’s (1989, 2004)lay epistemic theory of knowledge formation, henceforth referred to as LET. Persuasion concerns the formation or change of subjective knowledge in the form of beliefs, opinions, or attitudes. Hence, the LET should be highly relevant to persuasion. According to LET, the knowledge formation process consists of hypothesis testing and inference during which persons acquire beliefs on the basis of relevant evidence. More specifically, our LET-based model highlights the following basic parameters of the persuasion process: (a) the structure of evidence whereby conclusions are reached, (b) cognitive ability factors affecting the extent (or depth) as well as the direction of the inferential activity, and (c) motivational factors also potentially affecting its extent and direction. The concept of evidence Pivotal to our “story” is the concept of evidence. The term “evidence” refers to information relevant to a conclusion. “Relevance” implies a prior assumption of a linkage between general categories, such that affirmation of one in a specific case (i.e., observation of the evidence) affects one’s belief in the other (i.e., warrants a conclusion). Such prior linkage is mentally represented in the knower’s mind and constitutes a belief to which he or she subscribes. This notion of evidence is the integrative glue combining the dual modes of persuasion into one. Specifically, the distinction between heuristic (or peripheral) cues and message arguments is now assumed to represent a difference in the contents of evidence relevant to a conclusion rather than a qualitative difference in the persuasive process as such. Consider a communication by Dr. Smith, a noted oncologist, whereby “taking hormonal therapy (for example, tamoxifen or an aromatase inhibitor) for 5 years lowers by half a woman’s with a breast cancer history risk of the cancer’s recurrence in the same breast, therefore this treatment ought to be adopted.” This argument may seem as persuasive evidence to a recipient whose background knowledge included the assumptions that “if something contributes to the reduction of a cancer’s recurrence (then) it should be adopted.” This may be considered the major premise of a syllogism. Dr. Smith’s specific argument supplies the minor premise that “taking hormonal therapy for five years reduce the risk of recurrence.” In other words, his or her pronouncement constitutes evidence that, granting the major premise, warrants the conclusion that (the option of)“five years of hormonal therapy should be adopted.” Such orderly and logical processing of a message argument from evidence to conclusion has been typically considered the hallmark of persuasion by the “systematic” or the “central” route. But consider, by contrast, a recipient who did not know that “there is a chance of breast cancer recurring.” On the other hand, such a person might strongly subscribe to the assumption that “if an opinion is offered by an expert, (then) it is valid.” Again, this assumption may serve as a major premise of a syllogism, and the realization “Dr. Smith is an expert” may serve as a minor premise, hence providing evidence that (granting the major premise) leads to the conclusion “Dr. Smith’s opinion (that the treatment option of taking hormonal therapy for 5 years for a woman with breast cancer history should be adopted) is valid.” Such reliance on the source attributes (like expertise) has been typically considered characteristic of persuasion accomplished by the “peripheral” or the heuristic route. Yet from the present perspective, the two persuasion types (i.e., via the central/systematic and the peripheral/heuristic routes) share a fundamental similarity in that both are mediated by “if then” or syllogistic reasoning from evidence to conclusion. In other words, both cues/heuristics and message arguments (and/or thoughts these evoke in recipients’ minds) are functionally equivalent in the persuasion process: Both function as forms of evidence from which the recipient may draw her or his conclusions. The subjectivity of evidence The foregoing analysis has important implications. For one, it suggests that evidence is “in the mind of the beholder.” A given information (whether of the cue or the message argument variety) would serve as evidence only for those recipients for whom it matched the antecedent term (X) in a conditional “if X then Y” premise to which they subscribed. Thus, different persons might find the same information more or less convincing, depending on the degree to which they subscribed to the premise appropriately matched by the information, and there should be no universal differences in persuasiveness of different types of information. Thus, for some recipients, a given cue may be more persuasive than a given message argument, whereas for others, the argument would be more persuasive than the cue. Recent work on “epistemic authority” (Kruglanski et al., 2005) indeed suggests that such cues as the communicator’s role (e.g., as parent) or expertise status (e.g., as teacher) may constitute an overwhelmingly persuasive evidence, more persuasive in fact than most arguments the communicator may be able to come up with. Second, the present analysis offers a theoretical interpretation of the (heretofore intuitive and empirically derived) notion of “argument strength” in terms of the degree of the recipient’s belief in a given “if X then Y” premise. From this perspective, the argument X will be “strong” for recipient A if A believed in the premise “if X then Y”firmly. Argument X will be weaker if A believed this premise less strongly, and so forth. The appreciation hypothesis The foregoing, highly schematic, depiction of the concept of evidence conceals the considerable amount of cognitive work often involved in constructing the evidence from the various bits and pieces available to the recipient in a given persuasive setting. The evidence may have to be gleaned from a thicket of informational detail in which it is “imbedded.” Furthermore, the relevant major premise that lends evidence its relevance has to be retrieved from memory or made accessible beyond some functional threshold of activation. Such gleaning and retrieval processes often can be quite painstaking and laborious. It is here where motivation and cognitive capacity considerations come in: If the information is lengthy, complex, or unclear, the “distillation” of intelligible evidence may require considerable amount of processing motivation or capacity. Similarly, if the audience’s motivation and/or capacity is low, only relatively straightforward evidence will register and exert persuasive impact. In more general terms, appreciating the relevance of the information given to specific conclusion depends on the relation of task demands (determined, e.g., by the information’s length, complexity, and order of appearance) to the recipients’ (cognitive and motivational) processing resources. Appreciation of the information’s relevance, and hence the difference in persuasion based on more versus less relevant information, would be greater if the processing resources were ample (vs. restricted), given the demands of the information-processing task that the persuasive setting may present. Research in the unimodel paradigm Studies conducted within the unimodel paradigm related to the processing difficulty, motivation, and capacity parameters of our framework (see Erb et al., 2003; Kruglanski & Thompson, 1999a, Pierro, Mannetti, Erb, Spiegel, & Kruglanski, 2005). Some studies tested the appreciation hypotheses described above, whereby information that is relatively difficult to process exerts greater persuasive impact under high (vs. low) processing motivation (or cognitive capacity), whereas information that is relatively easy to process exerts greater persuasive impact under low (vs. high) processing motivation (or cognitive capacity); under these conditions, it will not have to compete for influence with the difficult to process information. Processing difficulty and motivation (issue involvement) In one study (Kruglanski & Thompson, 1999a, Study 4), brief arguments were followed by lengthy arguments. Under low involvement, the brief arguments affected judgments, whereas the lengthy ones did not. Under high involvement, however, it was the lengthy arguments that affected judgments but not the brief arguments. It is noteworthy that the brief initial arguments mimic here the typical effects of cue information under low elaboration likelihood conditions (cf. Petty et al., 1981), whereas the subsequent lengthy arguments replicate here the typical message arguments effects obtained in prior persuasion research (cf. Petty et al., 1981). Another study (Pierro et al., 2005, Study 1) substituting issue involvement for cognitive load showed that, consistent with the unimodel, when the source expertise information was brief, the difference between the expert and the inexpert conditions was significant under low involvement but not under high involvement. By contrast, when the source information was lengthy, the difference between the expert and the inexpert conditions was significant under high involvement but not under low involvement. In this case, effects of the lengthy source information mimic those of the message argument information in prior research. Similarly, effects of the brief source information replicate those of the typical source information in prior research. These results are compatible with the unimodel’s implication that what matters is the difficulty of processing information rather than the type of information processed (e.g., whether classified as a cue or as a message argument). Processing difficulty and cognitive capacity In addition to varying processing difficulty and motivation, we tested the interactive effects of processing difficulty and cognitive capacity. Whereas previous research (Petty, Wells, & Brock, 1976) demonstrated a decreased elaboration of message arguments under limited cognitive capacity, we sought to explore whether cognitive load should also impair the elaboration of equally lengthy and complex expertise information. To examine this possibility, Kruglanski and Thompson (1999a, Study 2) presented all participants with lengthy expertise information (a one-page curriculum vitae, conveying that the communicator was expert or inexpert), which was followed by an equally lengthy set of message arguments. Orthogonally, we manipulated cognitive load. Participants in the no-load condition were persuaded more by the expert than by the inexpert source, whereas those under load were not persuaded differentially as a function of communicator expertise. It would seem then that to be adequately processed, relatively lengthy and complex “heuristic information” (about the source) requires sufficient cognitive capacity. As predicted by the unimodel, the heuristic information appears to function identically to message argument information of comparable length and complexity, which also is processed less effectively when recipients’ mental capacity is taxed (Petty et al., 1976). Rule activation The following studies tested the notion that persuasive impact of message information will be enhanced by the activation of the appropriate inference rules that facilitate the processing of such information by lending it subjective relevance. Thus far in the persuasion literature, the notion of rule activation was reserved for the processing of heuristic information assumed to proceed in a “top-down” fashion (Chen & Chaiken, 1999; Eagly & Chaiken, 1993) and was implied to be irrelevant to the processing of message information assumed to proceed in a “bottom-up” fashion (Bobrow & Norman, 1975). According to the unimodel, rule activation should play the same role in the processing of message contents as it does in the processing of heuristic cues external to the message. In the case of heuristics, the heuristic cue, for example, that the message is “long,” represents a minor premise for which the rule “length equals strength” constitutes the major premise, jointly affording the conclusion that the lengthy message is compelling and should command agreement. The same process was expected to take place when dealing with message information. In a study designed to explore this possibility, we presented participants with a message extolling the virtues of a small school. Orthogonally, we activated the rules “less is better” or “more is better” in the context of a prior task. This priming manipulation resulted in a preference for a small school in the “less is better” versus “more is better” condition in a message describing its virtues. Similar results were found in a subsequent experiment (Erb, Fishbach, & Kruglanski, 2002, Study 2) where the “small is likable” and the “big is likable” rules were primed. These findings support the notion that the persuasiveness of message arguments (like the persuasiveness of heuristic cues) is positively affected by the activation of rules that lend relevance to the message contents, reducing the difficulty of processing the messages, and thus enhancing persuasion. Relevance override Persuasion studies often find that peripheral or heuristic cues exert their persuasive effects under low processing motivation or cognitive capacity, whereas message argument information does so under high motivation and capacity conditions (see, e.g., Maheswaran & Chaiken, 1991; Petty et al., 1981). Whereas the differences in length/complexity of cues versus message arguments in typical persuasion research may explain why the message arguments did not have much persuasive impact under low capacity and motivation conditions, it is unclear why the cues did not exert persuasive impact under high capacity conditions. Inspired by Petty’s (1994) insight, Pierro et al. (2004) have shown that in a large proportion of studies in the persuasion literature, the cues are perceived as less relevant to the attitudes measured than are the message arguments. This led Pierro et al. to formulate a relevance override hypothesis whereby the impact of information depends on its relative relevance to the attitude issue, as well as on its ordinal position and the recipient’s cognitive and motivational resources. Specifically, when the early information is less relevant to the attitude issue than the late information (defining an irrelevant–relevant sequence), the early information should exert impact under low capacity and motivation conditions but not under high capacity and motivation conditions where its effect should be overridden by the more relevant information that may be carefully processed under these ample resource conditions. This would explain why in most persuasion studies, the cues did not exert a persuasive effect under high motivation and capacity conditions. Our analysis also implies that when the early information is more relevant to the attitude issue than the later information (defining the relevant–irrelevant sequence), it will exert persuasive effect irrespective of resource (capacity and motivation) conditions. Under low resource conditions, it will be persuasive because it is the only information the recipient is able to process, and under high resource condition because its implications would override those of the subsequent, less relevant information. These notions were tested in three studies by Pierro et al. (2004). The results from all three experiments provided supporting evidence that more relevant information, irrespective of its type (it constituting either cues or “arguments”), had a persuasive impact if it appeared early regardless cognitive or motivational resources, or if it appeared late but under high (not low) capacity and motivation conditions. Biasing effects Early information can bias the processing of subsequent information provided the individual had sufficient motivation to process the latter after processing the former information. Both the ELM and the HSM hold that central route or systematic processing can occasionally be biased by heuristic or peripheral cues (Bohner, Chaiken & Hunyadi, 1994; Bohner, Ruder, & Erb, in press; Chaiken & Maheswaran, 1994; Darke et al., 1998; Mackie, 1987; Petty, Schuman, Richman, & Strathman, 1993). Significantly, within the dual-process models, the biasing hypothesis is (implied to be) asymmetrical: It is the heuristic or peripheral cues that are presumed to bias the subsequent processing of message information but not vice versa. The reason for the asymmetry is obvious: Because in prior persuasion studies, cues typically appeared before the message arguments, it does not make much sense to ask whether their processing might be biased by the (central or systematic) processing of the subsequent message arguments. But the unimodel removes the constraint on processing sequence; hence it affords the question of whether any information type might be biased by preceding information, providing one had sufficient motivation to even process the later appearing information. How might such a process work? Simply, the early information could make accessible certain conclusions serving as evidence for further inference rules in whose light the subsequent information might be interpreted (Higgins, Rholes, & Jones, 1977). We conducted two experiments to test this idea (Erb, Pierro, Mannetti, Spiegel, & Kruglanski, 2004). In the first, we looked at the biasing effects of early message arguments on processing subsequent message arguments and found that attitudes toward aspects highlighted in the subsequent arguments was biased by the quality of the initial message argument under high, but not low, processing motivation. In the second study, we looked at biasing effects of early message arguments on processing subsequent source information under high processing motivation and found that the perceived expertise of the communicator was biased by initial argument’s quality that in turn mediated the attitude judgments. The unimodel and cancer communication Cancer communication is the study and application of strategically designed messages delivered through selected media to convey relevant health information to targeted audiences (of health care consumers, cancer survivors, health care providers, researchers, patients, at-risk populations, and others) to promote cancer prevention and control, encourage cancer screening and early detection activities, reduce cancer morbidity and mortality, and enhance the quality of life (Kreps & Viswanath, 2001). Effective communications can influence people to engage in behaviors that will improve their health, such as stopping smoking or undergoing screening for certain types of cancer. In applying persuasion theory to the realm of cancer communication, several authors have noted the importance of tailoring the message to specific characteristics of the recipients (see Kreps & Chapelsky Massimilla, 2002, for a review; Rimer & Kreuter, in press). Tailored communication efforts use data gathered about individual characteristics to design and present personalized health promotion messages to individuals that match each person’s unique background and orientation (Rimer & Glassman, 1998). Tailored communication techniques such as tailored print communications (TPCs) and telephone delivered interventions have been found effective in promoting and fostering cancer prevention, detection, and screening behaviors including smoking cessation, physical activity, mammography and cervical cancer screening, and reduction of dietary fat, cholesterol, and saturated fat (see Kreps & Chapelsky Massimilla, 2002, for a review). Though research on such tailoring techniques reported mixed results, in much of the reviewed studies tailored communications were found to be more effective than controlled communications. In Skinner, Campbell, Rimer, Curry, and Prochaska (1999) research, five of seven studies found TPCs to result in greater behavioral change than nontailored communications. A series of randomized controlled clinical trials using interactive computerized education programs tailored for patients with diabetes, asthma, and rheumatoid arthritis showed significant improvements in targeted behaviors that, in some cases, were biochemically validated (see Krishna, Balas, Spencer, Griffin, & Boren, 1997, for a review). A randomized controlled trial with structured telephone counseling regarding adherence to colposcopy showed that 20% more patients in the intervention group than in the control group adhered to the recommended cancer diagnostic follow-up and treatment (Lerman et al., 1992). In addition, Rimer and Glassman (1999) found TPCs to produce significantly raised mammography rates over the control condition. In what follows, we consider the contrasting implications for tailoring offered by the unimodel versus the dual-mode persuasion frameworks. Different treatments of source characteristics: peripheral cues versus epistemic authority beliefs Both the dual-process models and the unimodel highlight the role of motivational and cognitive resources in the processing of persuasive information. They differ, however, in the relation they see between such resources and specific content categories of information. Essentially, the unimodel posits no systematic relation between resources and content. Rather, it posits a relation between resources and task difficulty: the greater the difficulty, the more extensive the required resources. By contrast, the dual-process models do imply a relation between resources and content. Both the ELM and the HSM assume that high motivation and cognitive capacity are required for the processing of message information, whereas low motivation and capacity are needed for the processing of nonmessage (i.e., peripheral or heuristic) information. In contrast, the unimodel treats these two types of information as functionally equivalent as they may both serve as evidence for given conclusions depending on the specific beliefs subscribed to by the recipient (the major premises on which basis conclusions may be drawn). Such flexibility of “beliefs” captured by the unimodel has unique implications for cancer communication, distinct from implications of the dual-process models. An important category of such differences relates to source effects. For some individuals, source information may appear as exceedingly relevant evidence for various conclusions, perhaps more so than certain message arguments. In such a case, source information may not only have persuasive impact under low resource conditions but also under high resource conditions (Kruglanski et al., 2005; Pierro et al., 2004). Moreover, according to the unimodel, whether an individual would deign to draw conclusions from the message arguments would depend on their own self-ascribed epistemic authority as a source of information. For instance, Internet savvy, educated consumers may weigh heavily their own epistemic authority in health-related domains and expend considerable efforts on analyzing information relevant to their condition and using it to draw their own conclusions as to the best course of treatment. By contrast, persons with low self-ascribed authority may be guided primarily by their family physician’s recommendations in all medical matters. Following the logic of the dual-process models, it might be assumed that when patients’ motivation for seeking out and processing cancer information is high and their cognitive resources are ample, features that may often characterize individuals with a diagnosis of cancer, source characteristics would not matter much anymore, and these individuals’ health decisions would be primarily determined by their systematic processing of specific cancer information delivered by whatever source. But according to the unimodel, cancer patients highly motivated to form accurate conclusions may be interested in whatever information they might find that is relevant to their specific concerns. Often, this could be information about specific sources (e.g., the reputation and track record of oncologists and other health communicators). In other instances, such information might consist of specific arguments (e.g., concerning the reliability of prostate-specific antigen [PSA] testing or the effectiveness of surgical vs. radiation procedures in treating prostate cancer) that individuals may feel capable of evaluating, providing that their self-ascribed epistemic authority in the domain of such arguments was sufficiently high. We do not mean to imply that all (or most) cancer patients are under high accuracy motivation, prompting them to engage in careful and extensive processing of relevant information. At times, such individuals might experience a high need for cognitive closure (Kruglanski, 2004; Kruglanski & Webster, 1996; Kruglanski, Pierro, Mannetti, & De Grada, 2006), and they might find unsettling the uncertainty associated with the careful weighing of multiple arguments. In such a case, they might simply “seize and freeze” on early, apparently relevant information, again irrespective of whether it was of the message argument or the “peripheral/heuristic” variety. Subjective relevance A unique implication of the unimodel is that different people may vary in their cancer-relevant “lay theories” and in the sources they believe to be credible with regard to cancer-related topics. For example, some people may believe that their parents know best about what kind of diet is appropriate. Others may trust their friends in these matters. Yet, other may assign considerable epistemic authority to various celebrities, and so forth (for discussion, see Kruglanski et al., 2005). These differences in the subjective relevance of various source or message factors to individuals’ judgments could have considerable implications for the kind of tailoring that may be effective in persuading them to adopt the recommended cancer-related practices. For instance, in an antismoking campaign directed at adolescents and college-age adults, one might attribute the message to members of the peer group known to constitute a high-level epistemic authority for these individuals (Kruglanski et al., 2005). By contrast, in tailoring a similar message to older adults, one might attribute the recommendations to renowned health experts. For people who assign great epistemic authority to celebrities, celebrity endorsements through media might be most effective. There is indeed evidence that celebrity endorsement led to an increase in PSA testing, a common screening mechanism for prostate cancer (Viswanath, 2005). Katie Couric, a well-known television personality in the United State, aided the promotion of screening for colorectal cancer that led to the coined phrase about the “Katie Couric effect” (Gorman, 2000; Newman, 2001). The unimodel suggest, additionally, that the use of celebrities such as Couric might be particularly effective for audiences who assign epistemic authorities to celebrities in general, and Couric in particular. The use of other celebrities might be indicated for audiences that bestow those particular celebrities with high epistemic authority. Finally, the use of celebrities might be ineffectual for recipients who doubt their qualifications to make authoritative recommendations on cancer-related issues. Processing resources and the cancer control continuum. Experts in the domain of cancer communication (e.g., Kreps & Viswanath, 2001; Viswanath, 2005) refer to the cancer control continuum ranging from cancer prevention, screening and detection phases through the diagnosis and treatment phases all the way to the end of life phases of terminally ill cancer patients. Our unimodel has implications for message tailoring at different points on this continuum. For instance, at the cancer prevention, screening and detection phases message recipients may be healthy and relatively young; they may be, therefore, unlikely to perceive the cancer diagnosis as self-relevant and may be unlikely to engage in extensive processing of cancer-relevant information. If so, the tailoring of cancer communications to these individuals should either (a) increase the perceived self-relevance of cancer to their own lives or (b) present the cancer-related communications in an easy to process, undemanding, manner. Furthermore, persons with a positive cancer diagnosis may be under considerable stress and hence unable to engage in deep information processing. It follows that they too may require easy to follow and simple communications relevant to their beliefs. Often, family members of cancer patients may constitute a group of individuals with sufficient motivation and cognitive resources to engage in the processing of (subjectively relevant) cancer-related information. They might be less motivated than their cancer-diagnosed kin to quickly arrive at cognitive closure or escape the topic altogether. Thus, they might constitute a group of individuals to whom more intricate, specific, and complex cancer communications might be directed. Conclusion The research described earlier supports the basic contention of the unimodel that judgmental parameters such as difficulty of processing (operationalized, e.g., by length/complexity of the information given, or rule accessibility), perceived relevance to the topic (operationalized by argument quality and/or expertise information), magnitude of processing motivation (operationalized by issue involvement or need for closure), or processing sequence (i.e., early- or late-appearing information) determine persuasion. Controlling for parameter values, the contents or types of information, for example, those inherent in the message arguments versus related to source information external to the message, do not appear to have unique persuasive effects as such. The unimodel offers a generalized alternative to the prevalent dual-mode approaches. At the same time, it shares some fundamental commonalities with dual-process models, and in many cases, it grows out of insights and arguments developed by the dual-process theorists. In this sense, it is perhaps better though of as an evolution in our thinking about persuasion, rather than representing a revolutionary break. Nonetheless, the unimodel affords three fundamental advantages to persuasion researchers. These are the advantages of (a) parsimony, (b) generative potential, and (c) a novel, flexible view of persuasion of potentially considerable practical significance. The parsimony advantage resides in the synthesis the model affords of contemporary persuasion work (Chaiken et al., 1989; Petty & Cacioppo, 1986) that stresses the role of cognitive and motivational resources with the work conducted in the 1970s on the logical structure of reasoning (McGuire, 1960; Wyer, 1970) and more generally on the function of evidence in opinion change. Perhaps of equal importance, it integrates current persuasion work with alternative social judgment phenomena in domains of impression formation (Brewer, 1988; Fiske & Neuberg, 1990), causal reasoning (cf. Chun, Spiegel, & Kruglanski, 2002; Trope & Alfieri, 1997), or indeed statistical reasoning that, according to the unimodel, are affected identically by the very same judgmental parameters that determine persuasion (see Chun & Kruglanski, in press; Erb et al., 2003; Kruglanski, Erb et al., in press). Beyond parsimony, the unimodel possesses considerable generative potential for subsequent research. The research described earlier manipulated length and ordinal position not only of heuristic but also of message argument information, manipulated the accessibility of both types of information, and we are beginning to cross these variables with a variety of persuasively relevant factors beyond involvement, such as the need for closure. Future research could also look at other dependent variables beyond attitude change per se, for example, the persistence of change, resistance to counterpersuasion, and strength of the attitude behavior relations. According to the unimodel, those phenomena are the consequences of deep (or elaborate) processing of any information, rather than the elaborate processing of message argument information exclusively (as implied by prior dual-process formulations). On a practical level, the unimodel offers exciting new possibilities to potential communicators. It suggests that brief and superficial persuasion may be accomplished via simplistic message arguments as well as via message-unrelated cues or heuristics. More importantly, perhaps, it implies that even where recipients are unfamiliar with the topic at hand and hence unable to appreciate issue-related arguments, it may be still possible to produce profound attitude change via heuristic evidence (e.g., about source expertise) that recipients may incorporate into their attitudes. In this sense, the unimodel offers a flexible view of persuasion in which the same persuasive outcomes may be attained via different means depending on the particular epistemic characteristics of persuasion recipients. Persuasion plays a particularly significant role in the prevention, early detection, and treatment of cancer and ultimately the survivorship of cancer patients. The present flexible view of persuasion suggests the importance of identifying the specific belief structures characterizing various targeted populations (the kind of “if X then Y” premises their members subscribe to) to which cancer-relevant informational campaigns are addressed in order to determine which information they would find compelling. It also suggests that importance of considering the contexts in which the recipients are likely to encounter the informational campaigns, including the phase they are at on the cancer control continuum (Viswanath, 2005) and tailoring their cancer communications by taking into account cognitive and motivational resources likely to be at the recipients’ disposal at the moment they encounter the cancer-relevant messages. These considerations and others suggest to potential utility the integrative view of persuasion offered by the unimodel in approaching the task of designing effective communications aimed at improving individuals ability to prevent, detect, and treat various types of cancer. Acknowledgment This work was supported by NSF grant SBR-9417422. References Bobrow , D. G. , & Norman , D. A. ( 1975 ). Some principles of memory schemata. In D. G. Bobrow & A. M. Collins (Eds.), Representation and understanding: Studies in cognitive science . New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bohner , G. , Chaiken , S., & Hunyadi , P. ( 1994 ). The role of mood and message ambiguity in the interplay of heuristic and systematic processing . European Journal of Social Psychology , 24 , 207 – 221 . Google Scholar Crossref Search ADS WorldCat Bohner , G. , Ruder , M., & Erb , H.-P. (in press). When expertise backfires: Contrast and assimilation effects in persuasion . British Journal of Social Psychology . OpenURL Placeholder Text WorldCat Brewer , M. B . ( 1988 ). A dual process model of impression formation. In R. S. Wyer & T. K. Srull (Eds.), Advances in social cognition (Vol. 1 , pp. 1 – 36 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chaiken , S. , Liberman , A., & Eagly , A. H. ( 1989 ). Heuristic and systematic processing within an beyond the persuasion context. In J. S. Uleman & J. A. Bargh (Eds.), Unintended thought (pp. 212 – 252 ). New York : Guilford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chaiken , S. , & Maheswaran , D. ( 1994 ). Heuristic processing can bias systematic processing: Effects of source credibility, argument ambiguity, and task importance on attitude judgments . Journal of Personality and Social Psychology , 66 , 460 – 473 . Google Scholar Crossref Search ADS PubMed WorldCat Chen , S. , & Chaiken , S. ( 1999 ). The heuristic-systematic model in its broader context. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social and cognitive psychology (pp. 73 – 96 ). New York : Guilford Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chun , W. Y. , & Kruglanski , A. W. ( in press ). The role of task demands and processing resources in the use of base rate and individuating information . Journal of Personality and Social Psychology . OpenURL Placeholder Text WorldCat Chun , W. Y. , Spiegel , S. & Kruglanski , A. W. ( 2002 ). Assimilative behavior identification can also be resource dependent: A Unimodel-Based Analysis of Dispositional Attribution Phases . Journal of Personality and Social Psychology , 83 , 542 – 555 . Google Scholar Crossref Search ADS PubMed WorldCat Darke , P. R. , Chaiken , S., Bohner , G., Einwiller , S., Erb , H.-P., & Hazelwood , D. ( 1998 ). Accuracy motivation, consensus information, and the law of large numbers: Effects on attitude judgments in the absence of argumentation . Personality and Social Psychology Bulletin , 24 , 1205 – 1215 . Google Scholar Crossref Search ADS WorldCat Eagly , A. H. , & Chaiken , S. ( 1993 ). The psychology of attitudes . Fort Worth, TX : Harcourt Brace Jovanovich . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Erb , H. P. , Kruglanski , A. W., Chun , Y. W., Pierro , A., Mannetti , L., & Spiegel , S. ( 2003 ). Searching for commonalities in human judgment: The parametric unimodel and its dual-model alternatives. In W. Stroebe & M. Hewstone (Eds.), European Review of Social Psychology (pp. 1 – 48 ). New York : Psychology Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Erb , H. P. , Pierro , A., Mannetti , L., Spiegel , S., & Kruglanski , A. W. ( 2004 ). Biased processing of persuasive information: On the functional equivalence of cues and message arguments . Unpublished manuscript, Bonn University , Bonn, Germany . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fiske , S. T. , & Neuberg , S. L. ( 1990 ). A continuum of impression formation, from category- based to individuating processes: Influences of information and motivation on attention and interpretation . Advances in Experimental Social Psychology , 23 , 1 – 12 . Google Scholar Crossref Search ADS WorldCat Gorman , C . ( 2000 ). Katie’s crusade. Colon cancer is one of the deadliest and most preventable malignancies. What you need to know about the disease—And the surprisingly painless test that could save your life . Time , 155 ( 10 ), 70 – 76 . OpenURL Placeholder Text WorldCat Higgins , E. T. , Rholes , W. S., Jones , C. R. ( 1977 ). Category Accessibility and Impression Formation . Journal of Experimental Social Psychology , 13 , 141 – 154 . Google Scholar Crossref Search ADS WorldCat Kreps , G. L. , & Chapelsky Massimilla , D. ( 2002 ). Cancer communications research and health outcomes: Review and challenge . Communication Studies , 53 , 318 – 336 . Google Scholar Crossref Search ADS WorldCat Kreps , G. L. , & Viswanath , K. ( 2001 ). Foreword. Communication interventions and cancer control: A review of the National Cancer Institute’s health communication intervention research initiative . Family and Community Health , 24 ( 3 ), ix – xiii . Google Scholar Crossref Search ADS PubMed WorldCat Krishna , S. , Balas , A., Spencer , D. C., Griffin , J. Z., & Boren , S. A. ( 1997 ). Clinical trials of interactive computerized patient education: Implications for family practice . The Journal of Family Practice , 45 ( 1 ), 25 – 33 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Kruglanski , A. W . ( 1989 ). Lay Epistemics and Human Knowledge: Cognitive and Motivational Bases . New York : Plenum . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Kruglanski , A. W . ( 2004 ). The psychology of closed-mindedness . New York : Psychology Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kruglanski , A. W. , Dechesne , M., Erb , H. P., Pierro , A., Mannetti , L., & Chun , W. Y. ( in press ). Modes, systems and the sirens of specificity: The issues in gist . Psychological Inquiry . OpenURL Placeholder Text WorldCat Kruglanski , A. W. , Erb , H. P., Pierro , A., Mannetti , L., & Chun , W. Y. ( in press ) On parametric continuities in the world of binary either ors . Psychological Inquiry . OpenURL Placeholder Text WorldCat Kruglanski , A. W. , Pierro , A., Mannetti , L., De Grada , E., ( 2006 ). Groups as Epistemic Providers: Need for Closure and the Unfolding of Group-Centrism . Psychological Review , 113 ( 1 ), 84 – 100 . Google Scholar Crossref Search ADS PubMed WorldCat Kruglanski , A. W. , Raviv , A., Bar-Tal , D., Raviv , A., Sharvit , K., Ellis , S., et al. . ( 2005 ). Says who?: Epistemic authority effects in social judgment . Advances in Experimental Social Psychology , 37 , 345 – 392 . Google Scholar Crossref Search ADS WorldCat Kruglanski , A. W. , & Thompson , E. P. ( 1999a ). Persuasion by a single route: A view from the unimodel . Psychological Inquiry , 10 ( 2 ), 83 – 109 . Google Scholar Crossref Search ADS WorldCat Kruglanski , A. W. , & Thompson , E. P. ( 1999b ). The illusory second mode or, the cue is the message . Psychological Inquiry , 10 ( 2 ), 182 – 193 . Google Scholar Crossref Search ADS WorldCat Kruglanski , A. W. , & Webster , D. M. ( 1996 ). Motivated closing of the mind: “Seizing” and “freezing.” Psychological Review , 103 , 263 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Lerman , C. , Hanjani , P., Caputo , C., Miller , S., Delmoor , E., Nolte , S., et al. . ( 1992 ). Telephone counseling improves adherence to colposcopy among lower income minority women . Journal of Clinical Oncology , 10 , 330 – 333 . Google Scholar Crossref Search ADS PubMed WorldCat Machie , R. R ., ( 1987 ). Vigilance research – Are we ready for countermeasures? Human Factors , 29 ( 6 ), 707 – 723 . Google Scholar Crossref Search ADS PubMed WorldCat Maheswaran , D. , & Chaiken , S. ( 1991 ). Promoting systematic processing in low motivation settings: Effect of incongruent information on processing and judgment . Journal of Personality and Social Psychology , 61 , 13 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat McGuire , W. J . ( 1960 ). A syllogistic analysis of cognitive relationships. In M. J. Rosenberg, C. I. Hovland, W. J. McGuire, R. P. Abelson, & J. W. Brehm (Eds.), Attitude organization and change (pp. 65 – 111 ). New Haven : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Newman , L . ( 2001 ). Panel examines impact of public figures on cancer awareness . The Journal of the National Cancer Institute , 93 ( 1 ), 8 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R. E . ( 1994 ). Two routes to persuasion: State of the art. In G. D’Ydewalle, P. Eelen, & P. Berteleson (Eds.), International perspectives on psychological science (Vol. 2 , pp. 229 – 247 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Cacioppo , J. T. ( 1986 ). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 19 , pp. 123 – 205 ). San Diego, CA : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Cacioppo , J. T., & Goldman , R. ( 1981 ). Personal involvement as a predictor of argument-based persuasion . Journal of Personality and Social Psychology , 41 , 847 – 855 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Schuman , D. W., Richman , S. A., & Strathman , A. J. ( 1993 ). Positive mood and persuasion: Different roles for affect under high- and low-elaboration conditions . Journal of Personality and Social Psychology , 64 ( 1 ), 5 – 20 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Wells , G. L., & Brock , T. C. ( 1976 ). Distraction can enhance or reduce yielding to propaganda: Though disruption versus effort justification . Journal of Personality and Social Psychology , 34 , 874 – 884 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Wheeler , S. C., & Bizer , G. Y. ( 1999 ). Is there one persuasion process or more? Lumping versus splitting in attitude change theories . Psychological Inquiry , 10 , 156 – 163 . Google Scholar Crossref Search ADS WorldCat Pierro , A. , Mannetti , L., Erb , H. P., Spiegel , S., & Kruglanski , A. W. ( 2005 ). Informational length and order of presentation as determinants of persuasion . Journal of Experimental Social Psychology , 41 , 458 – 469 . Google Scholar Crossref Search ADS WorldCat Pierro , A. , Mannetti , L., Kruglanski , A. W., & Sleeth-Keppler , D. ( 2004 ). Relevance override: On the reduced impact of “cues” under high motivation conditions of persuasion studies . Journal of Personality and Social Psychology , 86 , 251 – 264 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. I. , & Glassman , B. ( 1998 ). Tailoring communications for primary care settings . Methods of Information in Medicine , 37 ( 3 ), 171 – 178 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Rimer , B. K. I. , & Glassman , B. ( 1999 ). Is there a use for tailored print communications in cancer risk communication? The Journal of the National Cancer Institute , 25 , 140 – 148 . OpenURL Placeholder Text WorldCat Rimer , B. K. , & Kreuter , M. W. ( in press ). Advancing tailored health communication: A persuasion & message effects perspective . Journal of Communication . OpenURL Placeholder Text WorldCat Skinner , C. S. , Campbell , M. I. C., Rimer , B. K., Curry , S., & Prochaska , J. O. ( 1999 ). How effective is tailored print communication? Annals of Behavioral Medicine , 21 , 290 – 298 . Google Scholar Crossref Search ADS PubMed WorldCat Trope , Y. , & Alfieri , T. ( 1997 ). Effortfulness and flexibility of dispositional judgment processes . Journal of Personality and Social Psychology , 73 , 662 – 674 . Google Scholar Crossref Search ADS WorldCat Viswanath , K . ( 2005 ). Science and society: The communications revolution and cancer control . Nature Reviews Cancer , 5 , 828 – 835 . Google Scholar Crossref Search ADS PubMed WorldCat Wyer , R. S . ( 1970 ). The quantitative prediction of belief and opinion change: A further test of a subjective probability model . Journal of Personality and Social Psychology , 16 , 559 – 571 . Google Scholar Crossref Search ADS WorldCat © 2006 International Communication Association

Peters,, Ellen;Lipkus,, Isaac;Diefenbach, Michael, A.

doi: 10.1111/j.1460-2466.2006.00287.xpmid: N/A

Abstract We examine potential roles of 4 functions of affect in health communication and the construction of health preferences. The roles of these 4 functions (affect as information, as a spotlight, as a motivator, and as common currency) are illustrated in the area of cancer screening and treatment decision making. We demonstrate that experienced affect influences information processes, judgments, and decisions. We relate the functions to a self-regulation approach and examine factors that may influence the weight of cognitive versus affective processing of information. Affect’s role in health communication is likely to be nuanced, and it deserves careful empirical study of its effects on patients’ well-being. A major theme that emerges from judgment and decision-making research is that we frequently do not know our own “true” values in decisions (e.g., how much we value a treatment option or better patient/physician communication). Rather, when asked to form a judgment or to make a decision, we often appear to construct our values and preferences “on-the-spot” using cues from the decision situation and our own internal reactions (Payne, Bettman, & Schkade, 1999; Slovic, 1995). This kind of on the spot value judgment can have significant implications for a person’s well-being, such as when cancer patients are asked to make treatment decisions. In health communications, the goal of communication efforts is often informed choice; decisions should be based on patients’“accurate” understanding of the facts and be consistent with patient values (Briss et al., 2004; Edwards, Unigwe, Elwyn, & Hood, 2003; Frosch, Kaplan, & Felitti, 2001; Parascandola, Hawkins, & Danis, 2002). For example, a man who elects to be screened for prostate cancer by having a prostate-specific antigen test should understand and weigh the potential positive and negative outcomes of the test (e.g., a possible cancer diagnosis, anxiety while waiting for the results, the possible need to undergo additional tests such as a biopsy related to an abnormal finding that proves not to be cancer) with the likelihood of each outcome. However, although the goal of informed choice is an excellent one, it may not attend adequately to the possibility that value judgments and preferences of patients are malleable and are sometimes constructed on the spot. As a result, informed choice may be a less realistic ideal in some situations (e.g., when how information is presented as opposed to what information is presented exerts a significant impact on choices). In other medical situations, the goal of health communication is to persuade (e.g., a 14-year-old vulnerable to smoking should be persuaded not to initiate smoking). With this goal, communicators are attempting to harness the power of constructed values and preferences. We examine the roles of affect and the affect heuristic in health communication and the construction of health preferences. This study extends earlier work on the affect heuristic by examining four functions of affect in the construction process. We first review the four functions developed by Peters (in press) and illustrate these processes with examples relating to cancer screening and treatment. Research demonstrates that affect experienced in the moment, whether the result of considering a decision option or from an unrelated source, influences concurrent cognitions, information processes, judgments, and decisions. Subsequently, we relate the functions to a self-regulation approach to health threats that includes both affective and cognitive components. Finally, we examine factors that may influence the weight of more cognitive versus more affective processing of health information. What are the functions of affect in the process of constructing judgments and decisions? Recent research has developed and tested theories of judgment and decision making that incorporate affect as a key component in a process of constructing values and preferences (Lichtenstein & Slovic, in press). Within these theories, integral affect refers to positive and negative feelings1 about a stimulus that are generally based on prior experiences and thoughts and are experienced while considering the stimulus, whereas incidental affect is positive and negative feelings such as mood states that are independent of a stimulus but can be misattributed to it or can influence decision processes. Together, they are used to predict and explain a wide variety of judgments and decisions ranging from choices among jelly beans to life satisfaction and valuation of human lives (Kahneman, Schkade, & Sunstein, 1998; Schwarz & Clore, 1983; Slovic, Finucane, Peters, & MacGregor, 2002). Mild incidental affect and integral affect are ubiquitous in everyday life. Imagine finding a quarter on the sidewalk (a mild positive mood state is induced; there may be a “carryover” of incidental affect that colors your next decision even though the feelings are not normatively relevant to the decision) or considering whether you will have a bowl of oatmeal or a chocolate croissant for breakfast (mild positive and negative affect are experienced; these are termed integral affect because they are part of the decision maker’s internal representation of the decision options). Feelings experienced in the cancer context can be more extreme. For example, a breast cancer survivor may experience a euphoric mood state when told that she is in remission or a 65-year-old man with prostate cancer may use integral affect toward his options of watchful waiting and surgery to guide his treatment choice. Whether mild or strong, these feelings can impact the processing of information and, thus, what is judged or decided. Research has begun to delineate the various ways that affect alters information processing. In this study, we argue that affect has four separable roles important to health communication and decision-making processes. First, as suggested above, affect can act as information. Second, it can act as a spotlight focusing our attention on different information—numerical cues or risks versus benefits, for example—depending on the extent of our affect. Third, affect can motivate action or the processing of information. Finally, affect may serve as a common currency in judgments and decisions, allowing us to compare more effectively the values of very different decision options or information (e.g., compare apples to oranges). These functions are not mutually exclusive, and we consider their interrelations after introducing all four functions. We will focus primarily on valence (positive and negative) feelings because this evaluation has been found essential to meaning by researchers studying such diverse phenomena as language, attitudes, and the behavior of persons with brain lesions (see, e.g., Damasio, 1994; Osgood, Suci, & Tannenbaum, 1957). Consideration of how discrete emotion states impact preferences and when valence versus discrete emotions alter preferences are also important questions but are outside the scope of this study (Lerner & Tiedens, in press; Peters, Burraston, & Mertz, 2004). Affect as information One of the most comprehensive theoretical accounts of the role of affect and emotion in decision making was presented by the neurologist, Antonio Damasio (1994). In seeking to determine “what in the brain allows humans to behave rationally,” Damasio argued that a lifetime of learning leads decision options and attributes to become “marked” by positive and negative feelings linked directly or indirectly to somatic or bodily states. When a negative somatic marker is linked to an outcome, it acts as information by sounding an alarm that warns us away from that choice. When a positive marker is associated with the outcome, we are drawn toward that option. By consulting and being guided by these feelings, Damasio claims that we make better quality and more efficient decisions. Without these feelings, information in a decision lacks meaning and the resulting choice suffers. The affect heuristic built upon these and other ideas to suggest that affect may serve as a cue for many important judgments including probability judgments (Slovic et al., 2002). Zajonc (1980) proposed that affective reactions to stimuli are often the earliest reactions. Under the affect heuristic model, affect can be experienced first upon consideration of a familiar technology or it can be the result of further processing of information (LeDoux, 1996; Peters, Västfjäll, et al., 2006). In either event, the affect then acts as information guiding decisions and judgments such as risk and benefit perceptions. The affect heuristic is substantially similar to models of “risk as feelings” and “mood as information” (Loewenstein, Weber, Hsee, & Welch, 2001; Schwarz & Clore, 2003) and shares much in common with dual process theories, such as those by Epstein (1994) and Leventhal (e.g., Cameron & Leventhal, 2003; Leventhal, 1970; Leventhal, Diefenbach, & Leventhal, 1992). Whereas some theories focus exclusively on the use of mood states (incidental affect) in judgments (Forgas, 1995,2; Schwarz & Clore, 2003), use of the affect heuristic is characterized by reliance on feelings attributed to an option or stimulus and experienced while considering it in judgments and decisions. Alhakami and Slovic (1994) proposed that the strength of positive or negative affect associated with an activity (and experienced while considering that activity) guided perceptions of its risks and benefits. Thus, judgments about a technology such as a new medical treatment would be based not only on what people think about the treatment but also on how they feel about it. If feelings toward a technology are favorable, decision makers are moved toward judging its risks as low and its benefits as high; if their feelings are more negative, they tend to judge the opposite—high risk and low benefit. For example, virtual colonoscopy is currently under much scrutiny for the detection of colon cancer. Individuals with positive affect toward this technology (the affect may be due to prior learning that it is not invasive or it is less embarrassing) may interpret new information about risks and benefits in ways that are consistent with their affect (i.e., perceive it as low risk and high benefit). In some situations, decision options are unfamiliar to the decision maker and focused communication efforts may be able to assist the individual. In these cases, providing individuals with “affective cues” may help provide meaning to the information presented. For example, Peters, Slovic, & Hibbard (2004) were interested in the processes by which decision makers might bring meaning to unfamiliar, dry, cold facts about health plan quality. In a series of studies, they influenced the interpretation and comprehension of numerical information about health plan attributes by providing affective cues that can be used easily to evaluate the overall goodness or badness of a health plan. For example, in one of the studies, older adult participants were presented with identical attribute information (quality of care and member satisfaction) about two health plans. The information was presented in bar chart format with the actual score displayed to the right of the bar chart (see Figure 1). The information for half of the subjects was supplemented by the addition of affective categories (i.e., the category lines plus affective labels that placed the health plans into categories of poor, fair, good, or excellent). Figure 1 Open in new tabDownload slide Example of affective categories in health plan choice. Figure 1 Open in new tabDownload slide Example of affective categories in health plan choice. The attribute information was designed such that Plan A was good on both attributes, whereas Plan B was good on quality of care but fair on member satisfaction. The specific scores for quality of care and member satisfaction were counterbalanced across subjects, such that for half the subjects the average quality-of-care scores were higher, and for the other half average member satisfaction scores were higher. They predicted and found that affective categories influenced the choices. Specifically, older adults preferred health Plan A more often when the categories were present (Plan A was always in the good affective category when the affective labels were present). In a second study, they found that choices of older adults with high deliberative ability (high speed of processing measured in a digit symbol substitution task) were influenced less by the presence versus absence of affective categories than were choices of older adults with low deliberative ability. The results suggest that adults with high deliberative efficiency are better able to compare information and derive meaning from the data. However, for adults with low deliberative efficiency, affective categories appear to provide more information and influence its meaning—suggesting that affective labels may be especially useful among those who are taxed by too much information, time pressure, or the stress of illness. Allowing individuals such as a recently diagnosed cancer patient access to the meaning of important information like the quality of care offered at different hospitals should help them to process it more deeply and make better decisions. A condition that included the category lines but no affective labels had no impact on choice, suggesting that categorization alone cannot explain this effect. A third study provided direct evidence of the affective basis of this manipulation by demonstrating that decision makers accessed their feelings about the health plan highlighted by affective categories (e.g., Plan A in Figure 1) faster than their thoughts in the presence but not in the absence of the manipulation. Communication techniques such as these may prove useful in improving risk comprehension when cancer-related distress is high and objective risks are underweighted (Lerman et al., 1995). Although testing in a real-world environment is absolutely necessary, if affectively based attitudes are best modified with affective as opposed to cognitive information (Millar & Tesser, 1989), then cancer risk communication techniques with an affective basis may promote better comprehension of information like objective risks that might otherwise be ignored. The use of methods such as affective categories does call, however, for a different emphasis in health and other communications. Instead of a focus on providing complete and accurate information, the emphasis shifts to providing usable, meaningful, and accurate information that will support better choices. It brings with it a new level of responsibility for health communicators (including physicians and other health professionals) who would need to know how patients currently respond to information and would need to bring their expertise to bear not only on what information to provide but also on how to present that information in ways that ethically support good decision making. It also requires a delicate balance between informing patients and helping them take into account important information that might otherwise not enter into their decision processes. The provision of more subjective interpretations may be difficult and be resisted by health professionals who prefer to provide only “objective facts.” We hope this study helps shift the focus on patient best interests from exposing them to information to helping them comprehend and use that information. Affect also acts as information in the construction of risk estimates (Johnson & Tversky, 1983; Loewenstein et al., 2001; Peters & Slovic, 1996). Peters, Slovic, Hibbard, and Tusler (2006) linked affect to an adjustment process in fatality estimates. They found that decision makers asked to estimate the number of fatalities in the United States each year from various causes of death were influenced by the anchor provided (the actual number of deaths from a disease), a typical finding in such studies. The decision makers also appeared to adjust away from the anchor based on the extent of their worry about the disease under consideration; fatality estimates were significantly higher for causes of death about which they reported greater worry. These findings are among the first to link affect as a possible mechanism underlying the adjustment process (see also Peters, Slovic, & Gregory, 2003). They are also consistent with cancer results. For example, in the breast cancer literature, worry is associated with perceived risk (Erblich, Bovbjerg, & Valdimarsdottir, 2000; Lipkus et al., 2000).3 Although the direction of this effect is not entirely clear, experimental results have demonstrated that increasing incidental negative affect subsequently increases risk perceptions across a variety of domains (e.g., Johnson & Tversky). A pervasive fear of cancer may underlie in part the tendency for individuals to exaggerate their cancer risk (Croyle & Lerman, 1999). Reliance on affect as information may create communication challenges with respect to cancer risk. Specifically, reliance on affect may result in less use of (nonaffective) numerical information, making it difficult to educate patients about their objective risk and alleviate unnecessary cancer distress (Watson et al., 1999). In one nonhealth-related study, the presence of salient affective cues led to neglect of probabilistic information (Rottenstreich & Hsee, 2001). We speculate that individuals’ often strong negative affect about cancer may create insensitivity to its (often relatively low) objective risk. In support of this notion, Kraus, Malmfors, and Slovic (1992) found that expert toxicologists were able to assess accurately the risk of cancer posed by different exposure levels to a toxic agent, but the average individual from the public tended to believe that any exposure level was risky, possibly because negative feelings about cancer created an insensitivity to varying levels of the likelihood of cancer. If the strength of affect toward an outcome causes neglect of its likelihood of occurrence as hypothesized, then probability neglect may be particularly strong among patients who have stronger negative reactions to cancer initially. Anxiety in the face of low objective risk, however, may also encourage protective behaviors to avoid increased risk (e.g., using sunscreen to avoid skin cancer). Affect as information thus acts as a substitute for the assessment of other specified target attributes in judgment (e.g., what is the objective risk of cancer from a toxic agent? Kahneman, 2003). Without affect, information appears to have less meaning and to be weighed less in cancer judgment and choice processes (e.g., numerical risk information). Communication efforts should include statements that acknowledge a person’s possible emotional reactions to a threat (e.g., it is quite normal to be worried in a situation like this) or illustrate the prevalence of the threat (e.g., prostate cancer is the most diagnosed cancer in the United States). Another example, within the context of genetic testing for breast cancer susceptibility, illustrates this approach further. To improve understanding of testing-related probabilities and decision making, women eligible for genetic testing for BRCA1/2 are led through a structured exercise in which they imagine receiving a positive, negative, or indeterminate result. In each case, they are asked to imagine what they would feel and what they would think; the researcher’s goal was to identify potential maladaptive emotional and cognitive reactions that can be addressed by the genetic counselor (Diefenbach & Hamrick, 2003). This is one of the special efforts being developed to highlight the meaning of important, but underweighted, numerical information. Affect as a spotlight Considerably less work has been done on the other three proposed functions of affect in the construction of preferences. Peters (in press) proposed that affect plays a role as a spotlight in a two-step process. First, the extent or type of affective feelings (e.g., weak vs. strong affect or anger vs. fear) focuses the decision maker on new information. Second, the new information (rather than the initial feelings themselves) is used to guide the judgment or decision. Although incidental feelings have been shown to function as a spotlight in memory and judgment (e.g., mood-congruent biases on memory; Bower, 1981), little research has examined how integral feelings about a target might alter what information becomes more salient in decisions. Nabi (2003) experimentally manipulated integral affect (anger vs. fear) toward drunk driving. Consistent with the proposed two-step approach, this affect manipulation first made some stored knowledge more accessible (e.g., reasons to be angry about drunk driving were spotlighted), and second, the more accessible information had a greater impact in subsequent preferences. Peters et al. (2003) found a similar spotlight effect with affect influencing the dollar values salient to buyers and sellers of a lottery ticket. Both level of affect toward an object and the types of affect (e.g., anger vs. fear) appear to make different cues more salient. Depending upon how people feel about an object, they may focus on different information. The function of affect as a spotlight takes advantage of the role of feelings in directing cognitions to address the source of the feeling. Alhakami and Slovic (1994) demonstrated that the negative correlation between perceived risk and perceived benefit is mediated by affect. For example, decision makers with positive affect toward a radiation source (e.g., X rays) tend to perceive it as highly beneficial and low in risk; the reverse happens if decision makers have negative affect about a radiation source (e.g., nuclear power). Although this effect has been interpreted in terms of the role of affect as information (Slovic et al., 2002), it may be related to affect’s role as a spotlight. The affect-as-spotlight hypothesis predicts that new benefit information will be more salient for decision makers who had previously developed positive feelings about a treatment or screening option. As a result, they will spend more time looking at and will remember more of the new benefit information, whereas they spend less time looking at any new risk information and will remember them less well. It predicts the reverse for treatment and screening options people evaluate negatively (e.g., less time spent considering and poorer memory for new benefit information). Thus, current affective experiences will guide the processing of cancer communications from physicians, decision aids, the media, and other sources. For example, in choices among screening options for colon cancer, negative feelings about contact with one’s own fecal matter may lead decision makers to spend more time examining its risks (lower accuracy rate) and less time examining its benefits. Knowing this might lead to a communication strategy of increasing positive mood in order to alter the negative feelings elicited during consideration of that option (or one could try to attenuate the negative affect elicited by making any fecal contact as abstract or normal as possible). No studies exist to the best of our knowledge testing these hypotheses. On a larger conceptual level, this means that we may need to tailor our communication approaches to a person’s emotions (see also Loewenstein, 2005). Affect as a motivator of information processing and behavior In a third role, affect appears to function as a motivator of information processing and behavior. Classical theories of emotion include, as the core of an emotion, a readiness to act and the prompting of plans (Frijda, 1986). Although affect is generally a much milder experience compared to a full-blown emotion state, recent research has demonstrated that we tend to automatically classify stimuli around us as good or bad and that this tendency is linked to behavioral tendencies. Stimuli classified as good elicit a tendency to approach, whereas those classified as bad elicit avoidance tendencies (Chen & Bargh, 1999). Incidental mood states also motivate behavior as people tend to act to maintain or attain positive mood states (Isen, 2000). In addition, there is abundant research showing that some emotions, such as worry, have been used to motivate specific behavioral change, such as mammography screening (McCaul, Branstetter, Schroeder, & Glasgow, 1996). Diefenbach, Miller, and Daly (1999), for example, demonstrated that worry about breast cancer predicted mammography screening independent of perceived breast cancer risks and trait anxiety.4 Several reviews explicate the role of fear appeals as a motivator of behavior change (e.g., Leventhal, 1970; Sutton, 1982; Witte, 1998). Affect also appears to be linked to the extent of deliberative effort decision makers are willing to put forth in order to make the best decision (i.e., the extent of systematic processing). Peters et al.’s (2003) studies of how people price lottery tickets were sometimes played for real money and other times were hypothetical. Although the same general effects emerged in both types of studies, real play elicited stronger feelings about the lottery tickets. At the same time, the real-play participants showed more evidence of taking the effort to calculate an expected value (40% of real-play buyers gave the expected value as their response compared to 10% of hypothetical-play buyers). In addition, half the hypothetical-play participants showed evidence of less effort in another way by giving the same buying price as their earlier selling price; only 3% of real-play participants did so. Thus, real play seemed to motivate buyers and sellers to work harder, and the motivating effect of real play was mediated by affect. These findings suggest that individuals (whether patients, family members, or friends) who have strong affect about cancer may work harder to find and process information about treatment or other options. Of course, Isen’s (2000) work on decision makers motivated to maintain or attain positive moods suggests that decision makers in negative moods from cancer diagnosis or other reasons may avoid making choices if they do not believe that the work to make the choice will lead to more positive or at least less negative moods. Thus, increased negative affect about cancer risk may lead to coping efforts and, for example, increased genetic testing when the potential for controlling risk is high (e.g., breast cancer) but decreased testing when risk reduction is not possible (e.g., Huntington’s disease; Croyle & Lerman, 1999). An additional source of motivation from affect may come from the negative feelings elicited by ambivalence. At times, our emotions and thoughts do not perfectly align themselves as positive or negative. Instead, objects or tasks are evaluated as having both positive and negative qualities. For example, smokers often have conflicting (ambivalent) emotions and thoughts about smoking (Lipkus, Green, Feaganes, & Sedikides, 2001; Lipkus et al., 2005). The decision to get a mammogram may involve weighing its benefits (e.g., gaining peace of mind, being responsible for one’s health) against its potential costs (e.g., pain, fear of finding cancer). In this instance, ambivalence could hinder the decision to get a mammogram (Rimer et al., 2002). Thus, in situations where the targeted behavior is related to positive health outcomes, ambivalence may be detrimental; conversely, ambivalence may serve a useful purpose by motivating change in unhealthy behaviors (e.g., smoking, drinking).5 We suspect that having ambivalent thoughts and feelings (Priester & Petty, 1996) may act as an important cue that triggers information search and influences decision making. Specifically, because the presence of opposing positive and negative beliefs and feelings is experienced as emotionally aversive (and a negative mood state results), the ambivalent individual is expected to engage in decisional conflict management strategies (Weber, Baron, & Loomes, 2001). Within the framework of conflict and decision-making models, ambivalence is assumed to trigger both problem-focused coping (e.g., what will help me decide?) and emotion-focused coping (e.g., how do I handle my ambivalent feelings? Haenze, 2001; Lazarus, 1991; Luce, 1998; Luce, Payne, & Bettman, 2001). Problem-focused coping should cause the person to devote more effort and thought to what action(s) seems appropriate (Haenze, 2001; Jonas, Diehl, & Bromer, 1997; Maio, Bell, & Esses, 1996). Emotion-focused coping should cause the person to engage in strategies to alleviate negative emotions (Haenze, 2001). In both cases, the negative mood state generated from facing difficult trade-offs motivates action to reduce uncertainty in decisions. Strategies to reduce uncertainty may take the form of turning to friends, experts, or other sources (e.g., written materials) to solve the dilemma. For example, a prostate cancer patient who is torn about which treatment, if any, to pursue may use these sources to gain insight, help clarify values, and inform the final decision. Alternatively, the other functions of affect may become more relevant. For example, if a prostate cancer patient feels ambivalent about a type of treatment (e.g., seed implant), the ensuing negative mood state may act as a spotlight and cause the patient to focus more on the risks than the benefits of this procedure. As a result, the patient may decide against having this treatment. Ambivalence may incorporate our notions of affect as a spotlight on certain features of a treatment with the notion of affect as a motivator of action. Future research could examine the effects of ambivalence on preferences, judgments, and decisions and on the communication efforts that might best facilitate good decisions. Affect as common currency Affect is simpler in some ways than thoughts. Affect often comes in two “flavors,” positive and negative; thoughts include more, and more complex, cost-benefit and other trade-offs (Trafimow & Sheeran, 2004). Several theorists have suggested that, as a result, integral affect plays a role as a common currency, allowing decision makers to compare apples to oranges (Cabanac, 1992). Montague and Berns (2002) link this notion to “neural responses in the orbitofrontal-striatal circuit which may support the conversion of disparate types of future rewards into a kind of internal currency, that is, a common scale used to compare the valuation of future behavioral acts or stimuli” (p. 265). By translating more complex thoughts into simpler affective evaluations, decision makers can compare and integrate good and bad feelings rather than attempt to make sense out of a multitude of conflicting logical reasons. This function thus is an extension of the affect-as-information function into more complex decisions that require integration of information. It predicts that affective information can be more easily and effectively integrated into judgments than less affective information. In the health plan choice studies of Peters, Slovic, et al. (2004), affective categories were hypothesized to act as overt markers of affective meaning in choices. If correct, then these overt markers should help participants consider relevant information (i.e., not considered as much when affective categories are not present) and apply that information to a complex judgment. Thus, affective categories should influence not just the choice of a health plan, as shown in previous studies, but should also help decision makers take into account more information and be more sensitive to variation in information. In an initial test of this hypothesis, participants were asked to judge a series of eight health plans one at a time on a 7-point attractiveness scale ranging from 1 =extremely unattractive to 7 =extremely attractive. For each health plan, they received information about cost and two quality attributes presented with numerical scores (e.g., Plan A scored 72 out of 100 points when members of the health plan rated the “ease of getting referrals to see a specialist”). The eight health plans represented a 2 × 2 × 2 design of low and high scores on each of the three attributes; eight versions were constructed using a Latin square design. Participants who received the quality information with affective categories (poor, fair, good, excellent) took into account more information and showed significantly greater sensitivity to differences in quality among plans. Thus, providing information in a more affective format appeared to help these judges better integrate important quality information into their judgments. Of course, it is also possible that these results reflect affect’s function as a motivator of behavior. Specifically, the affective categories manipulation may have produced a greater affective reaction that motivated the judges to work harder and integrate more information. These questions are empirical and deserve further research. Communicating the risks and benefits of cancer treatment and screening options can be quite difficult due to the complexity of information. It may be that providing this information in a more affective format might help patients integrate more information and thus make more informed choices. Relations among the four functions of affect The four functions of affect clearly are interrelated (see Figure 2). Affect, experienced due to an incidental source or upon consideration of some stimulus, can provide direction to cognition and be used as information and/or it can promote an automatic readiness to act and motivate information processing and behaviors (Oatley & Jenkins, 1996). Figure 2 Open in new tabDownload slide Hypothesized relationships among the functions of affect. Figure 2 Open in new tabDownload slide Hypothesized relationships among the functions of affect. If affect is used as information, it can be used as a substitute for other information in making a judgment (the direct line from affect as information to judgment) or it can lead to the use of affect as common currency when multiple pieces of information need to be combined. In this case, the affective information from multiple sources will be highlighted in the integration process; less affective information may receive less weight in the judgment process because it is not as easily integrated as the affective information. Finally, affect can be used as information to guide the selection of other information to process. This new information is highlighted and then used in the subsequent judgment. New affect could result from this process and take on a function of its own (e.g., negative affect from a cancer diagnosis could lead a patient to examine the risks more than the benefits of some treatment; if the risks are relatively small and elicit only slight negative affect, the function of affect as common currency might take over and the combination of the original affect from the diagnosis and the relatively neutral affect from the risks might lead to overall lower negative affect toward the cancer). If affect is used as a motivator of information processing and behaviors, it could lead simply to a health behavior of getting a mammogram, for example, or it might lead to more deep processing of information, coping efforts (including the avoidance of information that might lead to further distress), or communication to others. Each of these motivated processes and behaviors could lead to new affect that then takes on another function. Health communication efforts often have as their goal inducing long-term positive change in health-related behaviors. However, in research guided by the elaboration likelihood model, sources of affect incidental to a decision option (e.g., a mood state, the source of a message such as a famous liked person) have been found to evoke less systematic processing and less enduring attitude change than might be helpful in health communication (Petty, Barden, & Wheeler, 2002), suggesting that the use of affect is generally detrimental to this cause. But in most of these cases, the judgments were not necessarily very important or personally relevant. In cases where the opposite holds (e.g., high personal relevance), emotions can evoke the opposite response and may themselves receive greater scrutiny (Petty, DeSteno, & Rucker, 2001). We believe that integral affective sources (e.g., a cancer patient’s feelings elicited as he considers the pros and cons of a course of treatment) may increase the elaborations of health message recipients by directing attention, memory, and judgment to address the affect-eliciting event (affect’s function as a spotlight). The ability to process the message also may increase through the functions of affect as information and affect as common currency as data become meaningful information through affect (Hsee, Loewenstein, Blount, & Bazerman, 1999); the motivation to process and the extent of processing may increase through the function of affect as a motivator. It would be useful to include a manipulation of argument quality in further health communication tests of the functions of affect in order to examine more closely the processes underlying these effects. Functions of affect in health behavior theories Considering different functions of affect may help in further development of health behavior theories. Traditionally, health behavior theories have a strong cognitive orientation, focusing on health beliefs, expectations, social norms, and goals to predict health behaviors such as decisions to quit smoking, adhere to a weight loss program, or obtain a colonoscopy (Janz & Becker, 1984, Fishbein & Ajzen, 2005). The exception to this predominant cognitive orientation in health behavior theorizing is the self-regulation framework (Cameron & Leventhal, 2003; Petrie & Weinman, 1997) that explicitly takes into account affective reactions to a stimulus as determinants of health behavior. The self-regulation framework has its origin in the parallel processing model (Leventhal, 1970) that hypothesizes that individuals process external and internal stimuli (e.g., a message from their health plan to obtain a mammogram; detection of a lump) on both cognitive and emotional levels simultaneously. The self-regulation model regards the individual as a “problem solver” who actively attempts to make sense of environmental and somatic stimuli using both cognitive and affective processes (Leventhal et al., 1997). On the cognitive processing level, a stimulus is categorized according to five cognitive attributes that encompass the dimensions of identity (i.e., the attempt to name or label a stimulus), timeline (i.e., acute vs. chronic), cause (e.g., genetic, accident, exposure), consequence (i.e., minor vs. life threatening), and controllability (remedy vs. no remedy; Lau, Bernard, & Hartman, 1989). On the affective processing level, the stimulus evokes an affective response that interacts with the cognitive appraisal of the stimulus (Diefenbach et al., 1999). For example, imagine a person taking a shower and detecting a lump in the axilar region. The immediate response might be one of fear (i.e., affective response) accompanied by the simultaneous thought that it might be cancer (i.e., identity attribute). The cancer label might trigger related thoughts such as, “this is serious” (i.e., consequence), “I have to deal with this for a long time” (i.e., timeline), “I don’t know where this comes from” (i.e., cause), and “there is treatment available” (i.e., controllability). Each of these attributes can trigger specific emotions and not all attributes might be present at the same time. Of course, this interplay between cognition and emotion might take a different trajectory as self-regulation theory does not specify a fixed sequence among cognitive or emotional processing (Brissette, Leventhal, & Leventhal, 2003). What is important, however, is that this model provides a framework for affective functioning within a larger context of cognitive activity and recognizes that behavioral actions can be triggered as much by affective reactions as by cognitive processing. To continue with the example above, the individual might conclude to consult a physician about the lump because of her evaluation that it might be cancerous. However, she might also seek professional advice because her anxiety becomes unbearable and she needs reassurance (note that high levels of worry also might be associated with avoidant behavior). Although research has confirmed the existence of the cognitive attributes describing a stimulus, much less research has focused on affective processing mechanisms (Cameron & Leventhal, 2003). We argue that consideration of affect as having multiple functions provides a promising approach to enriching health behavior theories. The example above suggests that the traditional self-regulation framework recognizes two of the four affect functions: affect as information and affect as motivator. The individual instantaneously categorizes the lump as “something bad” and experiences elevated levels of anxiety; these feelings may be used as information to judge her cancer risk as high and may motivate her to seek professional advice quickly about diagnosis and treatment as well as about how to cope with the feelings themselves. During further cognitive and affective processing, the role of affect as a spotlight might come into play. Elevated levels of worry might lead her to focus more on potential negative interpretations and outcomes (affect-as-spotlight function), possibly ignoring benign interpretations of the lump (e.g., the lump is a bruise). Finally, across all the various internal and external information available to the individual, the affective information (e.g., worry about the lump, relief from a news story about a similar woman who found out her lump was not cancerous) may be integrated more and more easily than nonaffective information into ongoing evaluations of her risk, thus providing evidence for the affect-as-common currency function. This also underscores the need for a comprehensive theoretical framework that attempts to integrate different areas of research. Trade-offs between affect and deliberation Affect and cognitions both play a role in constructing preferences. The weight of one versus the other in judgments and choices, however, may be dictated in part by contextual factors. One such factor might be the importance or relevance of the decision. Clearly, choices of a life-saving treatment (e.g., what is the best cancer therapy) versus a weekend movie differ in importance. We suspect that when decisions are not very important, individuals will spend little thought determining a course of action, but rather rely on their affect as a substitute (e.g., how do I feel about it?). At the other extreme, when the decision is important and consequential, individuals may rely on both what they think and how they feel about the pros and cons of each option. Under these situations, individuals are often motivated to reach the most accurate decisions; hence, we assume that these decisions will often involve more central/systematic processing than peripheral/heuristic processing (Forgas, 2000; Payne, Bettman, & Johnson, 1993). The most accurate decisions may rely on interactions between thoughts and feelings. For example, different affective states (e.g., anxiety) can promote more in-depth systematic processing of health-relevant information as well as prime information consistent with those mood states (Cameron, 2003; Forgas, 2000). Affect associated with an option through prior experiences may guide decision makers to better options (e.g., somatic markers; Damasio, 1994). However, incidental feelings such as mood states also can color the interpretation of information and, thus, bias the direction of its evaluation in potentially inappropriate situations (e.g., when critical life-saving treatments need to be made). In some circumstances, individuals may become aware of the influence of their feelings on decisions, think the influence is inappropriate, and try to correct for it by relying on strategies based on their lay theories of how feelings may influence decisions (Petty & Wegener, 1993; Wegener & Petty, 1997). This combination of processes may lead to more accurate decisions. Other factors that may influence the weight accorded affective versus more cognitive information include decision complexity and familiarity. When a decision is complex because the person is faced with several nondominating options (i.e., no single option is superior to all others on the outcomes of interest, e.g., choices among prostate cancer treatment; Steginga & Occhipinti, 2004), too much information, or too technical information, the person may rely most on how they feel rather than on what they think—perhaps because they may not trust their evaluation of the facts or their ability to integrate multiple and competing reasons. Similarly, when the decision to be made can be equated with familiar past experiences, the individual may rely more on how they felt about the outcome(s) obtained rather than go through the effort of evaluating the pros and cons of each option (i.e., the affect-as-information function). For example, many individuals aged 50 and older are faced with the option of which method they should use to screen for colorectal cancer. Assuming some effort was put forth earlier thinking through available screening options (e.g., fecal occult blood test, sigmoidoscopy, colonoscopy) and/or the person was pleased with how the procedure went (e.g., not too embarrassing or time consuming, was painless), when faced with the same situation, he or she may simply use the affective cues attached to outcome(s) experienced to determine which, if any, screening technique to use. Factors that interfere with a person’s ability to cognitively evaluate information also can cause the individual to weigh affect more than cognition in choice. Shiv and Fedorikhin (1999), for example, found that decision makers under cognitive load were more likely to choose an affectively pleasing chocolate cake over a fruit salad compared to those in a no-load condition. For better or worse, cognitive load is ubiquitous in life. We are surrounded by environmental noises, interference, or distractions by others (e.g., phone calls, loud music, children, etc.), and we are plagued with busy schedules. These cognitive loads may be a particular burden to cancer patients faced, not only with their daily lives but also with the added stress and decisions of a potentially life-threatening disease. Another related influential factor that may tip the scale toward relying more on affect than on cognitions is time pressure (e.g., the need to quickly choose a treatment, being faced with a busy physician who has only a few minutes to provide information and answer questions). Finucane, Alhakami, Slovic, and Johnson (2000) found that the inverse correlation between risks and benefits grows larger under time pressure. Their findings suggest that the affective processes underlying perceptions of risks and benefits may weigh more when time for deliberation is low. In choices among cancer treatments, patients may be asked to make critical decisions immediately after diagnosis (e.g., treatment decisions about mastectomy vs. breast-conserving procedures and type of adjuvant treatment, if any, decisions about whether to enter a clinical trial, decisions about whom to tell and when; Revenson & Pranikoff, 2005). Many patients may feel delaying a decision is not an option. As a result, they may base their final decisions on how they feel because of the lack of perceived opportunity to reflect on the option(s). Finally, an important factor that may shape our emotional reactions, and hence preferences, is the information presentation format, such as in the health plan choice studies reviewed previously. A classic example is framing of information. Loss-frame messages emphasize the negative outcomes or benefits lost by taking or not taking some form of action; gain-frame messages emphasize positive outcomes or costs avoided by taking or not taking some form of action. The extant literature shows that losses loom larger than gains and hence are more influential in determining decisions and actions (Tversky & Kahneman, 1981). Thus, individuals are more likely to have adverse emotional reactions to information presented in terms of losses than gains (e.g., medical decisions in which treatment outcomes are framed in terms of lives lost vs. lives saved). In sum, differential weighting of cognitions and affect and affect itself are likely to be influenced by several contextual variables, some of which have been presented here. Admittedly, this is not an exhaustive listing. The main point here is that the influence of affect on the interpretation of health communications and the construction of health preferences is itself shaped by several situational factors that need to be considered. Is affect rational? Emotion’s influence on health decisions can be overwhelming. It can cause undue anxiety or fear, overestimation of risks (e.g., with breast cancer patients), and avoidant choices among treatment options (e.g., avoiding a treatment with better long-term survival because it requires a colostomy). It can also be a distraction when it provides information or motivation to attend to or act on emotional information at the expense of other more important message content (e.g., a positive mammogram result and the emotions that ensue may distract a woman from the information that only 20% of women with positive mammograms have breast cancer). Often times when people consider the impact of emotion on health decisions, they think about these kinds of negative impacts. Damasio (1994), on the other hand, argues that integral affect in the form of somatic markers increases the accuracy and efficiency of the decision process, and its absence (e.g., in the brain-damaged patients) degrades decision performance. This view of affect likely is also true in the domain of cancer. Affect is rational in the sense that some level of integral affect is necessary for information to have meaning so that decisions can be made. This “affective rationality” is key for health communications that have normally focused less on the role of affect and more on deliberative means (Hibbard & Peters, 2003). Warning labels on cigarette packages in the United States, for example, focus on four statements informing smokers about health hazards; evidence concerning these labels suggests that they have little influence on tobacco sales and have become invisible compared to the more salient and colorful cigarette packages and advertising (Fischer, Krugman, Fletcher, Fox, & Rojas, 1993; Fischer, Richards, Berman, & Krugman, 1989). Survey research in Canada, however, suggests that their larger warning labels with color pictures and 16 separate messages about specific risks of smoking create more negative emotional reactions toward cigarettes and increase attempts to quit among smokers (Hammond, Fong, McDonald, Cameron, & Brown, 2003; Hammond, McDonald, Fong, Brown, & Cameron, 2004). Affect’s role in health communication is likely to be nuanced; it therefore deserves careful empirical study of its effects on patients’ well-being. Affect sometimes may help and other times hurt decision processes. Which occurs will depend on the affect elicited by the communication attempt, how affect influences the information processing that takes place in the construction of preferences and how that particular influence matches whatever processing will produce the best decision for the cancer patient in that situation. In other words, the presence of affect does not guarantee good or bad decisions; it does guarantee that communicated information will be processed in ways that are different from when it is not present. Understanding the latter processes presents important communication challenges. Acknowledgments This paper is based in part on the chapter “The functions of affect in the construction of preferences” by E. Peters, which will appear in S. Lichtenstein & P. Slovic (Eds.), The construction of preference. New York: Cambridge University Press. Preparation of this paper was supported in part by the National Science Foundation under grant SES-0111941, SES-0339204, and SES-0241313. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. Notes 1 " Emotion theorists continue to debate the structure of emotion. Although some theorists argue for a different rotation of emotion space (in particular, a 45° rotation that yields the two factors of valence and arousal), we have chosen to focus on the rotation that yields factors of positive and negative affect. 2 " Forgas (1995) considered only the influence of incidental affect in judgments. The first three functions proposed here, however, map somewhat onto his judgmental strategies. First, the integral affect of this study may be present as a component of a pre-existing evaluation and thus may be experienced and used as the information in Forgas’ direct access strategy. In addition, the use of incidental affect as information overlaps completely with Forgas’ heuristic strategy. Second, similar to his substantive processing strategy, incidental or integral affect as a spotlight can influence attention, encoding, retrieval, and associative processes. Third, although incidental affect may not influence judgments in Forgas’ motivated processing strategy, integral affect, through its functions as a motivator, can drive the goals and behaviors of motivated processing. 3 " At extreme levels of negative affect, it may cease to provide information into a decision process and may instead prompt defensive processing of information. It is not clear whether there exists a particular level of affect that will prompt this change although Witte (1998) suggests that this “boomerang effect” will not occur so long as the decision maker perceives that he or she can exert some control over the situation. 4 " Some researchers have suggested that very high levels of cancer worry promote defensive avoidance of cancer screening behaviors (e.g., Lerman et al., 1993). If correct, then the specific behaviors motivated by affect may change depending upon the level of affect experienced. It may be that one level of affect promotes healthy screening behaviors, whereas an increase over some threshold produces avoidance, particularly in the absence of a concrete action plan (Leventhal, 1970). 5 " In the context where ambivalence hinders adaptive behaviors (e.g., getting a mammogram), communications may target reducing ambivalence. Such approaches may reduce the uncertainty and negative feelings associated with ambivalence, perhaps by helping people envision and make salient the positive aspects of behavior change. References Alhakami , A. S. , & Slovic , P. ( 1994 ). A psychological study of the inverse relationship between perceived risk and perceived benefit . Risk Analysis , 14 , 1085 – 1096 . Google Scholar Crossref Search ADS PubMed WorldCat Bower , G . ( 1981 ). Mood and memory . American Psychologist , 36 , 129 – 148 . Google Scholar Crossref Search ADS PubMed WorldCat Briss , P. , Rimer , B., Reilley , B., Coates , R. C., Lee , N. C., Mullen , P., et al. . ( 2004 ). Promoting informed decisions about cancer screening in communities and healthcare systems . American Journal of Preventive Medicine , 26 , 67 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Brissette , I. , Leventhal , H., & Leventhal , E. A. ( 2003 ). Observer ratings of health and sickness: Can other people tell us anything about our health that we don’t already know? Health Psychology , 22 , 471 – 478 . Google Scholar Crossref Search ADS PubMed WorldCat Cabanac , M . ( 1992 ). Pleasure: The common currency . Journal of Theoretical Biology , 155 , 173 – 200 . Google Scholar Crossref Search ADS PubMed WorldCat Cameron , L. D . ( 2003 ). Anxiety, cognition, and responses to health threats. In L. D. Cameron & H. Leventhal (Eds.), The self-regulation of health and illness behaviour (pp. 157 – 183 ). London : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cameron , L. D. , & Leventhal , H. (Eds.). ( 2003 ). The self-regulation of health and illness behavior . London : Routledge . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chen , M. , & Bargh , J. A. ( 1999 ). Consequences of automatic evaluation: Immediate behavioral predispositions to approach or avoid the stimulus . Personality & Social Psychology Bulletin , 25 , 215 – 224 . Google Scholar Crossref Search ADS WorldCat Croyle , R. T. , & Lerman , C. ( 1999 ). Risk communication in genetic testing for cancer susceptibility . Journal of the National Cancer Institute Monographs , 25 , 59 – 66 . Google Scholar Crossref Search ADS WorldCat Damasio , A. R . ( 1994 ). Descartes’ error: Emotion, reason, and the human brain . New York : Avon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Diefenbach , M. A. , & Hamrick , N. ( 2003 ). Self-regulation and genetic testing: Theory, practical considerations, and interventions. In L. Cameron & H. Leventhal (Eds.), The self-regulation of health and illness behaviour . Amsterdam : Harwood Academic . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Diefenbach , M. A. , Miller , S. M., & Daly , M. ( 1999 ). Specific worry about breast cancer predicts mammography use in women at risk for breast and ovarian cancer . Health Psychology , 18 , 532 – 536 . Google Scholar Crossref Search ADS PubMed WorldCat Edwards , A. , Unigwe , S., Elwyn , G., & Hood , K. ( 2003 ). Personalized risk communication for informed decision making about entering screening programs . The Cochrane Database of Systemic Reviews , Issue No. 1 , Art No. CD001865. OpenURL Placeholder Text WorldCat Epstein , S . ( 1994 ). Integration of the cognitive and the psychodynamic unconscious . American Psychologist , 49 , 709 – 724 . Google Scholar Crossref Search ADS PubMed WorldCat Erblich , J. , Bovbjerg , D. H., & Valdimarsdottir , H. B. ( 2000 ). Psychological distress, health beliefs, and frequency of breast self-examination . Journal of Behavioral Medicine , 23 , 277 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat Finucane , M. L. , Alhakami , A., Slovic , P., & Johnson , S. M. ( 2000 ). The affect heuristic in judgments of risks and benefits . Journal of Behavioral Decision Making , 13 , 1 – 17 . Google Scholar Crossref Search ADS WorldCat Fischer , P. M. , Krugman , D. M., Fletcher , J. E., Fox , R. J., & Rojas , T. H. ( 1993 ). An evaluation of health warnings in cigarette advertisements using standard market research methods: What does it mean to warn? Tobacco Control , 2 , 279 – 285 . Google Scholar Crossref Search ADS WorldCat Fischer , P. M. , Richards , J. W., Berman , E. F., & Krugman , D. M. ( 1989 ). Recall and eye tracking study of adolescents viewing tobacco advertisements . Journal of the American Medical Association , 261 , 84 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat Fishbein , M. , & Ajzen , I. ( 2005 ). Theory-based change interventions: Comments on Hobbis and Sutton . Journal of Health Psychology , 10 , 27 – 31 . Google Scholar Crossref Search ADS PubMed WorldCat Forgas , J. P . ( 1995 ). Mood and judgment: The affect infusion model (AIM) . Psychological Bulletin , 117 , 39 – 66 . Google Scholar Crossref Search ADS PubMed WorldCat Forgas , J. P . ( 2000 ). Feeling and thinking: The role of affect in social cognition . Cambridge, UK : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Frosch , D. L. , Kaplan , R. M., & Felitti , V. ( 2001 ). Evaluation of two methods to facilitate shared decision making for men considering the prostate-specific antigen test . Journal of General Internal Medicine , 16 , 391 – 398 . Google Scholar Crossref Search ADS PubMed WorldCat Frijda , N. H . ( 1986 ). The emotions: Studies in emotion and social interaction . New York : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Haenze , M . ( 2001 ). Ambivalence, conflict, and decision making: Attitudes and feelings in Germany towards NATO’s military intervention in the Kosovo war . European Journal of Social Psychology , 31 , 693 – 706 . Google Scholar Crossref Search ADS WorldCat Hammond , D. , Fong , G., McDonald , P., Cameron , R., & Brown , K. ( 2003 ). Impact of the graphic Canadian warning labels on adult smoking behavior . Tobacco Control , 12 , 391 – 395 . Google Scholar Crossref Search ADS PubMed WorldCat Hammond , D. , McDonald , P., Fong , G., Brown , K., & Cameron , R. ( 2004 ). The impact of cigarette warning labels and smoke-free bylaws on smoking cessation . Canadian Journal of Public Health , 95 , 201 – 204 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Hibbard , J. H. , & Peters , E. ( 2003 ). Supporting informed consumer health care decisions: Data presentation approaches that facilitate the use of information in choice . Annual Review of Public Health , 24 , 413 – 433 . Google Scholar Crossref Search ADS PubMed WorldCat Hsee , C. K. , Loewenstein , G., Blount , S., & Bazerman , M. H. ( 1999 ). Preference reversals between joint and separate evaluations of options: A review and theoretical analysis . Psychological Bulletin , 125 , 576 – 590 . Google Scholar Crossref Search ADS WorldCat Isen , A. M . ( 2000 ). Some perspectives on positive affect and self-regulation . Psychological Inquiry , 11 , 184 – 187 . OpenURL Placeholder Text WorldCat Janz , N. K. , & Becker , M. H. ( 1984 ). The health belief model: A decade later . Health Education Quarterly , 11 , 1 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat Johnson , E. J. , & Tversky , A. ( 1983 ). Affect, generalization, and the perception of risk . Journal of Personality and Social Psychology , 45 , 20 – 31 . Google Scholar Crossref Search ADS WorldCat Jonas , K. , Diehl , M., & Bromer , P. ( 1997 ). Effects of attitudinal ambivalence on information processing and attitude-intention consistency . Journal of Experimental Social Psychology , 33 , 190 – 210 . Google Scholar Crossref Search ADS WorldCat Kahneman , D . ( 2003 ). A perspective on judgment and choice: Mapping bounded rationality . American Psychologist , 58 , 697 – 720 . Google Scholar Crossref Search ADS PubMed WorldCat Kahneman , D. , Schkade , D., & Sunstein , C. R. ( 1998 ). Shared outrage and erratic awards: The psychology of punitive damages . Journal of Risk and Uncertainty , 16 , 49 – 86 . Google Scholar Crossref Search ADS WorldCat Kraus , N. , Malmfors , T., & Slovic , P. ( 1992 ). Intuitive toxicology: Expert and lay judgments of chemical risks . Risk Analysis , 12 , 215 – 232 . Google Scholar Crossref Search ADS WorldCat Lau , R. R. , Bernard , T. M., & Hartman , K. A. ( 1989 ). Further explorations of common-sense representations of common illnesses . Health Psychology , 8 , 195 – 219 . Google Scholar Crossref Search ADS PubMed WorldCat Lazarus , R. S . ( 1991 ). Progress on a cognitive-motivational-relational theory of emotion . American Psychologist , 46 , 819 – 834 . Google Scholar Crossref Search ADS PubMed WorldCat LeDoux , J. E . ( 1996 ). The emotional brain: The mysterious underpinnings of emotional life . New York : Simon & Schuster . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lerman , C. , Daly , M., Sands , C., Balshem , A., Lustbader , E., Heggan , T., et al. . ( 1993 ). Mammography adherence and psychological distress among women at risk for breast cancer . Journal of the National Cancer Institute , 85 , 1074 – 1080 . Google Scholar Crossref Search ADS PubMed WorldCat Lerman , C. , Lustbader , E., Rimer , B. K., Daly , M., Miller , S., Sands , C., et al. . ( 1995 ). Effects of individualized breast cancer risk counseling: A randomized trial . Journal of the National Cancer Institute , 87 , 286 – 292 . Google Scholar Crossref Search ADS PubMed WorldCat Lerner , J. S. , & Tiedens , L. Z. ( 2006 ). Toward a model of emotion-specific influences on judgment and decision making: Portrait of the angry decision maker . Journal of Behavioral Decision Making , 19 , 115 – 137 . Google Scholar Crossref Search ADS WorldCat Leventhal , H . ( 1970 ). Findings and theory in the study of fear communications. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 5 , pp. 119 – 186 ). New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Leventhal , H. , Benjamini , Y., Brownlee , S., Diefenbach , M. A., Leventhal , E. A., Patrick-Miller , L., et al. . ( 1997 ). Illness representations: Theoretical foundations. In K. J. Petrie & J. A. Weinman (Eds.), Perceptions of health and illness: Current research and applications (pp. 19 – 45 ). Amsterdam : Harwood Academic . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Leventhal , H. , Diefenbach , M. A., & Leventhal , E. A. ( 1992 ). Illness cognition: Using common sense to understand treatment adherence and affect cognition interactions . Cognitive Therapy and Research , 16 , 143 – 163 . Google Scholar Crossref Search ADS WorldCat Lichtenstein , S. , & Slovic , P. (Eds.). ( in press ). The construction of preference . New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Lipkus , I. M. , Green , J. D., Feaganes , J. R., & Sedikides , C. ( 2001 ). The relationship between attitudinal ambivalence and desire to quit among college smokers . Journal of Applied Social Psychology , 31 , 113 – 133 . Google Scholar Crossref Search ADS WorldCat Lipkus , I. M. , Kuchibhatla , M., McBride , C. M., Bosworth , H. B., Pollak , K. I., Siegler , I. C ., et al. . ( 2000 ). Relationships among breast cancer perceived absolute risk, comparative risk, and worries . Cancer Epidemiology, Biomarkers & Prevention , 9 , 973 – 975 . OpenURL Placeholder Text WorldCat Lipkus , I. M. , Pollak , K. I., McBride , C. M., Schwartz-Bloom , R. D., Lyna , P., & Bloom , P. S. ( 2005 ). Assessing attitudinal ambivalence towards smoking and its association with desire to quit among teen smokers . Psychology and Health , 20 , 373 – 388 . Google Scholar Crossref Search ADS WorldCat Loewenstein , G . ( 2005 ). Hot-cold empathy gaps and medical decision making . Health Psychology , 24 , S49 – S56 . Google Scholar Crossref Search ADS PubMed WorldCat Loewenstein , G. F. , Weber , E. U., Hsee , C. K., & Welch , E. S. ( 2001 ). Risk as feelings . Psychological Bulletin , 127 , 267 – 286 . Google Scholar Crossref Search ADS PubMed WorldCat Luce , M. F . ( 1998 ). Choosing to avoid: Coping with negatively emotion-laden consumer decisions . Journal of Consumer Research , 24 , 409 – 433 . Google Scholar Crossref Search ADS WorldCat Luce , M. F. , Payne , J. W., & Bettman , J. R. ( 2001 ). The impact of emotional tradeoff difficulty in decision behavior. In E. U. Weber, J. Baron, & G. Loomes (Eds.), Conflict and tradeoffs in decision-making (pp. 86 – 109 ). Cambridge, UK : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Maio , G. R. , Bell , D. W., & Esses , V. M. ( 1996 ). Ambivalence and persuasion: The processing of messages of immigrant groups . Journal of Experimental Social Psychology , 32 , 513 – 536 . Google Scholar Crossref Search ADS PubMed WorldCat McCaul , K. D. , Branstetter , A. D., Schroeder , D. M., & Glasgow , R. E. ( 1996 ). What is the relationship between breast cancer risk and mammography screening? A meta-analytic review . Health Psychology , 15 , 423 – 429 . Google Scholar Crossref Search ADS PubMed WorldCat Millar , M. G. , & Tesser , A. ( 1989 ). The effects of affective-cognitive consistency and thought on the attitude-behavior relation . Journal of Experimental Social Psychology , 25 , 189 – 202 . Google Scholar Crossref Search ADS WorldCat Montague , R. P. , & Berns , G. S. ( 2002 ). Neural economics and the biological substrates of valuation . Neuron , 36 , 265 – 284 . Google Scholar Crossref Search ADS PubMed WorldCat Nabi , R. L . ( 2003 ). Exploring the framing effects of emotion: Do discrete emotions differentially influence information accessibility, information seeking, and policy preference? Communication Research , 30 , 224 – 247 . Google Scholar Crossref Search ADS WorldCat Oatley , K. , & Jenkins , J. M. ( 1996 ). Understanding emotions . Oxford, UK : Blackwell . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Osgood , C. E. , Suci , G. J., & Tannenbaum , P. H. ( 1957 ). The measurement of meaning . Urbana, IL : University of Illinois . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Parascandola , M. , Hawkins , J., & Danis , M. ( 2002 ). Patient autonomy and the challenge of clinical uncertainty . Kennedy Institute of Ethics Journal , 12 , 245 – 264 . Google Scholar Crossref Search ADS PubMed WorldCat Payne , J. W. , Bettman , J. R., & Johnson , E. J. ( 1993 ). The adaptive decision maker . New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Payne , J. W. , Bettman , J. R., & Schkade , D. A. ( 1999 ). Measuring constructed preferences: Towards a building code . Journal of Risk and Uncertainty , 19 , 243 – 270 . Google Scholar Crossref Search ADS WorldCat Peters , E . (in press). The functions of affect in the construction of preferences. In S. Lichtenstein & P. Slovic (Eds.), The construction of preference . New York : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Peters , E. , Burraston , B., & Mertz , C. K. ( 2004 ). An emotion-based model of stigma susceptibility: Appraisals, affective reactivity, and worldviews in the generation of a stigma response . Risk Analysis , 24 , 1349 – 1367 . Google Scholar Crossref Search ADS PubMed WorldCat Peters , E. , & Slovic , P. ( 1996 ). The role of affect and worldviews as orienting dispositions in the perception and acceptance of nuclear power . Journal of Applied Social Psychology , 26 , 1427 – 1453 . Google Scholar Crossref Search ADS WorldCat Peters , E. , Slovic , P., & Gregory , R. ( 2003 ). The role of affect in the WTA/WTP disparity . Journal of Behavioral Decision Making , 16 , 309 – 330 . Google Scholar Crossref Search ADS WorldCat Peters , E. , Slovic , P., & Hibbard , J. ( 2004 ). Evaluability manipulations influence the construction of choices among health plans (Report No. 04-02) . Eugene, OR : Decision Research . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Peters , E. , Slovic , P., Hibbard , J., & Tusler , M. ( 2006 ). Why worry? Perceived risk of medical errors and willingness to act to reduce errors . Health Psychology , 25 , 144 – 152 . Google Scholar Crossref Search ADS PubMed WorldCat Peters , E. , Västfjäll , D., Slovic , P., Mertz , C. K., Mazzocco , K., & Dickert , S. ( 2006 ). Numeracy and decision making . Psychological Science , 17 , 407 – 413 . Google Scholar Crossref Search ADS PubMed WorldCat Petrie , K. J. , & Weinman , J. A. ( 1997 ). Perceptions of health and illness: Current research and applications . Amsterdam : Harwood Academic . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Barden , J., & Wheeler , S. C. ( 2002 ). The elaboration likelihood model of persuasion: Health promotions that yield sustained behavioral change. In R. J. DiClemente, R. A. Crosby, & M. Kegler (Eds.), Emerging theories in health promotion practice and research (pp. 71 – 99 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , DeSteno , D., & Rucker , D. D. ( 2001 ). The role of affect in attitude change. In J. P. Forgas (Ed.), Handbook of affect and social cognition (pp. 212 – 236 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Wegener , D. T. ( 1993 ). Flexible correction processes in social judgment: Correcting for context-induced contrast . Journal of Experimental Social Psychology , 29 , 137 – 165 . Google Scholar Crossref Search ADS WorldCat Priester , J. R. , & Petty , R. E. ( 1996 ). The gradual threshold model of ambivalence: Relating the positive and negative bases of attitudes to subjective ambivalence . Journal of Personality and Social Psychology , 71 , 431 – 449 . Google Scholar Crossref Search ADS PubMed WorldCat Revenson , T. A. , & Pranikoff , J. R. ( 2005 ). A contextual approach to treatment decision making among breast cancer survivors . Health Psychology , 24 , S93 – S98 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. , Halabi , S., Skinner , C. S., Lipkus , I. M., Strigo , T. S., Kaplan , E. B., et al. . ( 2002 ). Effects of a mammography decision-making intervention at 12 and 24 months . Preventive Medicine , 22 , 247 – 257 . Google Scholar Crossref Search ADS WorldCat Rottenstreich , Y. , & Hsee , C. K. ( 2001 ). Money, kisses, and electric shocks: On the affective psychology of risk . Psychological Science , 12 , 185 – 190 . Google Scholar Crossref Search ADS PubMed WorldCat Schwarz , N. , & Clore , G. L. ( 1983 ). Mood, misattribution, and judgments of well-being: Information and directive functions of affective states . Journal of Personality and Social Psychology , 45 , 513 – 523 . Google Scholar Crossref Search ADS WorldCat Schwarz , N. , & Clore , G. L. ( 2003 ). Mood as information: 20 years later . Psychological Inquiry , 14 , 294 – 301 . Google Scholar Crossref Search ADS WorldCat Shiv , B. , & Fedorikhin , A. ( 1999 ). Heart and mind in conflict: Interplay of affect and cognition in consumer decision making . Journal of Consumer Research , 26 , 278 – 282 . Google Scholar Crossref Search ADS WorldCat Slovic , P . ( 1995 ). The construction of preference . American Psychologist , 50 , 364 – 371 . Google Scholar Crossref Search ADS WorldCat Slovic , P. , Finucane , M. L., Peters , E., & MacGregor , D. G. ( 2002 ). The affect heuristic. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment (pp. 397 – 420 ). New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Steginga , S. K. , & Occhipinti , S. ( 2004 ). The application of the heuristic-systematic processing model to treatment decision making about prostate cancer . Medical Decision Making , 24 , 573 – 583 . Google Scholar Crossref Search ADS PubMed WorldCat Sutton , S. R . ( 1982 ). Fear-arousing communications: A critical examination of theory and research. In J. R. Eiser (Ed.), Social psychology and behavioral medicine (pp. 303 – 337 ). London : Wiley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Trafimow , D. , & Sheeran , P. ( 2004 ). A theory about the translation of cognition into affect and behavior. In G. Maio & G. Haddock (Eds.), Contemporary perspectives in the psychology of attitudes: The Cardiff symposium (pp. 57 – 75 ). London : Psychology Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tversky , A. , & Kahneman , D. ( 1981 ). The framing of decisions and the psychology of choice . Science , 211 , 453 – 458 . Google Scholar Crossref Search ADS PubMed WorldCat Watson , M. , Lloyd , S., Davidson , J., Meyer , L., Eeles , R., Ebbs , S., et al. . ( 1999 ). The impact of genetic counseling on risk perception and mental health in women with a family history of breast cancer . British Journal of Cancer , 79 , 868 – 874 . Google Scholar Crossref Search ADS PubMed WorldCat Weber , E. U. , Baron , J., & Loomes , G. ( 2001 ). Conflict and tradeoffs in decision-making . Cambridge, UK : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wegener , D. T. , & Petty , R. E. ( 1997 ). The flexible correction model: The role of naive theories of bias correction. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 29 , pp. 141 – 208 ). New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Witte , K . ( 1998 ). Fear as motivator, fear as inhibitor: Using the extended parallel response model to explain fear appeal success and failures. In P. A. Anderson & L. K. Guerrero (Eds.), Handbook of communication and emotion: Research, theory, and contexts (pp. 424 – 451 ). San Diego, CA : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zajonc , R. B . ( 1980 ). Feeling and thinking: Preferences need no inferences . American Psychologist , 35 , 151 – 175 . Google Scholar Crossref Search ADS WorldCat © 2006 International Communication Association

Viswanath,, K.;Emmons, Karen, M.

doi: 10.1111/j.1460-2466.2006.00292.xpmid: N/A

Abstract Recent work on message effects theories offers a fruitful way to systematically explore how features, formats, structures of messages may attract audience attention and influence the audience and is of great relevance to public health communications. Much of this work, however, has been pursued primarily at the individual level of analysis. It is our contention that message effects on health outcomes could potentially be moderated and mediated by social contextual factors in public health such as social class, social organizations and neighborhoods among others, leading to differential effects among different audience sub-groups. This essay, through a selective review of literatures in communication and social epidemiology, will explore how major message effects may moderate and mediate the role of social determinants of health on cancer control, specifically cancer-related health disparities. A significant challenge in public health communications is how to transcend the clutter in the information environment to reach one’s intended target audience with effective messages. Practitioners and campaign planners have long focused on selecting appropriate channels and developing the “right messages” to cut through the noise to get noticed and influence the audience (Hornik, 2002). Although much of the work on “right messages” has been based on intuitive assumptions about the intended audience and the media to which they attend (e.g., television, radio, newspapers), recent work on message effects theories offers a fruitful way to systematically explore how features, formats, and structure of messages may attract audience attention and influence the audience. Even as the work in message effects has continued to contribute to our understanding of how the media influence cognitions, attitudes, and behavior, this work has been pursued primarily at the individual level of analysis. That is, the focus is usually on how individual characteristics such as personality or beliefs influence the reception and processing of messages that lead to certain outcomes. At the same time, a large body of work in public health—social epidemiology—has identified how “social determinants” such as social class, social networks, neighborhood conditions, and social cohesion among others could elucidate the different etiological pathways that lead to differential disease outcomes among population subgroups (Berkman & Kawachi, 2000). It is our contention that message effects on health outcomes could potentially be moderated and mediated by such social contextual factors in public health leading to differential effects among different audience subgroups. Hence, it is critical to understand how research in message effects in health may inform and be informed by work in social epidemiology. This article, through a selective review of literatures in communication and social epidemiology, will explore the relationship between major message effects and social determinants of health and how that may influence cancer control, specifically cancer-related health disparities. We will (a) start with a brief overview of some message effects, (b) review some of the major social determinants of health, (c) draw connections between message effects and social determinants through a selective use of examples, and (d) end with a speculation on how this research can inform cancer control, particularly cancer-related health disparities. Message effects The study of message effects is distinct from the study of media effects, a broader treatment of effects of media that may include message effects, but may also focus on influence of other elements, such as exposure to media content, channel use, and information flows. Message effects are a class of theories that examine how features, formats, and content of mediated messages produce effects on cognitive, attitudinal, and behavioral outcomes. Some major message formats include framing, sensation seeking, narratives, exemplars, and fear appeals. Discussion of all these message formats in the context of this essay is not feasible; thus, we have selected framing and narratives as two examples of message effects that we will explore further in the context of disparities and social determinants of health. Framing The term framing has been used in at least two different ways in the communication literature and describes the efforts of defining a health issue or problem both in strategic communication and as well as news media coverage of health. In one variant, a message is framed to emphasize the consequences of either adopting or failing to adopt health behaviors (Salovey, Schneider, & Apanovitch, 2002). Gain-framed messages emphasize the benefits from adopting a health behavior, although loss-framed messages focus on costs of not adopting a particular behavior (Salovey et al., 2002) (See Rothman, Bartels, Wlaschin, & Salovey, 2006). Messages are framed with the intention of manipulating cognitions and affect, and its use in promoting cancer screening and in treatment decisions has been extensively studied (Banks et al., 1995; Davey, Butow, & Armstrong, 2003; Finney & Iannotti, 2002; Salovey & Williams-Piehota, 2004; Schneider et al., 2001). A review of literature on effects of framing suggest that loss framing is more effective than gain framing on screening behaviors, though that was not the result in case of skin cancer screening (Detweiler, Bedell, Salovey, Pronin, & Rothman, 1999). Much of the discussion on gain- versus loss-framed messages has been pursued at the individual level, how message frames influence individual health cognitions, information processing, and behaviors though there are some exceptions (e.g., see Salovey et al., 2002). A second variant of framing is more characteristic of literature in media studies, where framing has been conceptualized as a way to “construct” and “define” a topic, a “central organizing principle that holds together a diverse array of symbols and idea elements,” giving a coherent structure to one’s world views (Gamson, 2001). Frames typically are a product of interaction between frame “sponsors” (e.g., social movement organizations, organized social groups including businesses, hospitals, government agencies, and civic groups) and the news media that cover these groups and agencies (Cappella & Jamieson, 1997). Media frames suggest how to “look at a problem” and the “solutions” to resolve the problem, public opinion about social problems, and knowledge and learning about those topics (Entman, 1993; Gamson & Modigliani, 1989; Scheufele, 1999). In that vein, the media could be helpful in (a) creating awareness about cancer and cancer-related risks and behaviors and (b) generating public support for actions to reduce or exacerbate those risks depending on framing. For example, within the debate on mammography as a screening tool, mammography has been framed as either an effective way to save lives through early detection or one that leads to “false positives,” resulting in distress and/or unnecessary treatment. This approach to framing studies has been more sensitive to the structural factors such as the relationship between institutions of media (journalists) and news sources, for example (Cappella & Jamieson, 1997). Narratives Another message format is narratives or stories (Green, 2006). Story telling, an age-old cultural phenomenon, is a major staple of mass media genres of news, advertising and, entertainment. In contrast to explicit attempts at persuasion, narratives or stories carry implicit meanings with a persuasive intent and could be of any length though longer stories have a better chance of addressing barriers that might lead to disparities. They may range form 30-second advertising spots or full-page ads (e.g., about how mammography saved the life of a woman through early detection) to full-length televised soap operas that may follow a character dealing with a diagnosis of breast cancer. Incidental exposure to fictional narratives may influence viewers’ cognitions such as their perceptions of social reality (Gerbner, Gross, Morgan, Signorielli, & Shanahan, 2002) or behaviors (Anderson & Bushman, 2002) or knowledge about medicine and health (Davin, 2003). Narratives, particularly personal stories, are routinely used by journalists to make their points or to embellish news reports (Hoffman-Goetz, 1999). The power of the narrative, some argue, stems from the possible reason that the audience member is absorbed by the features of the story and is “transported” to a different world and comes back changed by the experience, accepting beliefs congruent with the story (Green & Brock, 2000). Narrative techniques have been widely adopted by communication scholars and practitioners to promote “prosocial” behaviors whether through formats such as soap operas, telenovelas, comic books, or even text-based narratives (Slater, 2002). Within public health, narratives have been used to promote family planning, AIDS education, and cancer screening among others (Kreuter & McClure, 2004; Slater, 2002), with the belief that public health approaches that are culturally compatible with the intended audience are more effective though evidence of a need to construct messages separately for each individual is mixed (Institute of Medicine & Committee on Communication for Behavior Change in the 21st Century: Improving the Health of Diverse Populations, 2002). The body of work on message effects, in general, has drawn from psychological theories and examined how individual characteristics interact with message structure to lead to outcomes in knowledge, attitudes, and behavior. Yet, this body of work does not address whether message effects are uniform across all audience subgroups and if message characteristics interact with social contextual factors such as social class, social networks, and social organizations. We will next review some major determinants of health that illustrate why a more socially focused, macrolevel approach is needed, before discussing how literatures in message effects and social determinants may inform each other for a more refined understanding of media effects in cancer control. Health disparities and social determinants of health There are profound disparities in health among different racial/ethnic and social groups. These disparities also extend to behaviors that are associated with increased cancer risk, such as diet, smoking (Barbeau, Krieger, & Soobader, 2004), physical inactivity, obesity (National Center for Health Statistics, 2004), and cancer screening (Institute of Medicine, 1999). Recent studies have also documented that there are disparities in what treatments are offered to cancer patients and differences across groups in treatments selected (Institute of Medicine, 2003; Schneider, Zaslavsky, & Epstein, 2002). In general, minorities and low-socioeconomic status (SES) groups are less likely to have access to preventive services, less likely to receive cancer screening services, and more likely to be in later stages when cancer is detected than other groups (Agency for Healthcare Research and Quality, 2003; National Center for Health Statistics). Such differences in disease burden among population subgroups and the possible reasons for them led epidemiologists to focus on “social determinants” that could elucidate the different etiological pathways that lead to differential health outcomes (Berkman & Kawachi, 2000). The social epidemiological approach posits that health behaviors are a product of a complex interaction of factors across intrapersonal, interpersonal, organizational, environmental, and policy levels (Kawachi & Berkman, 2000), and that planned social change in public health should focus on interventions that cut across these multiple levels (Smedley & Syme, 2000). Major social determinants (reviewed in Table 1) that are thought to moderate and influence health behaviors include social class or socioeconomic status (Krieger, 2001; Lynch & Kaplan, 2000; Lynch, Smith, Kaplan, & House, 2000), social networks through social capital and social cohesion (Berkman & Kawachi, 2000), social organizations such as workplace (Linnan, Emmons, & Abrams, 2002; Sorensen et al., 2003), communities and neighborhoods, and social policies (Sorensen et al.). The causal pathways that connect social determinants with health outcomes have been a subject of some controversy (Kaplan, 2004), though it is generally agreed that the influence of social determinants is through both “proximal” and “distal” factors such as access to material and intellectual resources, social support and living conditions, unequal distribution of knowledge, and exposures to environmental stressors, among others (Institute of Medicine, 1999). Differential access to knowledge between high- and low-SES groups is also offered as one mechanism linking SES with disparities. In short, disparities in health may occur concomitantly with disparities in access to information and knowledge, that is, communication inequality. Table 1 Exemplars of Social Determinants and Health Outcomes Social Determinant . Health Outcomes and Possible Reasons . Social class/socioeconimic status Effects on life course—access to material and intellectual resources, social support, and living conditions (Krieger, 2001; Lynch & Kaplan, 2000; Lynch et al., 2000) Unequal distribution of prestige, power, resources (Phelan et al., 2004) Greater probability of living in poor neighborhoods and risk of environmental exposures (Phelan et al., 2004) Unequal burden in morbidity and mortality (Institute of Medicine, 1999) Social organizations Workplace Formal and Informal policies that regulate health behaviors such as smoking  Volunteer and civic groups Job strain, occupational cultures, and social norms influence access to health services and behaviors (Linnan et al., 2002; Sorensen et al., 2003) Social networks and social ties Social support and social cohesion (Subramanian et al., 2002) Social cohesion (Berkman, 2000) Health media use and health information (Viswanath et al., in press) Community Environment Influence of community structure and complexity (Berkman & Kawachi, 2000; Viswanath et al., 1991) Neighborhood conditions (King et al., 2002, Subramanian et al., 2002) Public policies Moderation of risk behaviors (Emmons, 2000; Emmons et al., 2001; Sorensen et al., 2003) Indoor air policies influence smoking behavior (Gilpin et al., 2004) Built environment may influence physical activity (King et al., 2002) Social Determinant . Health Outcomes and Possible Reasons . Social class/socioeconimic status Effects on life course—access to material and intellectual resources, social support, and living conditions (Krieger, 2001; Lynch & Kaplan, 2000; Lynch et al., 2000) Unequal distribution of prestige, power, resources (Phelan et al., 2004) Greater probability of living in poor neighborhoods and risk of environmental exposures (Phelan et al., 2004) Unequal burden in morbidity and mortality (Institute of Medicine, 1999) Social organizations Workplace Formal and Informal policies that regulate health behaviors such as smoking  Volunteer and civic groups Job strain, occupational cultures, and social norms influence access to health services and behaviors (Linnan et al., 2002; Sorensen et al., 2003) Social networks and social ties Social support and social cohesion (Subramanian et al., 2002) Social cohesion (Berkman, 2000) Health media use and health information (Viswanath et al., in press) Community Environment Influence of community structure and complexity (Berkman & Kawachi, 2000; Viswanath et al., 1991) Neighborhood conditions (King et al., 2002, Subramanian et al., 2002) Public policies Moderation of risk behaviors (Emmons, 2000; Emmons et al., 2001; Sorensen et al., 2003) Indoor air policies influence smoking behavior (Gilpin et al., 2004) Built environment may influence physical activity (King et al., 2002) Open in new tab Table 1 Exemplars of Social Determinants and Health Outcomes Social Determinant . Health Outcomes and Possible Reasons . Social class/socioeconimic status Effects on life course—access to material and intellectual resources, social support, and living conditions (Krieger, 2001; Lynch & Kaplan, 2000; Lynch et al., 2000) Unequal distribution of prestige, power, resources (Phelan et al., 2004) Greater probability of living in poor neighborhoods and risk of environmental exposures (Phelan et al., 2004) Unequal burden in morbidity and mortality (Institute of Medicine, 1999) Social organizations Workplace Formal and Informal policies that regulate health behaviors such as smoking  Volunteer and civic groups Job strain, occupational cultures, and social norms influence access to health services and behaviors (Linnan et al., 2002; Sorensen et al., 2003) Social networks and social ties Social support and social cohesion (Subramanian et al., 2002) Social cohesion (Berkman, 2000) Health media use and health information (Viswanath et al., in press) Community Environment Influence of community structure and complexity (Berkman & Kawachi, 2000; Viswanath et al., 1991) Neighborhood conditions (King et al., 2002, Subramanian et al., 2002) Public policies Moderation of risk behaviors (Emmons, 2000; Emmons et al., 2001; Sorensen et al., 2003) Indoor air policies influence smoking behavior (Gilpin et al., 2004) Built environment may influence physical activity (King et al., 2002) Social Determinant . Health Outcomes and Possible Reasons . Social class/socioeconimic status Effects on life course—access to material and intellectual resources, social support, and living conditions (Krieger, 2001; Lynch & Kaplan, 2000; Lynch et al., 2000) Unequal distribution of prestige, power, resources (Phelan et al., 2004) Greater probability of living in poor neighborhoods and risk of environmental exposures (Phelan et al., 2004) Unequal burden in morbidity and mortality (Institute of Medicine, 1999) Social organizations Workplace Formal and Informal policies that regulate health behaviors such as smoking  Volunteer and civic groups Job strain, occupational cultures, and social norms influence access to health services and behaviors (Linnan et al., 2002; Sorensen et al., 2003) Social networks and social ties Social support and social cohesion (Subramanian et al., 2002) Social cohesion (Berkman, 2000) Health media use and health information (Viswanath et al., in press) Community Environment Influence of community structure and complexity (Berkman & Kawachi, 2000; Viswanath et al., 1991) Neighborhood conditions (King et al., 2002, Subramanian et al., 2002) Public policies Moderation of risk behaviors (Emmons, 2000; Emmons et al., 2001; Sorensen et al., 2003) Indoor air policies influence smoking behavior (Gilpin et al., 2004) Built environment may influence physical activity (King et al., 2002) Open in new tab Communication inequality Communication inequality may be defined as differences in the generation, manipulation, and distribution of information among social groups; and differences in (a) access and use, (b) attention, (c) retention, and (d) capacity to act on relevant information among individuals (Viswanath, in press). For example, attention paid to health information in different sources may vary by SES groups (Figure 1). Attention to media and messages is a robust predictor of knowledge (Romantan et al., 2005; Viswanath, Randolph et al., in press). Figure 1 Open in new tabDownload slide Attention to health in different media among different educational groups. Figure 1 Open in new tabDownload slide Attention to health in different media among different educational groups. Recall and retention of information from diffusion of information is more likely by higher SES groups compared to lower SES groups, a phenomenon characterized as the “knowledge gap,” though there are exceptions to this trend (Kwak, 1999; Viswanath & Finnegan, 1996). Audiences may vary in their capacity to act on information provided either because of lack of understanding the message or because of a lack of opportunity to act on the message. For example, limited health literacy is reported to be related to less knowledge of illness management, lower participation in shared decision making in treatment, adherence and compliance and lower self-reported health status (Nielsen-Bohlman, Panzer, & Kindig, 2004). These issues of inequality are particularly pertinent to a discussion of message effects, and may suggest a potential causal pathway between health disparities and message effects, which will be discussed in more detail in sections that follow. Media and message effects: A structural view Given the influential role of the media in our society today, it is somewhat surprising that there has been very little emphasis on the role that media play in promulgating or redressing health disparities. In media studies, with few exceptions, research on effects of media and messages, in both campaign and general media effects areas, has long focused on the individual as a unit of analysis, as mentioned earlier. The focus has been primarily on the impact of media on individual cognition, affect and behavior. This is a reflection of the field with its roots in the social psychological experimental work that developed post-World War II. This literature has made rich contributions to our understanding of the role of media in attitude reinforcement and attitude change and the conditions under which such effects may or may not occur (Bryant & Zillman, 2002). Yet, others have argued for a more structural approach, with the assumption that media messages are a product of interaction among agencies, groups, and individual actors within those agencies (Olien, Donohue, & Tichenor, 1983; Viswanath & Demers, 1999). This structural view contends that audiences attend and react to mediated content based on their structural location in the environment (Viswanath & Demers, 1999) and the social roles they play at any given time (McLeod, Kosicki, & Pan, 1991). For example, preferences for media channels, attribution of credibility and trust of information sources and reactions to media coverage depends on the social class and perceived power status of the audience groups (Finnegan, Viswanath, Kahn, & Hannan, 1993; Gaziano & McGrath, 1986; Olien et al., 1983). Audiences bring their collective experiences as members of social groups with specific characteristics (e.g., social class, occupational structure, race and ethnicity, and gender) in reacting to and processing of media messages. These structural determinants may mediate the impact of messages through such factors as access, preferences, appeals, attention and processing and capacity to act on the information. Drawing from both the literatures of media studies and social epidemiology, there are three broad structural determinants of health that may mediate message effects: (a) socioeconomic position, (b) social networks and social capital, and (c) social organizations including media institutions and practices (Table 2).1 Table 2 Social Determinants, Message Effects, and Potential Pathways Leading to Health Consequences Structural Factors . Media Effect . Potential Pathways . Examples of Health Consequences . Socioeconomic status Framing Increase attention, salience Low attention to health-related topics and low salience may result in greater “knowledge gaps” on cancer prevention, poor health behaviors, less participation in screening, lower participation in informed decision making, increase in disparities between low and high SES Activate existing schema or develop new schema (new ways looking at a problem and a solution) High attention and salience may engage the audience, greater learning on the part of low SES, potential for behavior change, more informed decision making, reduction or elimination of disparities Lower SES groups may have more difficulties sorting through complex health messages and thus may reject messages or have more difficulty reconciling conflicting messages Narratives/exemplars Liking Greater engagement may lead to greater knowledge, positive role models, greater identification with the messages, more positive health behaviors Attention Retention Absorption/transportation Social networks and social capital Framing Collective identity with the issue or topic Provides an interpretation of health messages within the context of a social relationship; individuals from a lower SEP background may have more exposure to individuals with less health care access and less trust of the health care system, which influences social norms regarding health Additional source of information Social networks provide opportunities to see similar others address health concerns; modeling of how similar others coped with disease burden interpersonal discussions may increase salience of health topics, and a consequence may increase depth of information processing about it Development and reinforcement of social norms Narratives/exemplars Modeling Reinforcement of norms Social organizations (churches, worksites, media institutions, and practices) Framing Worry or fear Cultivates either excessive worry or disregard for cancer risk behaviors Appeal to cultural values Media response to advocacy efforts may disproportionately increase attention to specific issues; advocacy related to disparities has been limited Reliance on powerful sources for news and interpretation Consequences on how the stories on health may or may not appeal to low- versus high-SES groups Health concerns of lower SES get limited attention in the public arena Potential for limited action to ameliorate disease burden among low SES Narratives in TV Repeated exposure Higher than actual portrayal of violence on television engenders belief in a violent world; repeated exposure may serve as a cue and behavioral model Cultivation of TV worldviews Focus on TV violence may increase salience of neighborhood cues that support belief about a violent world, reduced community engagement, reduced opportunities for social exchange, knowledge, reduced physical activity, and other behavior changes Structural Factors . Media Effect . Potential Pathways . Examples of Health Consequences . Socioeconomic status Framing Increase attention, salience Low attention to health-related topics and low salience may result in greater “knowledge gaps” on cancer prevention, poor health behaviors, less participation in screening, lower participation in informed decision making, increase in disparities between low and high SES Activate existing schema or develop new schema (new ways looking at a problem and a solution) High attention and salience may engage the audience, greater learning on the part of low SES, potential for behavior change, more informed decision making, reduction or elimination of disparities Lower SES groups may have more difficulties sorting through complex health messages and thus may reject messages or have more difficulty reconciling conflicting messages Narratives/exemplars Liking Greater engagement may lead to greater knowledge, positive role models, greater identification with the messages, more positive health behaviors Attention Retention Absorption/transportation Social networks and social capital Framing Collective identity with the issue or topic Provides an interpretation of health messages within the context of a social relationship; individuals from a lower SEP background may have more exposure to individuals with less health care access and less trust of the health care system, which influences social norms regarding health Additional source of information Social networks provide opportunities to see similar others address health concerns; modeling of how similar others coped with disease burden interpersonal discussions may increase salience of health topics, and a consequence may increase depth of information processing about it Development and reinforcement of social norms Narratives/exemplars Modeling Reinforcement of norms Social organizations (churches, worksites, media institutions, and practices) Framing Worry or fear Cultivates either excessive worry or disregard for cancer risk behaviors Appeal to cultural values Media response to advocacy efforts may disproportionately increase attention to specific issues; advocacy related to disparities has been limited Reliance on powerful sources for news and interpretation Consequences on how the stories on health may or may not appeal to low- versus high-SES groups Health concerns of lower SES get limited attention in the public arena Potential for limited action to ameliorate disease burden among low SES Narratives in TV Repeated exposure Higher than actual portrayal of violence on television engenders belief in a violent world; repeated exposure may serve as a cue and behavioral model Cultivation of TV worldviews Focus on TV violence may increase salience of neighborhood cues that support belief about a violent world, reduced community engagement, reduced opportunities for social exchange, knowledge, reduced physical activity, and other behavior changes Note: SES = socioeconomic status. Open in new tab Table 2 Social Determinants, Message Effects, and Potential Pathways Leading to Health Consequences Structural Factors . Media Effect . Potential Pathways . Examples of Health Consequences . Socioeconomic status Framing Increase attention, salience Low attention to health-related topics and low salience may result in greater “knowledge gaps” on cancer prevention, poor health behaviors, less participation in screening, lower participation in informed decision making, increase in disparities between low and high SES Activate existing schema or develop new schema (new ways looking at a problem and a solution) High attention and salience may engage the audience, greater learning on the part of low SES, potential for behavior change, more informed decision making, reduction or elimination of disparities Lower SES groups may have more difficulties sorting through complex health messages and thus may reject messages or have more difficulty reconciling conflicting messages Narratives/exemplars Liking Greater engagement may lead to greater knowledge, positive role models, greater identification with the messages, more positive health behaviors Attention Retention Absorption/transportation Social networks and social capital Framing Collective identity with the issue or topic Provides an interpretation of health messages within the context of a social relationship; individuals from a lower SEP background may have more exposure to individuals with less health care access and less trust of the health care system, which influences social norms regarding health Additional source of information Social networks provide opportunities to see similar others address health concerns; modeling of how similar others coped with disease burden interpersonal discussions may increase salience of health topics, and a consequence may increase depth of information processing about it Development and reinforcement of social norms Narratives/exemplars Modeling Reinforcement of norms Social organizations (churches, worksites, media institutions, and practices) Framing Worry or fear Cultivates either excessive worry or disregard for cancer risk behaviors Appeal to cultural values Media response to advocacy efforts may disproportionately increase attention to specific issues; advocacy related to disparities has been limited Reliance on powerful sources for news and interpretation Consequences on how the stories on health may or may not appeal to low- versus high-SES groups Health concerns of lower SES get limited attention in the public arena Potential for limited action to ameliorate disease burden among low SES Narratives in TV Repeated exposure Higher than actual portrayal of violence on television engenders belief in a violent world; repeated exposure may serve as a cue and behavioral model Cultivation of TV worldviews Focus on TV violence may increase salience of neighborhood cues that support belief about a violent world, reduced community engagement, reduced opportunities for social exchange, knowledge, reduced physical activity, and other behavior changes Structural Factors . Media Effect . Potential Pathways . Examples of Health Consequences . Socioeconomic status Framing Increase attention, salience Low attention to health-related topics and low salience may result in greater “knowledge gaps” on cancer prevention, poor health behaviors, less participation in screening, lower participation in informed decision making, increase in disparities between low and high SES Activate existing schema or develop new schema (new ways looking at a problem and a solution) High attention and salience may engage the audience, greater learning on the part of low SES, potential for behavior change, more informed decision making, reduction or elimination of disparities Lower SES groups may have more difficulties sorting through complex health messages and thus may reject messages or have more difficulty reconciling conflicting messages Narratives/exemplars Liking Greater engagement may lead to greater knowledge, positive role models, greater identification with the messages, more positive health behaviors Attention Retention Absorption/transportation Social networks and social capital Framing Collective identity with the issue or topic Provides an interpretation of health messages within the context of a social relationship; individuals from a lower SEP background may have more exposure to individuals with less health care access and less trust of the health care system, which influences social norms regarding health Additional source of information Social networks provide opportunities to see similar others address health concerns; modeling of how similar others coped with disease burden interpersonal discussions may increase salience of health topics, and a consequence may increase depth of information processing about it Development and reinforcement of social norms Narratives/exemplars Modeling Reinforcement of norms Social organizations (churches, worksites, media institutions, and practices) Framing Worry or fear Cultivates either excessive worry or disregard for cancer risk behaviors Appeal to cultural values Media response to advocacy efforts may disproportionately increase attention to specific issues; advocacy related to disparities has been limited Reliance on powerful sources for news and interpretation Consequences on how the stories on health may or may not appeal to low- versus high-SES groups Health concerns of lower SES get limited attention in the public arena Potential for limited action to ameliorate disease burden among low SES Narratives in TV Repeated exposure Higher than actual portrayal of violence on television engenders belief in a violent world; repeated exposure may serve as a cue and behavioral model Cultivation of TV worldviews Focus on TV violence may increase salience of neighborhood cues that support belief about a violent world, reduced community engagement, reduced opportunities for social exchange, knowledge, reduced physical activity, and other behavior changes Note: SES = socioeconomic status. Open in new tab Socioeconomic position Some have argued that SES is the “fundamental cause” of health disparities possibly due to unequal distribution of prestige, power, resources, and knowledge translating to inequalities in access to treatment, living in poor neighborhoods and environmental exposures (Phelan, Link, Diez-Roux, Kawachi, & Levin, 2004). Unequal knowledge between low- and high-SES groups has been offered as one reason for health disparities (Link & Phelan, 1995). Research on the “knowledge gap hypothesis” suggests that information is unequally distributed in the social system, with members from higher SES groups having more and accurate information compared to people from lower SES (Tichenor, Donohue, & Olien, 1980; Viswanath & Finnegan, 1996). For example, members of higher SES, in general, learn more from information campaigns compared to people of lower SES groups (Hornik, 2002; Viswanath & Finnegan, 1996). The socioeconomic gradient has been particularly observed related to sustained information flow, such as that found in public health campaigns. Two interrelated sets of explanations at both the individual and structural level have been offered to explain the knowledge gaps: At the individual level, some have observed that relevance and utility for information, motivation, attention, level of controversy about the topic, and interpersonal discussion may be critical in explaining the knowledge gap (Ettema, Brown, & Luepker, 1983). At the structural level, the complexity (size and diversity) of the community, level of controversy, and social conflict about an issue and relevance of topic to a social group (Rucinski, 2004) may explain the knowledge gaps. That is, for example, high levels of social conflict in the community increases media coverage, raises salience, and attracts citizens’ attention to the controversy, thereby equalizing information among both high- and low-SES groups. This dichotomy between individual- and structural-level determinants of knowledge gaps has been difficult to explain deterring any efforts to bridge the gaps. The work in message effects offers a fruitful way to bridge the explanation for knowledge gaps at individual and structural levels by identifying ways to increase message salience and attract audience attention, potentially equalizing media effects across SES groups. We will elaborate this point discussing two message effects, “framing” and “narratives,” drawing examples from current problems in public health. Framing, in theory, may work in a number of ways to equalize knowledge levels across different SES groups. One, media frames influence the way audiences perceive issues, given that frames suggest a way to look at a problem and the solutions to resolve the problem (Entman, 1993; Pan & Kosicki, 2001). For example, the recent attention to the “epidemic of obesity” serves as an excellent example of how media framing may play a role in bridging gaps. The federal government and global health agencies have declared obesity as a crisis warranting preventive action.2 There are active efforts to stem obesity and recently several federal agencies increased research funding for obesity prevention and treatment. In line with these developments, media attention to obesity has been steadily increasing. For example, the number of news stories on obesity in major television networks alone increased by more than 1000% in 5 years: from 5 in 2000, to 12 in 2001, 24 in 2002, 40 in 2003, and 56 in 2004.3 Reports about community actions, such as eliminating vending machines from schools to promotion of physical activity among youth and adults, are beginning to accumulate. One might hypothesize that the framing of obesity as an “epidemic” played a significant role in this social change. This scenario is not dissimilar to how cardiovascular disease emerged as a major health problem in the media in the 1970s and 1980s (Viswanath & Finnegan, 2002). These actions at the level of agencies and the consequent media attention may also influence individual processing of information. Attention and framing in the media may increase salience among individuals, thus leading to greater engagement with the issue. Also, media frames offer a new way of looking at issues or activating existing schema in the audiences’ minds, allowing for processing of information and formation of opinions (Price, Tewksbury, & Powers, 1997) leading to greater learning across different SES groups. Several other topics of relevance to cancer control are influenced by how sponsors and mass media frame the messages: smoking (smokers’ rights vs. second-hand smoke effects), screening behaviors, and cancer treatment (being in a clinical trial vs. being a guinea pig à la Tuskegee study).4 All these topics have been a source of sustained media coverage at one time or other in the past few years, although the impact of use of different frames has not been systematically explored. Framing effects are a product of how media portray an issue and prior schema activated by the frames within the minds of the audience. For example, in the case of Whites, clinical trial frame may activate a schema focused on scientific advances, whereas in African Americans it may activate a schema of racism and denial of treatment. It remains to be seen how media frames will generate attention and activate prior schema and how these could affect disparities in health knowledge and behaviors. On another message format, narratives, most of the limited available evidence suggests that when stories are used to present cultural themes in interventions as a context and background to provide health information and in constructing the messages, they may be more effective (Kreuter & McClure, 2004). For example, Kreuter et al. (2004) conducted a study with African American women to examine whether tailoring on four sociocultural constructs, religiosity, collectivism, racial pride, and time orientation, could enhance the impact of cancer communications. The communications were delivered in the context of a tailored women’s health magazine, and culturally tailored magazines received more attention, were liked better, yielded better retention of information, and were shared with others (Kreuter et al., 2004). Within media studies, culture serves as a resource from which media producers draw their stories and exemplars, and the audience processes information through the filters of culture. As a message format, narratives should be intuitively appealing and effective in processing information, given that they may engender high salience and arousal when properly constructed. Although most studies have used narratives on specific audience segments (e.g., African Americans), what is not clear is the extent to which narratives may interact with social structural variables, such as social class and ethnicity, in reducing communication inequalities in health. More theoretical and comparative work remains to be done in this area. Social networks and social capital The role of social networks and social ties in health promotion has begun to accumulate in such areas as family planning and tobacco use (Valente, Hoffman, Ritt-Olson, Lichtman, & Johnson, 2003). Much attention on health and social ties focuses on social capital—solidary ties and norms of reciprocity that result from social interaction and that supposedly foster collaboration (Putnam, 2000). Social capital is considered a critical moderator between sociodemographic characteristics (e.g. social class) and health outcomes (Hawe & Shiell, 2000; HHoltgrave & Crosby, 2003; Kawachi, Kennedy, & Glass, 1999; Lochner, Kawachi, Brennan, & Buka, 2003; Putnam; Subramanian, Kim, & Kawachi, 2002). Of note, however, is that the causal nature of the relationship as well as its precise role remains a subject of debate (Baum, 1999; Macinko & Starfield, 2001; Mackenbach, 2002; Pearce & Davey Smith, 2003). In addition to the generation of social capital, social networks may also provide information on health or supplement information on health that is available in the secular environment (Viswanath, Randolph, & Finnegan, in press). Health behaviors occur in the context of social norms, acceptable beliefs, and behaviors, and social networks influence what norms are held and acceptable within a group (Valente et al., 2003). It is quite likely that networks could potentially mediate message effects either by providing exemplars or by framing and interpreting health messages in the environment. Interpersonal discussions among network members may increase salience for a health topic, increase the chance of evaluating the information and enhance learning about health and knowledge about the topic (Valente & Saba, 2001). Audience may engage in “reflective integration,” when they deliberate on information that they have attended to and talk to other people about (Kosicki & McLeod, 1990). This is of particular relevance to cancer given the complexity of information processing that people must engage in at different points in the cancer control continuum (Hiatt & Rimer, 1999). Research on how social ties may moderate message effects in different stages of the continuum, from prevention, detection, diagnosis, and treatment, to end of life or survivorship, could be of immense value to both theory building and application. For example, patients are typically under a great deal of stress while they are weighing the pros and cons of different treatment options, and the choices they make may be influenced by experiences of others (Rimer, Briss, Zeller, Chan, & Woolf, 2004). Here, the literature on narratives can make a contribution in interpreting the messages and in helping in decision making. For example, a story by a woman on how she coped with hair loss during chemotherapy may serve as a model and example to other women about to undergo chemotherapy. The prospect of prostate cancer surgery, because of serious side effects such as impotence and incontinence, poses a serious dilemma for many men. Narratives relating the experience of other patients who have gone through the decision-making process may help people make decisions on the surgery. In fact, testimonials and stories have often been used to provide information to patients; theory-based interventions using exemplars or narratives could contribute to more informed decision making (see Rimer et al., 2004, for a discussion on IDM). It is, however, critical to examine the extent to which narratives are representative of audience’s class and cultural backgrounds in addition to individual situations. Social organizations, media institutions, and practices The role of social organizations (e.g., worksites, churches, schools, and health care systems) in health promotion is well documented (Emmons, 2000). Health promotion efforts through organizations provide the opportunity to intervene with large populations of individuals directly as well as indirectly through change in the policies, norms, and cultures of the organization. Such approaches have been successful in a variety of behavioral risk factor areas, including smoking (Jeffery et al., 1994; Sorensen et al., 1996), nutrition (Byers et al., 1995; Sorensen et al., 1999), physical activity (Sorensen et al., 2005), alcohol (Roman & Blum, 2002), HIV (Wilson, Holman, & Hammock, 1996), and cancer screening (Tilley et al., 1999). Here, we offer a brief summary of several different institutional channels that can impact on health outcomes (e.g., health care, churches, worksites, schools), with a particular emphasis on media institutions and their relevance to message effects. Interventions in the health care system can take advantage of the natural focus on health and utilize the relationship a patient has with his or her provider to stimulate preventive health care behaviors. Several studies have shown the health care system to be an effective channel for delivery of health promotion interventions (Curry, Ludman, et al., 2003; Emmons et al., 2005; Kristal, Curry, Shattuck, Feng, & Li, 2000). The national effort in increasing health care providers’ counseling of patients about smoking cessation is an excellent example of what could be accomplished in this setting through a systems-oriented approach (Rigotti et al., 2002; Taylor & Curry, 2004), though provider-delivered interventions may also exacerbate health disparities as some groups do not have access to health services (Institute of Medicine, 2003) or because of patient–provider communication patterns (Cooper & Roter, 2003). Churches have increasingly been used as a channel to reach some groups, given the centrality of the institution in the African American and Hispanic communities. Many of these interventions have been successful in bringing about behavior change in a range of risk behaviors and diseases including smoking, nutrition, obesity, colon cancer prevention, and mammography (Campbell et al., 2004; Darling, Nelson, & Fife, 2004; Derose, Duan, & Fox, 2002; Giarratano, Bustamante-Forest, & Carter, 2005; Peterson, Atwood, & Yates, 2002; Resnicow et al., 2004). Church-based interventions that place health in the context of spirituality and religion are effective in bringing about change (Yanek, Becker, Moy, Gittelsohn, & Koffman, 2001). Much of this work draws on a framing perspective, presenting messages about the importance of health behavior change in the context of spiritual rather than health-related benefits. For example, Wellness for African Americans Through Churches Project focused on improving nutrition, physical activity, and colorectal cancer screening among rural African Americans using tailored approach compared to a lay health advisor and control groups. Tailored, personalized messages included scriptural passages and religious content that supported behaviors targeted in the study. Subjects in the group that received the tailored approach showed significant increases in fruit and vegetable consumption, physical activity, and fecal occult blood testing compared to the lay health group. In addition, the tailored group participants showed greater exposure and recall of intervention messages (Campbell et al., 2004). Similarly, one study reported that “culturally appropriate educational programs to Hispanics have resulted in percentage of Latina women getting mammograms” (Darling et al., 2004). The body of work on interventions using organizations such as worksites, health care systems, churches, and schools as channels for intervention delivery to defined populations (e.g., workers, school personnel, students) is promising. What is not clear is the interaction between message formats and the channels. Several questions remain to be addressed in this realm: What factors at the organizational level are critical to understand the nature of message formats to be chosen for interventions? For example, how does heterogeneity/homogeneity of the occupational distribution within an organization help or hinder message dissemination and processing? What role do peers, colleagues, and fellow members play in the interpretation of messages about health, and what impact does that have on health outcomes? Do the experiences of fellow members and peers serve as exemplars in mediating messages about health? Media organizations, media messages, and relevance for health Although the role of organizations, their characteristics, culture, and policies have been extensively explored in health promotion in cancer control, the role of media organizations has been less extensively studied, even though their importance has been recognized in public health. Specifically, we know less about how the organizational structure of mass media and the occupational practices of media personnel lead to the production of certain message formats and how they may potentially influence health behaviors. Second, the occupational practices of media personnel, journalists, for example, may potentially widen health disparities rather than narrow them as they rely on official sources, scientists, and medical practitioners to report news. Such reliance may influence not only what they report on but also is likely that they report from the perspective of such powerful sources. Third, the language and literacy level of news reports, particularly in the print-based media, may impede low-literacy group from learning from news media. We will elaborate these points next. Within media studies, two bodies of work connecting message production with message effects offer some interesting and useful pointers for future work on message effects in health communication. Our premise is that understanding message effects in public health, especially in a nonintervention context, requires connecting message effects with the message production process (Finnegan & Viswanath, 2002; Shoemaker & Reese, 1996). One body of work stems from the cultivation analysis that examines the impact of television on perceptions of reality, and another examines media sociology and media effects, primarily in public affairs. Work in the area of cultivation analysis, despite the controversy it engendered, offers a useful model for relating media production with message effects. The argument behind the cultivation hypothesis is that heavy exposure to television content, images, ideas, themes, and stories is likely to influence the audience in a number of ways, including its effect on their cognitions and behaviors (Gerbner, Gross, Morgan, & Signorielli, 1980). The hypothesis is that heavy TV exposure often leads individuals to accept the TV-portrayed world as “real,” cultivating a stilted view of the world, especially a “mean-world syndrome” given that so much of televised content, drama, news, and movies, contain violence (Shanahan & Morgan, 1999; Signorielli, 1990). This finding now is well documented, even though the degree of association is attenuated when controlling for other factors (Gerbner et al., 1980; Gunter & Wobner, 1983; Hawkins & Pingree, 1990; Hirsch, 1980; Shanahan & Jones, 1999). Cultivation analysis encourages not only the study of media and message effects but also an analysis of the organizational logic of TV that results in production of specific kinds of messages, as well as a content analysis of television programs to document the amount of violence depicted in the programs. Studies of content analysis of television programs have repeatedly shown that prime time television dramas portray significantly higher levels of violence and crime than is the case in the real world (Gerbner et al., 2002). Television, just as other mass media, thus performs a social control function, defining what is acceptable and normative and what is unacceptable and deviant, in essence defining that violence is normatively acceptable (Shanahan & Morgan, 1999). The consequences of this are not trivial. For example, some recent work has shown a relationship between TV viewing and obesity. One plausible mechanism linking TV with obesity is portrayal of violence on TV as prime time television dramas portray significantly higher levels of violence and crime than is the case in the real world (Gerbner et al., 2002). The consequence is that people who spend more time watching television are more likely to overestimate crime and violence in the real world compared to people who spend less time with TV (Morgan & Shanahan, 1991; Signorielli & Morgan, 1990). The “fear of crime” engendered by heavy TV viewing could deter the viewer from engaging in any physical activity outside the home, thus potentially leading to more TV viewing and a sedentary life. Studies testing the “theories of neighborhood disorder” suggest that a heightened fear of crime in the neighborhood could influence physical activity (King, Stokols, Talen, Brassington, & Killingsworth, 2002; Newman, 1973; Perkins, Wandersman, Rich, & Taylor, 1993; Taylor, 1988). Thus, heightened TV viewing, particularly violent content and perceived disorder in neighborhoods, may lead to lower physical activity. Both obesity and TV viewing are more prevalent among low-SES groups, thus likely be one significant contributor to existing health disparities. Other studies have also documented how media narratives portraying smoking in movies (Dalton et al., 2003) or advertising of junk food on TV (Borzekowski & Robinson, 2001; Jeffrey, McLellarn, & Fox, 1982) could potentially influence health behaviors or viewers. Unfortunately, study of the media production process that leads to such message genres and formats has been limited. A systematic exploration of this area of message effects and production could make a significant contribution to public health communication and suggest ways of intervening at organizational and policy levels. The impact of TV messages on public health including perceptions of violence and rising obesity particularly on individuals living in lower income neighborhoods warrants more systematic work. The effects of exposure to TV genres combined with other daily hassles faced by low-SES groups such as those living in unsafe neighborhoods, lack of access to green space/playgrounds, and lack of choices in the local grocery stores may have multiplicative effect leading to greater burden of obesity among low-SES compared to high-SES groups. News media organizations and message framing Earlier, we discussed the importance of framing and its effects on health behaviors. It is also important to consider how media frames are generated within news organizations and the implications for public health. Media sociology and the impact of media framing on public cognitions, cynicism, and behavior has been one area that has been thoroughly and systematically examined in media studies (Cappella & Jamieson, 1997; Reese, Gandy, & Grant, 2001). Two broad groups of studies explain how the interaction and interrelationship between sources and journalists and the occupational practices of journalists lead to framing effects in news. First, community and advocacy groups have often used media coverage to draw public attention to major health problems and effectuate policy changes in such cases as AIDS, vaccination, binge drinking, drunk driving, and tobacco use. Community activism may attract the attention of journalists for several reasons. In the case of activism, confrontation, the role of drama, the subject of conflict, and a sense of moral injustice could influence reporters to pay attention and provide ideas for story angles (Gamson & Modigliani, 1987; Hilgartner & Bosk, 1988; Olien et al., 1983; Tichenor et al., 1980). Reporters are attracted to stories of outrage and deviance with greater moral clarity, that is, where the definitions of good and evil are clear and unambiguous. Cancer health disparities with their unjustified and disproportionate burden on the lower SES, working poor, racial, and minority ethnic groups provide the necessary moral outrage and, in theory, should be an appealing frame for journalists. Second, in order to make sense of events and structure their occupational routines, journalists rely on certain sources for a steady supply of information or “information subsidies,” which are reported as news (Fishman, 1980; Gandy, 1982). These routines include the beat system, an arrangement where reporters routinely go to certain geographical or positional locations, such as the mayor’s office or the city council, for gathering information. Although beats are common in print media, they are somewhat less so in television and radio where “general assignment” reporters cover a variety of stories. Third, either alone or in combination with the beat system, news or campaign events and news releases are another vital source of information. Reporters rely on news/press releases to be alerted about news and may interview sources as a follow-up or rewrite the news release into a regular story (Stryker, 2002). These occupational practices could sometimes lead to coverage that focuses more on superficial causes of social problems instead of structural determinants, overreliance on powerful sources, and “episodic” approach to coverage rather than consistent follow-up (Cappella & Jamieson, 1997; Iyengar, 1991). This rich tradition has a lot to offer to examine the impact of media framing on public health practices and behaviors (Stryker, 2003; Wallack & Dorfman, 1996; Yanovitzky, 2002). For example, a study of framing analysis focused on activists agitating for funding breast cancer documented the evolution of this issue from a problem of individual women, into one that is viewed as a major public health problem, with considerable resources devoted to it (Kolker, 2004). Media coverage has been reported to be associated with mammography screening, though it is not clear whether it is the sheer amount of attention or the framing of the issue that was influential (Yanovitzky, 2002). The tobacco control movement has been extraordinarily successful in furthering an anti–tobacco use agenda through successful organization and media advocacy (Wolfson, 2001). Recently, the Breast Cancer Coalition has used shareholder activism to attract attention to Avon Products Ltd.’s use of estrogenic chemical preservatives in its products (Fintor, 2003). This brief discussion raises certain questions about the role of framing at the structural level and how intense coverage by media in defining an issue could potentially influence audiences: How do media framing and social construction of cancer control problems influence audience knowledge, attitudes, and behavior about cancer? Are the effects of framing likely to vary by socioeconomic position of the audience as well as their race and ethnic background? To what extent does the selection of frames from mainstream culture apply to and resonate with those groups that may not see themselves as a part of that culture? How do the media frame cancer risks, and to what extent are these risks influenced by the structural conditions of the community (Dunwoody & Griffin, 1999)? For example, is it likely that cancer morbidity and mortality of certain racial and ethnic groups will receive differential attention and framing in the media based on proportion of minorities in the community in which those media are published (Gandy, 1999)? Last, will SES- and race- and ethnicity-based communication inequalities in effects be influenced by the way issues are framed in the media? Conclusions: Relevance for cancer control Healthy People 2010 proposes to eliminate health disparities by 2010. (U.S. Department of Health and Human Services, 2000) The continuing disparities in access to preventive services and treatment among different population subgroups make it unrealistic that we will come anywhere close to this goal in the next few years, even though there is rapid progress in such biomedical scientific areas of pharmacogenomics, genetics, and treatment. In cancer, despite the tremendous progress in basic biomedical sciences and the understanding of carcinogenesis, more than 50% of cancers are attributable to “lifestyle” factors such as tobacco use, energy imbalance (poor diet and sedentary life style), viral infection through sexual activities, alcohol abuse, and exposure to sun (Curry, Byers, & Hewitt, 2003; Hiatt & Rimer, 1999). The National Cancer Institute’s Cancer Control Program Review Group called for basic and behavioral research in population sciences to complement biomedical approaches in reducing cancer risk, morbidity, and mortality. Fundamental to this approach is the concept that cancer control can be advanced only by drawing from different disciplines including social and behavioral sciences such as epidemiology, surveillance, and biomedical sciences (Hiatt & Rimer, 1999), including communication sciences (Viswanath, 2005). In this article, while acknowledging the work done so far on communication in cancer control, we argue that in addition to focusing on individual level of analysis, studies that take into account structural factors that influence health behaviors and message effects can significantly enhance our understanding of human health behavior in a social context and in offering avenues for planned social change. Cancer control strategies seldom attract the attention from the media they deserve possibly because of lack of drama, novelty or confrontation, and complexity (Viswanath, 2005). Similarly, cancer-related disparities attract less attention from either the press or the public. Yet, attracting and sustaining media and audience attention through strategic message design may offer one pathway to media coverage and media effects. In this context, this essay is a theoretical speculation on how literature on message effects and media effects can be bridged with literature from social epidemiology and social ecological model of health to pursue an agenda for a systematic study and promotion of public health communication and the role it can play in the elimination of health disparities. We believe that research melding these two literatures offers great promise in examining the role of communication in cancer control along all stages of the continuum—from prevention, detection, diagnosis, treatment, and survival/end-of-life stages and equalizing services to sections of the population. Notes 1 " We are not suggesting that other social determinants such as social policies or neighborhoods are not important. These three are chosen for illustrative purposes, and lack of space precludes us from elaborating on others. 2 " See Secretary of U.S. Department of Health Human Services, Tommy Thompson’s testimony to the United States House Committee on Energy and Commerce, February 12, 2003. Available at http://energycommerce.house.gov/108/Hearings/02122003hearing786/Thompson1294.htm 3 " Although this is not a systematic content analysis, it does indicate increasing media attention to obesity. Inclusion of print media as well as local television news in an analysis of coverage on obesity may actually provide a more complete and accurate picture on how obesity is evolving into a public health problem. 4 " Tuskegee experiments are infamous experiments where African Americans suffering from syphilis were left untreated. References Agency for Healthcare Research and Quality . ( 2003 ). National healthcare disparities report . Rockville, MD : Agency for Healthcare Research and Quality, U.S. Department of Health and Human Services . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Anderson , C. A. , & Bushman , B. J. ( 2002 ). Psychology. The effects of media violence on society . Science , 295 , 2377 – 2379 . Google Scholar Crossref Search ADS PubMed WorldCat Banks , S. M. , Salovey , P., Greener , S., Rothman , A. J., Moyer , A., Beauvais , J., et al. . ( 1995 ). The effects of message framing on mammography utilization . Health Psychology , 14 ( 2 ), 178 – 184 . Google Scholar Crossref Search ADS PubMed WorldCat Barbeau , E. M. , Krieger , N., & Soobader , M. J. ( 2004 ). Working class matters: Socioeconomic disadvantage, race/ethnicity, gender, and smoking in NHIS 2000 . American Journal of Public Health , 94 , 269 – 278 . Google Scholar Crossref Search ADS PubMed WorldCat Baum , F . ( 1999 ). Social capital: Is it good for your health? Issues for a public health agenda . Journal of Epidemiology and Community Health , 53 ( 4 ), 195 – 196 . Google Scholar Crossref Search ADS PubMed WorldCat Berkman , L. , & Kawachi , I. ( 2000 ). A historical framework for social epidemiology. In L. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 3 – 12 ). New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Borzekowski , D. L. , & Robinson , T. N. ( 2001 ). The 30-second effect: An experiment revealing the impact of television commercials on food preferences of preschoolers . Journal of the American Dietetic Association , 101 ( 1 ), 42 – 46 . Google Scholar Crossref Search ADS PubMed WorldCat Bryant , J. , & Zillman , D. (Eds.). ( 2002 ). Media effects: Advances in theory and research (2nd ed.). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Byers , T. , Mullis , R., Anderson , J., Dusenbury , L., Gorsky , R., Kimber , C., et al. . ( 1995 ). The costs and effects of a nutritional education program following work-site cholesterol screening . American Journal of Public Health , 85 , 650 – 655 . Google Scholar Crossref Search ADS PubMed WorldCat Campbell , M. K. , James , A., Hudson , M. A., Carr , C., Jackson , E., Oakes , V., et al. . ( 2004 ). Improving multiple behaviors for colorectal cancer prevention among African American church members . Health Psychology , 23 , 492 – 502 . Google Scholar Crossref Search ADS PubMed WorldCat Cappella , J. N. , & Jamieson , K. H. ( 1997 ). Spiral of cynicism: The press and the public good . New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cooper , L. , & Roter , D. ( 2003 ). Patient-provider communication: The effect of race and ethnicity on process and outcomes of healthcare. In B. Smedley, A. Stith, & A. Nelson (Eds.), Unequal treatment: Confronting racial and ethnic disparities in healthcare . Washington, DC : The National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Curry , S. J. , Byers , T., & Hewitt , M. (Eds.). ( 2003 ). Fulfilling the potential of cancer prevention and early detection . Washington, DC : Institute of Medicine, National Research Council, The National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Curry , S. J. , Ludman , E. J., Graham , E., Stout , J., Grothaus , L., & Lozano , P. ( 2003 ). Pediatric-based smoking cessation intervention for low-income women: A randomized trial . Archives of Pediatrics and Adolescent Medicine , 157 , 295 – 302 . Google Scholar Crossref Search ADS PubMed WorldCat Dalton , M. A. , Sargent , J. D., Beach , M. L., Titus-Ernstoff , L., Gibson , J. J., Ahrens , M. B., et al. . ( 2003 ). Effect of viewing smoking in movies on adolescent smoking initiation: A cohort study . Lancet , 362 , 281 – 285 . Google Scholar Crossref Search ADS PubMed WorldCat Darling , C. M. , Nelson , C. P., & Fife , R. S. ( 2004 ). Improving breast health education for Hispanic women . Journal of the American Medical Women’s Association , 59 ( 3 ), 171, 228 – 229 . OpenURL Placeholder Text WorldCat Davey , H. M. , Butow , P. N., & Armstrong , B. K. ( 2003 ). Cancer patients’ preferences for written prognostic information provided outside the clinical context . British Journal of Cancer , 89 , 1450 – 1456 . Google Scholar Crossref Search ADS PubMed WorldCat Davin , S . ( 2003 ). Healthy viewing: The reception of medical narratives . Sociology of Health and Illness , 25 , 662 – 679 . Google Scholar Crossref Search ADS PubMed WorldCat Derose , K. P. , Duan , N. H., & Fox , S. A. ( 2002 ). Women’s receptivity to church-based mobile mammography . Journal of Health Care for the Poor and Underserved , 13 , 199 – 213 . Google Scholar Crossref Search ADS PubMed WorldCat Detweiler , J. B. , Bedell , B. T., Salovey , P., Pronin , E., & Rothman , A. J. ( 1999 ). Message framing and sunscreen use: Gain-framed messages motivate beach-goers . Health Psychology , 18 ( 2 ), 189 – 196 . Google Scholar Crossref Search ADS PubMed WorldCat Dunwoody , S. , & Griffin , R. J. ( 1999 ). Structural pluralism and media accounts of risk. In D. Demers & K. Viswanath (Eds.), Mass media, social control, and social change: A macrosocial perspective . Ames : University of Iowa Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Emmons , K. M . ( 2000 ). Behavioral and social science contributions to the health of adults in the United States. In B. D. Smedley & S. L. Syme (Eds.), Promoting health: Intervention strategies from social and behavioral research (pp. 254 – 321 ). Washington, DC : National Academy Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Emmons , K. M. , Stoddard , A. M., Fletcher , R., Gutheil , C., Suarez , E. G., Lobb , R., et al. . ( 2005 ). Cancer prevention among working class, multiethnic adults: Results of the healthy directions—health centers study . American Journal of Public Health , 95 , 1200 – 1205 . Google Scholar Crossref Search ADS PubMed WorldCat Entman , R. M . ( 1993 ). Framing: Towards clarification of a fractured paradigm . Journal of Communication , 43 ( 4 ), 51 – 58 . Google Scholar Crossref Search ADS WorldCat Ettema , J. S. , Brown , J. W., & Luepker , R. V. ( 1983 ). Knowledge gap effects in a health information campaign . Public Opinion Quarterly , 47 , 516 – 527 . Google Scholar Crossref Search ADS PubMed WorldCat Finnegan , J. R. , & Viswanath , K. ( 2002 ). Communication theory and health behavior change: The media studies framework. In K. Glanz, F. M. Lewis, & B. K. Rimer (Eds.), Health behavior and health education: Theory, research, and practice . San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Finnegan , J. R. , Viswanath , K., Kahn , E., & Hannan , P. ( 1993 ). Exposure to the sources of heart disease prevention information: Community type and social group differences . Journalism Quarterly , 70 , 569 – 584 . Google Scholar Crossref Search ADS WorldCat Finney , L. J. , & Iannotti , R. J. ( 2002 ). Message framing and mammography screening: A theory-driven intervention . Behavioral Medicine , 28 ( 1 ), 5 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat Fintor , L . ( 2003 ). Breast cancer group targets ‘shareholder activism’ at Avon . Journal of the National Cancer Institute , 95 , 1106 – 1108 . Google Scholar Crossref Search ADS PubMed WorldCat Fishman , M . ( 1980 ). Manufacturing the news . Austin : University of Texas Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gamson , W. A . ( 2001 ). Foreword. In S. D. Reese, J. R. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on our understanding of the social world . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gamson , W. A. , & Modigliani , A. ( 1987 ). The changing culture of affirmative action. In R. D. Braungart (Ed.), Research in political sociology (Vol. 3 , pp. 137 – 177 ). Greenwich, CT : JAI Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gamson , W. A. , & Modigliani , A. ( 1989 ). Media discourse and public opinion on nuclear power . American Journal of Sociology , 95 ( 1 ), 1 – 37 . Google Scholar Crossref Search ADS WorldCat Gandy , O. H . ( 1982 ). Beyond agenda setting: Information subsidies and public policy . Norwood, NJ : Ablex Publishing . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gandy , O. H . ( 1999 ). Community Pluralism and the “tipping point”: Editorial responses to race and related structural change. In D. Demers & K. Viswanath (Eds.), Mass media, social control, and social change: A macrosocial perspective (pp. 159 – 182 ). Ames : Iowa State University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gaziano , C. , & McGrath , K. ( 1986 ). Measuring the concept of credibility . Journalism Quarterly , 63 , 451 – 462 . Google Scholar Crossref Search ADS WorldCat Gerbner , G. , Gross , L., Morgan , M., & Signorielli , N. ( 1980 ). The “mainstreaming” of America: Violence profile no. 11 . Journal of Communication , 30 ( 3 ), 10 – 29 . Google Scholar Crossref Search ADS WorldCat Gerbner , G. , Gross , L., Morgan , M., Signorielli , N., & Shanahan , J. ( 2002 ). Growing up with television: Cultivation processes. In J. Bryant & D. Zillman (Eds.), Media effects: Advances in theory research (2nd ed., pp. 43 – 68 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Giarratano , G. , Bustamante-Forest , R., & Carter , C. ( 2005 ). A multicultural and multilingual outreach program for cervical and breast cancer screening . Journal of Obstetric Gynecologic and Neonatal Nursing , 34 , 395 – 402 . Google Scholar Crossref Search ADS WorldCat Gilpin , E. A. , Lee , L., & Pierce , J. P. ( 2004 ). Changes in population attitudes about where smoking should not be allowed: California versus the rest of the USA . Tobacco Control , 13 ( 1 ), 38 – 44 . Google Scholar Crossref Search ADS PubMed WorldCat Green , M. C . ( 2006 ). Narratives and cancer communication . Journal of Communication , 56 , S163 – S183 . Google Scholar Crossref Search ADS WorldCat Green , M. C. , & Brock , T. C. ( 2000 ). The role of transportation in the persuasiveness of public narratives . Journal of Personality and Social Psychology , 79 , 701 – 721 . Google Scholar Crossref Search ADS PubMed WorldCat Gunter , B. , & Wobner , M. ( 1983 ). Television viewing and public perceptions of hazards to life . Journal of Environmental Psychology , 3 , 325 – 335 . Google Scholar Crossref Search ADS WorldCat Hawe , P. , & Shiell , A. ( 2000 ). Social capital and health promotion: A review . Social Science and Medicine , 51 , 871 – 885 . Google Scholar Crossref Search ADS PubMed WorldCat Hawkins , R. B. , & Pingree , S. ( 1990 ). Divergent psychological processes in constructing social reality from mass media content. In N. Signorielli & M. Morgan (Eds.), Cultivation analysis: New directions in media effects research (pp. 33 – 50 ). Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hiatt , R. A. , & Rimer , B. K. ( 1999 ). A new strategy for cancer control research . Cancer, Epidemiology, Biomarkers and Prevention , 8 , 957 – 964 . OpenURL Placeholder Text WorldCat Hilgartner , S. , & Bosk , C. L. ( 1988 ). The rise and fall of social problems—A public arenas model . American Journal of Sociology , 94 ( 1 ), 53 – 78 . Google Scholar Crossref Search ADS WorldCat Hirsch , P . ( 1980 ). The scary world of the nonviewer and other anomalies: A reanalysis of Gerbner et al.’s findings on cultivation analysis . Communication Research , 7 , 403 – 456 . Google Scholar Crossref Search ADS WorldCat Hoffman-Goetz , L . ( 1999 ). Cancer experiences of African-American women as portrayed in popular mass magazines . Psycho-oncology , 8 ( 1 ), 36 – 45 . Google Scholar Crossref Search ADS PubMed WorldCat Holtgrave , D. R. , & Crosby , R. A. ( 2003 ). Social capital, poverty, and income inequality as predictors of gonorrhea, syphilis, chlamydia and AIDS case rates in the United States . Sexually Transmitted Infections , 79 ( 1 ), 62 – 64 . Google Scholar Crossref Search ADS PubMed WorldCat Hornik , R . ( 2002 ). Public health communication: Making sense of contradictory evidence. In R. Hornik (Ed.), Public health communication: Evidence for behavior change . New York : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Institute of Medicine . ( 1999 ). The unequal burden of cancer . Washington, DC : National Academy Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Institute of Medicine . ( 2003 ). Unequal treatment: Confronting racial and ethnic disparities in health care . Washington, DC : National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Institute of Medicine & Committee on Communication for Behavior Change in the 21st Century: Improving the Health of Diverse Populations . ( 2002 ). Speaking of health: Assessing health communication strategies for diverse populations . Washington, DC : National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Iyengar , S . ( 1991 ). Is anyone responsible? How television frames public issues . Chicago : University of Chicago Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Jeffery , R. W. , Kelder , S. H., Forster , J. L., French , S. A., Lando , H. A., & Baxter , J. E. ( 1994 ). Restrictive smoking policies in the workplace: Effects on smoking prevalence and cigarette consumption . Preventive Medicine , 23 ( 1 ), 78 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat Jeffrey , D. B. , McLellarn , R. W., & Fox , D. T. ( 1982 ). The development of children’s eating habits: The role of television commercials . Health Education Quarterly , 9 ( 2/3 ), 174 – 189 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Kaplan , G. A . ( 2004 ). What’s wrong with social epidemiology, and how can we make it better? Epidemiologic Reviews , 26 , 124 – 135 . Google Scholar Crossref Search ADS PubMed WorldCat Kawachi , I. , & Berkman , L. ( 2000 ). Social cohesion, social capital, and health. In L. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 174 – 190 ). New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kawachi , I. , Kennedy , B. P., & Glass , R. ( 1999 ). Social capital and self-rated health: A contextual analysis . American Journal of Public Health , 89 , 1187 – 1193 . Google Scholar Crossref Search ADS PubMed WorldCat King , A. C. , Stokols , D., Talen , E., Brassington , G. S., & Killingsworth , R. ( 2002 ). Theoretical approaches to the promotion of physical activity: Forging a transdisciplinary paradigm . American Journal of Preventive Medicine , 23 ( 2 Suppl .), 15 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat Kolker , E. S . ( 2004 ). Framing as a cultural resource in health social movements: Funding activism and the breast cancer movement in the US 1990-1993 . Sociology of Health and Illness , 26 , 820 – 844 . Google Scholar Crossref Search ADS PubMed WorldCat Kosicki , G. M. , & McLeod , J. M. ( 1990 ). Learning from political news: Effects of media images and information processing strategies. In S. Kraus (Ed.), Mass communication and political information processing . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kreuter , M. W. , & McClure , S. M. ( 2004 ). The role of culture in health communication . Annual Review of Public Health , 25 , 439 – 455 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Skinner , C. S., Steger-May , K., Holt , C. L., Bucholtz , D. C., Clark , E. M., et al. . ( 2004 ). Responses to behaviorally vs culturally tailored cancer communication among African American women . American Journal of Health Behavior , 28 , 195 – 207 . Google Scholar Crossref Search ADS PubMed WorldCat Krieger , N . ( 2001 ). Historical roots of social epidemiology: Socioeconomic gradients in health and contextual analysis . International Journal of Epidemiology , 30 , 899 – 900 . Google Scholar Crossref Search ADS PubMed WorldCat Kristal , A. R. , Curry , S. J., Shattuck , A. L., Feng , Z., & Li , S. ( 2000 ). A randomized trial of a tailored, self-help dietary intervention: The Puget Sound Eating Patterns study . Preventive Medicine , 31 , 380 – 389 . Google Scholar Crossref Search ADS PubMed WorldCat Kwak , N . ( 1999 ). Revisiting the knowledge gap hypothesis: Education, motivation, and media use . Communication Research , 26 , 385 – 413 . Google Scholar Crossref Search ADS WorldCat Link , B. G. , & Phelan , J. ( 1995 ). Social conditions as fundamental causes of disease . Journal of Health and Social Behavior , 35 , 80 – 94 . Google Scholar Crossref Search ADS WorldCat Linnan , L. A. , Emmons , K. M., & Abrams , D. B. ( 2002 ). Beauty and the beast: Results of the Rhode Island smoke free shop initiative . American Journal of Public Health , 92 ( 1 ), 27 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat Lochner , K. A. , Kawachi , I., Brennan , R. T., & Buka , S. L. ( 2003 ). Social capital and neighborhood mortality rates in Chicago . Social Science and Medicine , 56 , 1797 – 1805 . Google Scholar Crossref Search ADS PubMed WorldCat Lynch , J. , & Kaplan , G. ( 2000 ). Socioeconomic position. In L. Berkman & I. Kawachi (Eds.), Social epidemiology (pp. 13 – 35 ). New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lynch , J. W. , Smith , G. D., Kaplan , G. A., & House , J. S. ( 2000 ). Income inequality and mortality: Importance to health of individual income, psychosocial environment, or material conditions . BMJ , 320 , 1200 – 1204 . Google Scholar Crossref Search ADS PubMed WorldCat Macinko , J. , & Starfield , B. ( 2001 ). The utility of social capital in research on health determinants . Milbank Quarterly , 79 , 387 – 427 , IV. Google Scholar Crossref Search ADS PubMed WorldCat Mackenbach , J. P . ( 2002 ). Income inequality and population health: Evidence favoring a negative correlation between income inequality and life expectancy has disappeared . BMJ , 324 ( 7328 ), 1 – 2 . Google Scholar Crossref Search ADS PubMed WorldCat McLeod , J. M. , Kosicki , G. M., & Pan , Z. ( 1991 ). On understanding and misunderstanding media effects. In J. Curran & M. Gurevitch (Eds.), Mass media and society (pp. 235 – 266 ). New York : Edward Arnold . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Morgan , J. , & Shanahan , J. ( 1991 ). Do VCR’s change the TV picture? VCR’s and the cultivation process . American Behavioral Scientist , 35 , 122 – 135 . Google Scholar Crossref Search ADS WorldCat National Center for Health Statistics . ( 2004 ). Health, United States, 2004, with chart book on trends in the health of Americans . Retrieved February 2, 2005, from http://www.cdc.gov/nchs/hus.htm Newman , O . ( 1973 ). Defensible space . New York : Macmillan . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nielsen-Bohlman , L. , Panzer , A. M., & Kindig , D. A. (Eds.). ( 2004 ). Health literacy: A prescription to end confusion . Washington, DC : Institute of Medicine, National Research Council, The National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Olien , C. N. , Donohue , G. A., & Tichenor , P. J. ( 1983 ). Structure, communication, and social power: Evolution of the knowledge gap hypothesis. In E. Wartella & D. C. Whitney (Eds.), Mass communication review yearbook 4 (pp. 455 – 462 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pan , Z. , & Kosicki , G. M. ( 2001 ). Framing as a strategic action in public deliberation. In S. D. Reese, O. H. Gandy, & A. E. Grant (Eds.), Framing public life: Perspectives on media and our understanding of the social world (pp. 35 – 66 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pearce , N. , & Davey Smith , G. ( 2003 ). Is social capital the key to inequalities in health? American Journal of Public Health , 93 ( 1 ), 122 – 129 . Google Scholar Crossref Search ADS PubMed WorldCat Perkins , D. , Wandersman , A., Rich , R., & Taylor , R. ( 1993 ). The physical environment of street crime: Defensible space, territoriality, and incivilities . Journal of Environmental Psychology , 13 , 29 – 49 . Google Scholar Crossref Search ADS WorldCat Peterson , J. , Atwood , J. R., & Yates , B. ( 2002 ). Key elements for church-based health promotion programs: Outcome-based literature review . Public Health Nursing , 19 , 401 – 411 . Google Scholar Crossref Search ADS PubMed WorldCat Phelan , J. C. , Link , B. G., Diez-Roux , A., Kawachi , I., & Levin , B. ( 2004 ). “Fundamental causes” of social inequalities in mortality: A test of the theory . Journal of Health and Social Behavior , 45 , 265 – 285 . Google Scholar Crossref Search ADS PubMed WorldCat Price , V. , Tewksbury , D., & Powers , E. ( 1997 ). Switching trains of thought: The impact of news frames on readers cognitive responses . Communication Research , 24 , 481 – 506 . Google Scholar Crossref Search ADS WorldCat Putnam , R . ( 2000 ). Bowling alone: The collapse and revival of the American community . New York : Simon & Schuster . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Reese , S. D. , Gandy , O. H., & Grant , A. E. (Eds.). ( 2001 ). Framing public life: Perspectives on media and our understanding of the social world . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Resnicow , K. , Campbell , M. K., Carr , C., McCarty , F., Wang , T., Periasamy , S., et al. . ( 2004 ). Body and soul. A dietary intervention conducted through African American churches . American Journal of Preventive Medicine , 27 ( 2 ), 97 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Rigotti , N. A. , Quinn , V. P., Stevens , V. J., Solberg , L. I., Hollis , J. F., Rosenthal , A. C., et al. . ( 2002 ). Tobacco-control policies in 11 leading managed care organizations: Progress and challenges . Effective Clinical Practice , 5 ( 3 ), 130 – 136 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Rimer , B. K. , Briss , P. A., Zeller , P. K., Chan , E. C., & Woolf , S. H. ( 2004 ). Informed decision making: What is its role in cancer screening? Cancer , 101 ( Suppl .), 1214 – 1228 . Google Scholar Crossref Search ADS PubMed WorldCat Roman , P. M. , & Blum , T. C. ( 2002 ). The workplace and alcohol problem prevention . Alcohol Research & Health , 26 ( 1 ), 49 – 57 . OpenURL Placeholder Text WorldCat Romantan , A. , Hornik , R., Weiner , J., Price , V., Cappella , J., & Viswanath , K. ( 2005 , January 2021). Learning about cancer: A comparative analysis of the performance of eight measures of incidental exposure to cancer information in the mass media . Paper presented at the HINTS Data Users Conference , St. Pete Beach, FL . Rothman , A. J. , Bartels , R. D., Wlaschin , J., & Salovey , P. ( 2006 ). The strategic use of gain and loss framed messages to promote healthy behavior: How theory can inform practice . Journal of Communication , 56 , S202 – S220 . Google Scholar Crossref Search ADS WorldCat Rucinski , D . ( 2004 ). Community boundedness, personal relevance, and the knowledge gap . Communication Research , 31 ( 4 ), 472 – 495 . Google Scholar Crossref Search ADS WorldCat Salovey , P. , Schneider , T. R., & Apanovitch , A. M. ( 2002 ). Message framing in the prevention and early detection of illness. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook: Theory and practice (pp. 391 – 406 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Salovey , P. , & Williams-Piehota , P. ( 2004 ). Field experiments in social psychology: Message framing and the promotion of health protective behaviors . American Behavioral Scientist , 47 , 488 – 505 . Google Scholar Crossref Search ADS WorldCat Scheufele , D. A . ( 1999 ). Framing theory of media effects . Journal of Communication , 49 ( 1 ), 103 – 122 . Google Scholar Crossref Search ADS WorldCat Schneider , E. C. , Zaslavsky , A. M., & Epstein , A. M. ( 2002 ). Racial disparities in quality of care for enrollees in medicare managed care . JAMA , 287 , 1288 – 1294 . Google Scholar Crossref Search ADS PubMed WorldCat Schneider , T. R. , Salovey , P., Apanovitch , A. M., Pizarro , J., McCarthy , D., Zullo , J., et al. . ( 2001 ). The effects of message framing and ethnic targeting on mammography use among low-income women . Health Psychology , 20 , 256 – 266 . Google Scholar Crossref Search ADS PubMed WorldCat Shanahan , J. , & Jones , V. ( 1999 ). Cultivation and social context. In D. Demers & K. Viswanath (Eds.), Mass media and social control (pp. 31 – 50 ). Ames : Iowa University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Shanahan , J. , & Morgan , M. ( 1999 ). Television and its viewers: Cultivation theory and research . Cambridge, UK : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Shoemaker , P. , & Reese , S. ( 1996 ). Mediating the message: Theories of influences on mass media content (2nd ed.). New York : Longman . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Signorielli , N . ( 1990 ). Television’s mean and dangerous world: A continuation of the cultural indicators perspective. In N. Signorielli & M. Morgan (Eds.), Cultivation analysis: New directions in media effects research (pp. 85 – 106 ). Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Signorielli , N. , & Morgan , M. (Eds.). ( 1990 ). Cultivation analysis: New directions in media effects research . Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Slater , M. D . ( 2002 ). Entertainment education and the persuasive impact of narratives. In M. C. Green, J. F. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 157 – 181 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smedley , B. D. , & Syme , S. L. ( 2000 ). Promoting health: Intervention strategies from social and behavioral research. In B. D. Smedley & S. L. Syme (Eds.), Promoting Health: Intervention strategies from social and behavioral research . (pp. 254 – 321 ). Washington, DC : National Academy Press . Sorensen , G. , Barbeau , E., Stoddard , A. M., Hunt , M. K., Kaphingst , K., & Wallace , L. ( 2005 ). Promoting behavior change among working-class, multiethnic workers: Results of the healthy directions—Small business study . American Journal of Public Health , 95 , 1389 – 1395 . Google Scholar Crossref Search ADS PubMed WorldCat Sorensen , G. , Emmons , K., Hunt , M. K., Barbeau , E., Goldman , R., Peterson , K., et al. . ( 2003 ). Model for incorporating social context in health behavior interventions: Applications for cancer prevention for working-class, multiethnic populations . Preventive Medicine , 37 , 188 – 197 . Google Scholar Crossref Search ADS PubMed WorldCat Sorensen , G. , Stoddard , A., Peterson , K., Cohen , N., Hunt , M. K., Stein , E., et al. . ( 1999 ). Increasing fruit and vegetable consumption through work sites and families in the treatwell 5-a-day study . American Journal of Public Health , 89 ( 1 ), 54 – 60 . Google Scholar Crossref Search ADS PubMed WorldCat Sorensen , G. , Thompson , B., Glanz , K., Feng , Z., Kinne , S., DiClemente , C., et al. . ( 1996 ). Worksite-based cancer prevention: Primary results from the Working Well trial . American Journal of Public Health , 86 , 939 – 947 . Google Scholar Crossref Search ADS PubMed WorldCat Stryker , J. E . ( 2002 ). Reporting medical information: Effects of press releases and newsworthiness on medical journal articles’ visibility in the news media . Preventive Medicine , 35 , 519 – 530 . Google Scholar Crossref Search ADS PubMed WorldCat Stryker , J. E . ( 2003 ). Media and marijuana: A longitudinal analysis of news media effects on adolescents’ marijuana use and related outcomes, 1977-1999 . Journal of Health Communication , 8 , 305 – 328 . Google Scholar Crossref Search ADS PubMed WorldCat Subramanian , S. V. , Kim , D. J., & Kawachi , I. ( 2002 ). Social trust and self-rated health in US communities: A multilevel analysis . Journal of Urban Health , 79 ( 4 Suppl. 1 ), S21 – S34 . Google Scholar Crossref Search ADS PubMed WorldCat Taylor , R . ( 1988 ). Human territorial functioning . New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Taylor , C. B. , & Curry , S. J. ( 2004 ). Implementation of evidence-based tobacco use cessation guidelines in managed care organizations . Annals of Behavioral Medicine , 27 ( 1 ), 13 – 21 . Google Scholar Crossref Search ADS PubMed WorldCat Tichenor , P. J. , Donohue , G. A., & Olien , C. N. ( 1980 ). Community conflict and the press . Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tilley , B. C. , Vernon , S. W., Myers , R., Glanz , K., Lu , M., Hirst , K., et al. . ( 1999 ). The Next Step Trial: Impact of a worksite colorectal cancer screening promotion program . Preventive Medicine , 28 , 276 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat U.S. Department of Health and Human Services . ( 2000 ). Healthy People 2010 . Available at http://www.healthypeople.gov/Document/tableofcontents.htm#under. Valente , T. W. , Hoffman , B. R., Ritt-Olson , A., Lichtman , K., & Johnson , C. A. ( 2003 ). Effects of a social-network method for group assignment strategies on peer-led tobacco prevention programs in schools . American Journal of Public Health , 93 , 1837 – 1843 . Google Scholar Crossref Search ADS PubMed WorldCat Valente , T. W. , & Saba , W. P. ( 2001 ). Campaign exposure and interpersonal communication as factors in contraceptive use in Bolivia . Journal of Health Communication , 6 , 303 – 322 . Google Scholar Crossref Search ADS PubMed WorldCat Viswanath , K . ( 2005 ). The communications revolution and cancer control . Nature Reviews. Cancer , 5 , 828 – 835 . Google Scholar Crossref Search ADS PubMed WorldCat Viswanath , K. , Breen , N., Meissner , H., Moser , R. P., Hesse , B, Steele , W. R., et al. . ( 2006 ). Cancer knowledge and displarities in the information age . Journal of Health Communication , 11 ( Suppl. 1 ), 1 – 17 . Google Scholar Crossref Search ADS PubMed WorldCat Viswanath , K . ( in press ). Public communications and its role in reducing and eliminating health disparities. In G. E. Thomson, F. Mitchell, & M. Williams (Eds.), Examining the health disparities research plan of the National Institutes of Health: Unfinished business (pp. 231 – 270 ). Washington, DC : Institute of Medicine . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Viswanath , K. , & Demers , D. ( 1999 ). Mass media from a macrosocial perspective. In D. Demers & K. Viswanath (Eds.), Mass media, social control and social change: A macrosocial perspective (pp. 3 – 28 ). Ames : Iowa State University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Viswanath , K. , & Finnegan , J. R. ( 1996 ). The knowledge gap hypothesis: Twenty five years later. In B. Burleson (Ed.), Communication yearbook 19 (pp. 187 – 227 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Viswanath , K. , & Finnegan , J. R. ( 2002 ). Community health campaigns and secular trends: Insights from the Minnesota Heart Health Program and Community Trials in Heart Disease Prevention. In R. Hornik (Ed.), Public health communication: Evidence for behavior change (pp. 289 – 312 ). New York : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Viswanath , K. , Finniegan , J. R., Hannan , P. J., & Luepker , R. V. ( 1991 ). Health and knowledge gaps: Some lessons from the Minnesota Heart Health Program . American Behavioral Scientist , 34 , 712 – 726 . Google Scholar Crossref Search ADS WorldCat Viswanath , K. , Randolph , W., & Finnegan , J. R. ( in press ). Connecting the dots between social capital and public health outcomes: The role of communication and community pluralism . American Journal of Public Health . OpenURL Placeholder Text WorldCat Wallack , L. , & Dorfman , L. ( 1996 ). Media advocacy: A strategy for advancing policy and promoting health . Health Education Quarterly , 23 , 293 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat Wilson , M. G. , Holman , P. B., & Hammock , A. ( 1996 ). A comprehensive review of the effects of worksite health promotion on health-related outcomes . American Journal of Health Promotion , 10 , 429 – 435 . Google Scholar Crossref Search ADS PubMed WorldCat Wolfson , M . ( 2001 ). The fight against big tobacco: The movement, the state, and the public’s health . New York : Aldine de Gruyter . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Yanek , L. R. , Becker , D. M., Moy , T. F., Gittelsohn , J., & Koffman , D. M. ( 2001 ). Project Joy: Faith based cardiovascular health promotion for African American women . Public Health Reports , 116 ( Suppl. 1 ), 68 – 81 . Google Scholar Crossref Search ADS PubMed WorldCat Yanovitzky , I . ( 2002 ). Effect of news coverage on the prevalence of drunk-driving behavior: Evidence from a longitudinal study . Journal of Studies on Alcohol , 63 , 342 – 351 . Google Scholar Crossref Search ADS PubMed WorldCat © 2006 International Communication Association

Rimer, Barbara, K.;Kreuter, Matthew, W.

doi: 10.1111/j.1460-2466.2006.00289.xpmid: N/A

Abstract Tailored health communication (THC) is any combination of information and behavior change strategies intended to reach one specific person based on information unique to that person, related to the outcome of interest, and derived from an individual assessment. THCs have been studied as a means to facilitate behavior change by influencing some key intermediate steps that precede the behavioral outcome. These include the extent to which people attend to communications, think about them, find them relevant and salient, and intend to take action. Evidence shows that THCs have achieved modest success in changing a number of cancer-related behaviors, including smoking, diet, exercise, and cancer screening. However, it is likely that THCs could be more effective if they were developed with a greater understanding of message effects and what we refer to as the behavioral pathway. Instead of using unidimensional approaches to influence behavior change, a message effects approach would help researchers identify key leverage points for impact on such intermediate outcomes as persuasion and yielding. Such a strategy also might be used to determine when THCs are the preferred approach and when generic, targeted, or combinations of THCs and targeted communications might be appropriate. Viewing THCs from the perspective of the behavioral pathway might indicate use of different messages, sources, and formats to influence different people at different points on the pathway. We provide a brief history of THCs and suggest how integrating a broader perspective of health behavior and health communication theories could enrich THCs. What is tailored health communication? Tailoring is a process for creating individualized communications. It is an assessment-based approach in which data from or about a specific individual and related to a given health outcome are used to determine the most appropriate information or strategies to meet that person’s unique needs (Kreuter, Farrell, Olevitch, & Brennan, 1999; Kreuter & Skinner, 2000; Kreuter, Strecher, & Glassman, 1999; Rimer & Glassman, 1998). To date, researchers have usually inferred these needs based on an individual’s responses to assessment items measuring behavioral determinants. Ideally, the concept of “needs” might be expanded to also include specific message tactics or other elements of communication likely to enhance effectiveness for a given person. Tailored health communications (THCs) may be produced in any media and in nearly limitless formats, for example, letters, booklets, and calendars. Because tailored print materials have been the format studied most extensively to date, many of the examples here will draw upon this literature. THCs usually are personalized, but merely being personalized is not sufficient to consider them THCs. A brief history of tailored health communication Through the early to mid-1980s, most printed health information was mass produced in brochures, pamphlets, and booklets for undifferentiated audiences. The high cost of printing made it necessary to print many copies of materials to be cost effective. Thus, developers sought to produce materials and programs that would meet the needs of many potential users. As behavioral scientists grew increasingly aware of the heterogeneity within mass audiences and learned more about methods used in communication and other fields to create persuasive messages (Prochaska, Redding, & Evers, 2002; Rimer & Glassman, 1998; Slater & Flora, 1991), they began to develop different versions of materials for distinct population subgroups. Although principles of audience segmentation had been well known to communication scholars and widely used in advertising and marketing, including social marketing (Maibach, Rothschild, & Novelli, 2002; Pepper, 2000), their application to creating targeted health education materials was innovative. In early examples of this work, audience segmentation based on demographic and behavioral variables was used to design smoking cessation self-help guides for blue-collar and minority smokers (Strecher, Rimer, & Monaco, 1989), African Americans (Robinson, Orleans, James, & Sutton, 1992), older smokers (Rimer et al., 1994), pregnant women (S. W. Davis, Cummings, Rimer, Sciandra, & Stone, 1992), and women with young children (Keintz, Fleisher, & Rimer, 1994). Such segmentation generally was based on principles of social marketing and on the assumption that important differences between groups could be addressed by creating different versions of health communications. Decisions about what information was provided to which groups usually was based on audience research guided by theories used most frequently in that period (e.g., stages of change [SOC; Prochaska, 2004; Prochaska et al., 2002] and the Health Belief Model [Janz, Champion, & Strecher, 2002]). Early stage-based behavior change materials like Freedom From Smoking For You and Your Family (Strecher et al., 1989) foreshadowed tailoring by permitting users to select from a self-help smoking cessation guide those chapters that applied to their particular stage of readiness to quit. Clear Horizons, developed by Rimer et al. (1994), used a similar approach and was significantly more effective than generic smoking cessation materials for smokers aged 50 and older. The emergence of tailored health communication was part of a growing marketing approach to customize information. Like many scientific advances, the rise of THC was made possible by the convergence of several innovations that coincided in time (Rogers, 2003). One was the trend toward customization of health and other information. Another was the growing use of the Transtheoretical Model that described “stages of change” in an individual’s readiness to adopt or modify health behaviors (Prochaska, 2004; Prochaska, DiClemente, Velicer, & Rossi, 1993; Prochaska et al., 2002). Its SOC variable identified discrete population subgroups (those in precontemplation, contemplation, preparation, action, and maintenance) that could be targets for change using targeted approaches (focusing on the group), tailored (individual) approaches, or combinations of these (Rakowski et al., 2003). Its processes of change specified different ways in which change could be facilitated for individuals or groups. Ultimately, different message topics could be introduced and addressed based on knowing a person’s SOC for a given behavior, along with other relevant information. Although there are other stage models, none have achieved the level of dissemination of SOC. The model’s utility for intervention planning and delivery facilitated its adoption among practitioners and researchers developing prevention programs and messages (Samuelson, 1997). In most early studies of THC, stage of readiness was a central tailoring variable (Brug, Campbell, & van Assema, 1999; Campbell et al., 1994; Prochaska, et al., 1993; Rakowski et al., 2003; Rimer et al., 2001; Skinner, Strecher, & Hospers, 1994; Strecher, 1999). Population-wide customization of health information to different individuals was dependent upon processing large volumes of data quickly. Major improvements in computing provided health communication researchers the tools to create customized interventions. Also, producers had to be able to create and print attractive documents at affordable cost. As color printers became cheaper, the tools for customization became even more accessible. Computers first were used this way to create health risk appraisals (HRAs), which calculated an individual’s mortality risk based on his or her risk factor profile and algorithms derived from national data that weighted the relative importance of each risk factor to leading causes of mortality (Wagner, Beery, Schoenbach, & Graham, 1982). In most cases, however, the behavior change recommendations provided to HRA users had no explicit basis in theories of health behavior change (Becker & Janz, 1987). There is little evidence to support the efficacy of HRAs in promoting behavior change (Kreuter & Strecher, 1996). In summary, THCs arose from the confluence of these innovations as a computer-based, theory-driven approach to customize health information to different individuals. Selecting appropriate tailoring constructs and variables Program objectives, relevant literature, a theoretical foundation, and audience research should determine appropriate tailoring variables, just as they should be the basis for any evidence-based health communication. Although tailored interventions now have been developed for a wide range of health problems, including injury prevention and immunization, most early tailoring studies addressed outcomes related to health promotion, prevention, and early detection of cancer (Kreuter et al., 1999). The topics included cancer prevention behaviors (e.g., diet, smoking, and exercise), cancer screening (e.g., getting a mammogram or colorectal cancer screening [CRCS]), and other cancer-related topics such as informed decision making about use of prostate and breast and CRCS (e.g., Pignone, Harris, & Kinsinger, 2000; Rimer et al., 2002). Researchers often have selected promising variables or constructs from different health behavior models, using integrated approaches to THC construction. This approach to tailoring has been labeled behavioral construct tailoring (Kreuter et al., 2004). Behavioral construct tailoring also is consistent with intervention planning models widely used in public health and health education that identify and address important and changeable determinants of a given behavior regardless of their theoretical origins (e.g., PRECEDE–PROCEED [Green & Kreuter, 2005]). Kreuter et al. (2005) conducted a randomized study of 1,227 African American women from public health centers in St. Louis to compare the effects of cancer prevention magazines tailored on behavioral constructs only (e.g., self-efficacy, perceived barriers, knowledge, motivational readiness), cultural constructs only (religiosity, collectivism, racial pride, and time orientation), or both behavioral and cultural constructs. Reactions to the magazines (e.g., attention, liking, yielding, perceived relevance, sharing magazines with others) were highly favorable among women in all groups, but rates of mammography use and increases in fruit and vegetable consumption at follow-up were lowest among women whose magazines were tailored only on cultural variables with no tailoring based on variables from theories of health behavior change (Kreuter et al., 2005). Tailoring as a strategy to enhance motivation to process health information THCs have been used to enhance the relevance and salience of information. These are important steps along a hypothesized pathway to behavior change, whereby greater perceived relevance and salience increase motivation to process information and enhance message receptivity, information processing, and behavior change (see Figure 1). Figure 1 Open in new tabDownload slide Hypothetical use of tailored health communications to affect behavior change pathway. Figure 1 Open in new tabDownload slide Hypothetical use of tailored health communications to affect behavior change pathway. Tailoring could enhance motivation to process health information in at least four ways: (a) match content to an individual’s information needs and interests, (b) frame health information in a context that is meaningful to the person, (c) use design and production elements to capture the individual’s attention, and (d) provide information in the amount, type, and through channels of delivery preferred by the individual, thus potentially reducing barriers to exposure of individuals to communication interventions. Such an approach then could increase attention, lead to subsequent yielding, and, ultimately, enhance the likelihood of behavior change. To date, most THC interventions have focused primarily on the first of these approaches, although there are examples of other levels of tailoring. By providing content of interest or concern to specific individuals, readers should be motivated to pay closer attention to the information, process it more carefully, and be more likely to use it to make decisions and take actions to improve health (Kreuter et al., 1999). Most but not all THC studies show that THCs indeed lead to these expected outcomes (e.g., Brug et al., 1999; Campbell et al., 1994; Kreuter & Strecher, 1996; Prochaska et al., 1993; Rimer et al., 2002; Skinner et al., 1994). Still needed are studies that explicate the pathway that leads from exposure variables to behavior change. Newer methods of analysis, like structural equation modeling, may aid in identifying the communication and behavioral impacts of THCs. Moreover, laboratory studies could be used to advantage in this regard. This explanation of tailoring effects—that behavior change occurs through increasing motivation to process information—is consistent with Petty and Cacioppo’s Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1981). ELM asserts that under certain conditions like elevated motivation and ability, people are active information processors—considering messages carefully, relating them to other information they have encountered, and comparing them to their own past experiences (Cacioppo, Harkins, & Petty, 1981; Petty, 2006; Petty, Cacioppo, Strathman, & Priester, 1994). One condition under which people are motivated to process information actively is when they perceive the information to be personally relevant. Fazio’s (Fazio & Towles-Schwen, 1999; Fazio, Powell, & Williams, 1989) MODE model, an integrative dual-process framework of the attitude–behavior relationship, expands upon ideas introduced in ELM (Petty, Haugtvedt, & Smith, 1995) and is complementary to theories of behavior change, such as the Theory of Reasoned Action and the Theory of Planned Behavior (Montano & Kasprzyk, 2002). MODE is an acronym for m otivation and o pportunity as de terminants of the process through which attitudes will affect behavior. In short, the model proposes that when people are motivated and have the opportunity to do so, they will engage in a more thoughtful, deliberative process of making decisions about their behavior. When motivation and/or opportunity are lacking, people’s actions will be guided by whatever existing global attitudes can be spontaneously retrieved from memory. Attitudes that are based on thoughtful deliberation (and/or personal experience or repeated expression) are more accessible (Petty et al., 1995). Thus, if tailored messages enhance motivation and opportunity to process health information and deliver compelling new ideas (or reinforce existing ones) that are favorable toward a given health behavior, the resulting attitudes may be more accessible to that person and, in turn, have greater influence on behavior. Evidence for this explanation of tailoring effects was provided in a randomized experiment in which participants received either tailored materials, a standard American Heart Association (AHA) brochure, or the AHA content formatted to look like the tailored materials. Analyses showed significantly greater cognitive activity among those receiving the tailored materials. Specifically, they reported significantly more positive thoughts about the materials, made more positive personal connections to the materials, had more positive self-assessment thoughts, and had more positive thoughts indicating behavioral intentions than those who received AHA or AHA-formatted materials (Kreuter, Bull, Clark, & Oswald, 1999). Across all groups, greater processing was significantly associated with intentions to try behavioral recommendations made in the tailored materials. Among those who received tailored messages, the relationship between processing and behavioral intentions was consistent with a mediation explanation. However, formal mediation analyses were not conducted and would be needed to understand the processes underlying tailoring effects. Using THC to influence the behavior change pathway: Hypothetical examples Message development THCs could be used in new ways to enhance knowledge about how to construct messages that lead to behavior change by influencing critical steps on the pathway to behavior change (see Figure 2). Some stage models assume that different messages are needed for people in different stages of readiness (e.g., precontemplation vs. maintenance in the SOC formulation). Researchers using this paradigm have created variations on messages rather than wholly different strategies for communication. Yet, one could use different approaches to message design for different types of people in different behavioral readiness states. Figure 2 Open in new tabDownload slide Using tailored health communications along the behavioral pathway to colorectal cancer screening. Figure 2 Open in new tabDownload slide Using tailored health communications along the behavioral pathway to colorectal cancer screening. Weinstein’s Precaution Adoption Process Model (Weinstein & Sandman, 2002) proposes that there are distinct phases to taking precautions about one’s health. Individuals must first be aware of a problem, engaged in thinking about it, deciding whether or not to take action, taking action, and finally maintaining any change that was made. Communication objectives would be different for individuals in each phase, as are the theories or change mechanisms that would inform communication development for individuals in each phase. If a person is unaware of CRCS, it may be important to capture his or her attention, perhaps using a compelling narrative or image to encourage the person to think about CRCS. If he or she is aware of CRCS but not particularly engaged in the topic, communication strategies would aim to get the person to think about it and apply it to his or her situation. The person might be given data about his or her risk and then encouraged to think about how a diagnosis of colorectal cancer could affect his or her live. This may be particularly effective for people with high family risk. If he or she was thinking about CRCS but had not yet decided whether to be screened, theories of persuasion and attitude should apply. A loss-framed message could highlight the consequences of not being screened, using individualized information about the person to make the case more salient and compelling and to facilitate behavior change (Salovey, Schneider, & Bailey, 1999). Tailored lists of an individual’s reasons for getting screened (pros) and reasons for not getting screened (cons) might help them evaluate the issues and move toward action. Such approaches are consistent with the Transtheoretical Model of Change (Prochaska et al., 2002) in helping people change the balance of pros and cons and also with ELM in encouraging more effortful processing (Petty et al., 1995). Tailored telephone calls have been used to help people make informed decisions about whether to be screened, offering on-the-spot scheduling to overcome logistic barriers (N. A. Davis, Lewis, Rimer, Harvey, & Koplan, 1997). Finally, reminders are an important behavioral strategy to encourage maintenance of behaviors, like cancer screening (Task Force on Community Preventive Services, 2005). This potential to integrate health behavior and communication models to design more effective health communication strategies offers a great opportunity for collaboration and leadership among multiple disciplines in research, application, and theory development. In the section below, we propose additional ways that THCs could be used to enhance understanding of the pathway to behavior change and thereby enhance understanding of message effects. Fishbein and colleagues (Institute of Medicine, 2003) argued that although behavioral prediction theories can help to identify specific attitudes and beliefs that may be impediments to behavior change, they do not tell us how to formulate effective messages to change these attitudes. More refined theories are needed that disaggregate key steps in the behavior change process, for example, reception, acceptance, yielding, and impact, and then permit creation of interventions to address these points. THCs could be constructed to deliver specific tailored messages to address each point in the continuum, similar to the way messages have been created to reflect different SOC (Figure 1). As described by McGuire (2001), certain elements of a communication might be designed primarily to capture attention, whereas other elements would influence important attitudes that are determinants of the behavioral outcome. Applied to tailoring, this would require specification of what kinds of individual data are needed to generate each component and how those data would be translated into specific messages. To date, developing and testing messages in this manner has not been an explicit strategy for most THC programs, although it is consistent with some behavior change theories used for tailoring, for example, SOC (Prochaska, 2004; Prochaska et al., 2002) and Precaution Adoption Model (Weinstein & Sandman, 2002). Such an approach to message development would also be congruent with theories, such as Goal Directed Behavior (Perugini & Bagozzi, 2001) that focus on actions required to translate goals into behaviors. Specifying the behavior pathway could help researchers to identify leverage points for interventions. Researchers also could test different kinds of messages for different kinds of people at different points along the behavior pathway as defined by different theories. Such messages would reflect understanding of the literature and theories appropriate for each step. For example, research on messages to influence yielding would benefit from literature and theories related to attitude change, persuasion, and attention. Other approaches might be evaluated for their effectiveness in increasing attention to tailored communications. Interventions could be tailored on reading level or learning style. Graphics could be matched to the recipients’ age, race, sex, or other attributes (Skinner et al., 2002). Some tailored programs have included pictures of the intended recipient or family members (Kreuter, Caburnay, Chen, & Donlin, 2004). Even overt claims of relevance (e.g., This information is designed just for you based on the answers you provided) could be manipulated experimentally, an approach that has been shown to increase perceived relevance among recipients (Burnkrant & Unnava, 1989). Skinner et al. (2002) varied graphics and messages about testing for BRCA 1 and 2 depending on whether women said they wanted a lot or a little information; art work changed depending on whether women were Black, White, or Jewish, and there were other kinds of variations as well. It was not possible to test whether such fine variations resulted in more behavior change. Unfortunately, even when innovative approaches such as these are used to create THCs, most research is not powered to assess the impact of particular message design approaches. Small laboratory-based studies might enable researchers to answer questions without requiring large sample sizes. Message source could be varied to determine whether more credible or more trusted sources exert more impact. In a multicomponent intervention, Campbell et al. (1999) provided tailored church bulletins to promote fruit and vegetable consumption to African American churchgoers. Based on random assignment, some church bulletins used an expert scientific approach to addressing nutrition, whereas others used a spiritually based approach with a nutrition message from the church’s pastor. Message trust was significantly higher among those who received the spiritually based bulletin. Kalichman and Coley (1995) randomly assigned 100 Black women in an urban health clinic in Milwaukee to view one of three videos on HIV testing. The first had a Black man as the narrator, the second was identical but used a Black woman as narrator, and the third had the Black woman deliver the same content but also stressed culturally relevant losses as consequences of not being tested (e.g., “…not getting tested puts your family at risk of losing you to the disease” [p. 249]). Participants rated characters in this last video as significantly more concerned about Black families and the Black community, women like me, and as being like people I know compared to characters in the other videos. These findings suggest that it may be productive to tailor the source of health information to individual characteristics, including not only demographics but also culturally relevant values and beliefs. The communication channel through which a tailored intervention is delivered could be customized based on individuals’ access to health care, their health care needs, or personal preferences. Whether doing so enhances the effects of tailoring is an empirical question. Nested substudies could be used to assess the impact of specific message strategies or use adaptive designs to permit testing discrete hypotheses about particular message strategies. Using ELM as an example, tailored messages also could be developed to increase time spent thinking about a topic, with the assumption that more effortful thinking is more likely to lead to behavior change. Messages could be varied for people who are very resistant to effortful processing. Studies using a similar design have already suggested that message effectiveness may be enhanced if tailored to a person’s health locus of control (Holt, Clark, Kreuter, & Scharff, 2000). Tailored messages may be especially useful when emotional arousal facilitates behavior change. For example, Kruglanski (2002), Lang (2002), and others have shown that some people are more susceptible to messages that heighten emotional arousal. Responses to persuasive stimuli may vary by other personality characteristics, such as trait anxiety and self-monitoring (DeBono & McDermott, 1994; DeBono & Snyder, 1989). Williams-Piehota, Pizarr, Schneider, Mowad, and Salovey (2005) found that a minimal intervention that involved brief counseling and a tailored brochure, matched/mismatched to individuals’ monitor blunting styles, increased self-reported mammography use 6 months after the intervention for blunters but not monitors. There were corresponding changes in some but not all the other predictor variables. If personality traits can reliably predict responses to different types of persuasive messages, it may be possible to measure these variables and use an individual’s responses to customize how health messages tailored on other variables are delivered. On a population basis, it might not be feasible to collect detailed information on variables such as monitor–blunter status. But in a health plan or other more constrained setting, where patients are expected to provide certain kinds of personal information, it might be practicable to include personality scales. The wide range of possible tailoring variables and strategies for customizing content and delivery of health information raises questions about how much tailoring is required to achieve different communication objectives, and whether the incremental cost of more tailoring, or even more units of intervention, is worth the investment. Identifying the optimum dose of tailoring required to induce behavior change or even antecedent changes along the behavior change pathway presents significant challenges. First, there are at least two dimensions of dose to be considered: extent of tailoring (i.e., highly tailored vs. minimally tailored) and amount of exposure (e.g., in minutes or pages, number of tailored materials received, over what period of time). Second, there currently are no universally accepted models of tailoring. This includes what is defined as tailoring and whether subsequent communications use the initial data or whether new data are the expected norm for subsequent THCs. There is some research on this topic, but the results are not yet published. Without some agreement on how best to address different outcomes, the comparability of studies will remain limited. Different dose models for the same outcome could emerge simply because the studies started with different assumptions about the key determinants of change. Moreover, there are important moderating variables like message and production quality that can be difficult to assess objectively and are seldom reported in published research. For example, if message quality was low, a greater dose of tailoring might diminish effects. Few studies of tailoring dose have been conducted to date, and their results have been mixed. Dijkstra (2005) recently demonstrated that smoking cessation materials that were minimally tailored using mostly personal identification and demographic information led to greater self-reported quitting among college students than did more elaborately tailored materials. Bull, Kreuter, and Scharff (1999) found no effects of minimally tailored materials on physical activity in a community sample of adults. Moreover, there has been little attention to evaluating the quality of messages or of the overall communication product. The cost-effectiveness of tailoring (vs. not) and of different doses of tailoring is not known. Only a few authors have provided any data on cost-effectiveness (Lipkus, Rimer, Halabi, & Strigo, 2000; Saywell, et al., 2003). Generally speaking, we would expect that THC programs and materials are more expensive to develop than nontailored ones, with delivery costs between the two likely more comparable. Adding dose or complexity to a tailored program usually involves more assessments, data processing, computer programming, and message development, thus increasing costs. With additional assessment points, problems of participant fatigue and attrition become more important. However, cost-effectiveness also would be influenced by quantity of units produced, and tailoring efficiency is likely to increase with the number of units produced (Glasgow, Goldstein, Ockene, & Pronk, 2004). Tailored interventions are highly scalable and may be most appropriate when there is the potential to reach large numbers of people. Moreover, as Abrams, Leslie, Mermelstein, Kobus, and Clayton (2003) have argued, if tailoring permits greater population reach by getting the attention of intended recipients in a way that generic materials are unlikely to achieve, THCs could be more cost effective. Quantifying these costs and determining whether they are justified by superior effectiveness should be a priority for health communication researchers. Understanding message effects Most THC studies have been designed to assess behavioral impact. These studies provide less information about message effects than they do about overall program effects. However, an important strength of the THC literature is use of process measures, such as those developed by Brug et al. (1999). These intermediate outcomes can provide insights about the behavior change pathway. For example, Rimer et al. (2002) found that women who received a combined intervention of tailored print booklet (TP) and tailored telephone counseling (TC) were more likely to have had mammograms 12 and 24 months later than women in the usual care group or those who received only tailored print materials. This raised the question of why the combination of TP + TC was more effective than the TP alone. The process items may have provided some insight into the answer. Women in the TP + TC group were more likely to say that the newsletter helped them “think about the limits of mammography.” Those who responded that the newsletter helped in this way were more likely to have had mammograms. Women who received TP alone also were less likely than those who received TP + TC to have read most or all of the material, kept it, believed that the newsletter contained information especially for them, or said it helped them think about benefits and overcome barriers. (Findings were similar at 12 and 24 months.) Although reported reactions to the materials from women in both groups were extremely positive—more than 85% in the highest affirmative category on each item—the experience of the booklet seems to have been different for the two intervention groups and was significantly more favorable for the TP + TC group. These data are not definitive in explaining how the intervention exerted its effect. However, process items can be useful in understanding how women responded to different interventions, but it is not sufficient. Future directions The first generation of research on THC focused appropriately on the basic question of whether tailored materials were more effective than nontailored materials. It is now time to progress to a next generation of tailoring studies to explore how and under what conditions tailoring works and how its effects can be optimized. Integration of a persuasion and message effects perspective will help in both types of inquiry. More attention should be paid to issues, such as the appropriate control group for THC studies, and to creating standards and metrics so that THC studies can be compared at the message level. When developing any health communication, tailored or not, certain operational tasks must be completed (Kreuter & Wray, 2003). At the simplest level, these tasks could include choosing credible sources, developing a message strategy, defining the appropriate sources of data, and determining the settings and/or channels for optimal communication delivery. Although each of these decisions presents an opportunity to tailor health communications, it should not be assumed that THCs are always the ideal or even necessary approach to intervention. To date, tailoring has been used primarily as a message strategy, providing specific content to individuals based upon information from or about them, typically responding to a theory-driven assessment of behavioral precursors. There is considerable need and opportunity to explore a much wider range of tailoring strategies, formats, effects, and mechanisms for effectiveness. Collaborations between behavioral and communication scientists may be important in refining understanding of message effects within tailoring paradigms. Specifically, this paper identifies multiple intermediary points on the behavioral pathway and suggests that different tailoring variables and strategies and even different message formats might be better suited for different points along the way. This assumption should be empirically tested, as should questions about the extent to which and mechanisms through which different intermediary variables contribute to the end goal of behavior change, and whether some message strategies and individual data are more useful than others in influencing different points along the behavioral pathway. Moreover, to date, most printed health THC strategies have used researcher-defined data as the basis for developing THCs. Using a combination of quantitative and qualitative data, we generally decide what interventions people should receive. Computerized online interventions offer more opportunities for collaborative, real-time design. Products could range from those that are expert designed to those that are user designed, with combinations in between. Although some interventions have offered people choices about the amount of information they could receive on various topics (e.g., Skinner et al., 2002) and similar options, we are aware of no research that has assessed the efficacy of user-designed versus expert-designed THCs. Finally, little is known about the conditions under which tailoring may be most effective. Kreuter and Wray (2003) concluded that tailoring is not the best or most practical communication strategy for all situations, but should be considered when (a) there is a high level of variability in the population of interest on key determinants of a given outcome, (b) there is general awareness and understanding of the outcome in the population of interest, and (c) there is some mechanism for gathering data from or about the population of interest. These assumptions have not been tested, however. Methodological advances, such as fractional designs, sometimes referred to as adaptive designs, may permit more efficient testing of tailoring approaches without requiring many separate studies (Collins, Murphy, & Bierman, 2004). As with any communication-based intervention for health behavior change, there are limits to what can be achieved with tailoring. Information alone, even tailored information, cannot change many important determinants of health and health-related behaviors. Limited access to health care and other structural barriers outside an individual’s personal control may be the most important intervention points for some problems. We might expect tailoring to be less effective when key determinants of change related to either the behavior or the population of interest are not amenable to change by information. Indirect evidence from some tailoring studies supports this point. For example, in a recent study of mammography use among lower income African American women aged 40 and older, tailoring was most effective among women who had been screened previously but were currently behind schedule, but no more effective than usual care among women who never had mammograms (Kreuter et al., 2005). Perhaps, the women who had never been screened faced a different set of barriers to mammography than those who previously had a mammogram. There are many critical questions to be answered about THCs. We look toward a productive next generation of research to provide answers. Advances in computer hardware and software and health informatics, including greater use of electronic medical records and personal health records, will present new opportunities to design communication to meet the information and communication needs and preferences of recipients, using accurate, appropriate data. Acknowledgments This research was supported in part by the National Cancer Institute’s (NCI) Center of Excellence in Cancer Communication Research program (CA-095815-02) to M.W.K. and National Institutes of Health (NIH) Grant 1 R01 CA105786-01 and NCI Grant 5-P30-CA16086-28 to B.K.R. The authors thank Linda Kastleman at the University of North Carolina and Chris Casey, Danielle Davis, and Keri Jupka at Saint Louis University for assistance in the preparation of the manuscript. References Abrams , D. B. , Leslie , F., Mermelstein , R., Kobus , K., & Clayton , R. R. ( 2003 ). Transdisciplinary tobacco use research . Nicotine & Tobacco Research , 5 ( Suppl. 1 ), S5 – S10 . Google Scholar Crossref Search ADS WorldCat Becker , M. , & Janz , N. ( 1987 ). On the effectiveness and utility of health hazard/health risk appraisal in clinical and non-clinical settings: Behavioral science perspectives on health hazard/health risk appraisal . Health Services Research , 22 , 537 – 551 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Brug , J. , Campbell , M., & Van Assema , P. ( 1999 ). The application and impact of computer-generated personalized nutrition education: A review of the literature . Patient Education and Counseling , 36 ( 2 ), 145 – 156 . Google Scholar Crossref Search ADS PubMed WorldCat Bull , F. , Kreuter , M., & Scharff , D. ( 1999 ). Effects of tailored, personalized, and general materials on physical activity . Patient Education and Counseling , 36 ( 2 ), 181 – 192 . Google Scholar Crossref Search ADS PubMed WorldCat Burnkrant , R. E. , & Unnava , H. R. ( 1989 ). Self-referencing: A strategy for increasing processing of message content . Personality and Social Psychology Bulletin , 15 , 628 – 638 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. , Harkins , S., & Petty , R. ( 1981 ). The nature of attitudes and cognitive responses and their relationships to behavior. In R. Petty, T. Ostrom, & T. Brock (Eds.), Cognitive responses in persuasion (pp. 31 – 54 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Campbell , M. K. , Bernhardt , J., Waldmiller , M., Jackson , B., Potenziani , D., Weathers , B., et al. . ( 1999 ). Varying the message source in computer tailored nutrition education . Patient Education Counseling , 36 ( 2 ), 159 – 167 . Google Scholar Crossref Search ADS WorldCat Campbell , M. K. , DeVellis , B. M., Strecher , V. J., Ammerman , A. S., DeVellis , R. F., & Sandler , R. S. ( 1994 ). Improving dietary behavior: The effectiveness of tailored messages in primary care settings . American Journal of Public Health , 84 , 783 – 787 . Google Scholar Crossref Search ADS PubMed WorldCat Collins , L. M. , Murphy , S. A., & Bierman , K. A. ( 2004 ). A conceptual framework for adaptive interventions . Prevention Science , 5 ( 3 ), 181 – 192 . Google Scholar Crossref Search ADS WorldCat Davis , S. W. , Cummings , K. M., Rimer , B. K., Sciandra , R., & Stone , J. C. ( 1992 ). The impact of tailored self-help smoking cessation guides on young mothers . Health Education Quarterly , 19 , 495 – 504 . Google Scholar Crossref Search ADS PubMed WorldCat Davis , N. A. , Lewis , M. J., Rimer , B. K., Harvey , C. M., & Koplan , J. P. ( 1997 ). Evaluation of a phone intervention to promote mammography in a managed care plan . American Journal of Health Promotion , 11 , 247 – 249 . Google Scholar Crossref Search ADS PubMed WorldCat DeBono , K. G. , & McDermott , J. B. ( 1994 ). Trait anxiety and persuasion: Individual differences in information processing strategies . Journal of Research in Personality , 28 , 395 – 407 . Google Scholar Crossref Search ADS WorldCat DeBono , K. G. , & Snyder , M. ( 1989 ). Understanding consumer decision-making processes: The role of form and function in product evaluation . Journal of Applied Social Psychology , 19 , 416 – 424 . Google Scholar Crossref Search ADS WorldCat Dijkstra , A . ( 2005 ). Working mechanisms of computer tailored health education: Evidence from smoking cessation . Health Education Research , 20 , 527 – 539 . Google Scholar Crossref Search ADS PubMed WorldCat Fazio , R. H. , Powell , M. C., & Williams , C. J. ( 1989 ). The role of attitude accessibility in the attitude-to-behavior process . Journal of Consumer Research , 16 , 280 – 288 . Google Scholar Crossref Search ADS WorldCat Fazio , R. H. , & Towles-Schwen , T. ( 1999 ). The MODE model of attitude-behavior processes. In S. Chaiken & Y. Trope (Eds.), Dual-process models in social psychology (pp. 97 – 116 ). New York : Guilford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Glasgow , R. E. , Goldstein , M. G., Ockene , J. K., & Pronk , N. P. ( 2004 ). Translating what we have learned into practice: Principles and hypotheses for addressing multiple behaviors in primary care . American Journal of Preventive Medicine , 27 ( 2 ), 88 – 101 . Google Scholar Crossref Search ADS PubMed WorldCat Green , L. , & Kreuter , M. ( 2005 ). Health program planning . Mountain View, CA : Mayfield . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Holt , C. , Clark , E., Kreuter , M., & Scharff , D. ( 2000 ). Does locus of control moderate the effects of tailored health education materials? Health Education Research , 15 , 393 – 403 . Google Scholar Crossref Search ADS PubMed WorldCat Institute of Medicine . ( 2003 ). Speaking of health: Assessing health communications strategies for diverse populations . Washington, DC : National Academies Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Janz , N. , Champion , V., & Strecher , V. ( 2002 ). The health behavior model. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research, and practice (pp. 99 – 120 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kalichman , S. , & Coley , B. ( 1995 ). Context framing to enhance HIV-antibody-testing messages targeted to African American women . Health Psychology , 14 , 247 – 254 . Google Scholar Crossref Search ADS PubMed WorldCat Keintz , M. K. , Fleisher , L., & Rimer , B. K. ( 1994 ). Reaching mothers of preschool-aged children with a targeted quit smoking intervention . Journal of Community Health , 19 ( 1 ), 25 – 40 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Bull , F. C., Clark , E. M., & Oswald , D. L. ( 1999 ). Understanding how people process health information: A comparison of tailored and untailored weight loss materials . Health Psychology , 18 , 487 – 494 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Caburnay , C. A., Chen , J., & Donlin , M. ( 2004 ). Effectiveness of individually tailored calendars in promoting childhood immunization in urban public health centers . American Journal of Public Health , 94 ( 1 ), 122 – 127 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Farrell , D., Olevitch , L., & Brennan , L. ( 1999 ). Tailoring health messages: Customizing communication with computer technology . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kreuter , M. W. , & Skinner , C. ( 2000 ). Tailoring: What’s in a name? Health Education Research , 15 ( 1 ), 1 – 4 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Skinner , C. S., Holt , C. L., Clark , E. M., Haire-Joshu , D., Fu , Q., et al. . ( 2005 ). Cultural tailoring for mammography and fruit and vegetable intake among low-income African American women in urban public health centers . Preventive Medicine , 41 ( 3 ), 53 – 62 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Skinner , C. S., Steger-May , K., Holt , C. L., Bucholtz , D. C., Clark , E. M., et al. . ( 2004 ). Responses to behaviorally vs. culturally tailored cancer communication for African American women . American Journal of Health Behavior , 28 , 195 – 207 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , & Strecher , V. ( 1996 ). Do tailored behavior change messages enhance the effectiveness of health risk appraisals? Results from a randomized trial . Health Education Research , 11 ( 1 ), 97 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , Strecher , V., & Glassman , B. ( 1999 ). One size does not fit all: The case for tailoring print materials . Annals of Behavioral Medicine , 21 , 276 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Kreuter , M. W. , & Wray , R. J. ( 2003 ). Tailored and targeted health communication: Strategies for enhancing information relevance . American Journal of Health Behavior , 27 ( Suppl .), S227 – S232 . Google Scholar Crossref Search ADS PubMed WorldCat Kruglanski , A . ( 2002 , May). Appreciation and override processes in attitude change . Paper presented at the meeting of the National Cancer Institute’s Behavioral Research Program on Integrating Theories of Message Effects and Health Behavior Change to Improve Cancer Control , Philadelphia . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A . ( 2002, May ). Using the limited capacity model of mediated message processing to study and design health communication messages . Paper presented at the meeting of the National Cancer Institute’s Behavioral Research Program on Integrating Theories of Message Effects and Health Behavior Change to Improve Cancer Control , Philadelphia . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lipkus , I. M. , Rimer , B. K., Halabi , S., & Strigo , T. S. ( 2000 ). Can tailored interventions increase mammography use among HMO women? American Journal of Preventive Medicine , 18 ( 1 ), 1 – 10 , Google Scholar Crossref Search ADS PubMed WorldCat Maibach , E. , Rothschild , M., & Novelli , W. ( 2002 ). Social marketing. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research, and practice (pp. 437 – 462 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC McGuire , W. J . ( 2001 ). Input and output variables currently promising for constructing persuasive communications. In R. E. Rice & C. K. Atkin (Eds.), Public communication campaigns (pp. 22 – 48 ). Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Montano , D. , & Kasprzyk , D. ( 2002 ). The theory of reasoned action and the theory of planned behavior. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research, and practice (pp. 67 – 99 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pepper , D . ( 2000 ). Is your company ready for one-to-one marketing? Cambridge, MA : Harvard Business School Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Perugini , M. , & Bagozzi , R. P. ( 2001 ). The role of desires and anticipated emotions in goal-directed behaviours: Broadening and deepening the theory of planned behaviour . British Journal of Social Psychology , 40 ( 1 ), 79 – 98 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R . ( 2006 ). Fundamental processes of attitude change . Journal of Communication , 56 , S81 – S104 . Google Scholar Crossref Search ADS WorldCat Petty , R. , & Cacioppo , J. ( 1981 ). Attitudes and persuasion: Classic and contemporary approaches . Dubuque, IA : WC Brown . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. , Cacioppo , J., Strathman , A., & Priester , J. R. ( 1994 ). To think or not to think: Exploring two routes to persuasion. In S. Shavitt & T. Brock (Eds.), Persuasion: Psychological insights and perspectives (pp. 113 – 147 ). Boston : Allyn and Bacon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Haugtvedt , C. P., & Smith , S. M. ( 1995 ). Elaboration as a determinant of attitude strength: Creating attitudes that are persistent, resistant and predictive of behavior. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 93 – 130 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pignone , M. P. , Harris , R., & Kinsinger , L. ( 2000 ). Videotape-based patient decision aid for colon cancer screening—A randomized trial . Annals of Internal Medicine , 133 , 761 – 769 . Google Scholar Crossref Search ADS PubMed WorldCat Prochaska , J. O . ( 2004 ). Population treatment for addictions . Current Directions in Psychological Science , 13 , 242 – 247 . Google Scholar Crossref Search ADS WorldCat Prochaska , J. O. , DiClemente , C. C., Velicer , W. F., & Rossi , J. S. ( 1993 ). Standardized, individualized, interactive, and personalized self-help programs for smoking cessation . Health Psychology , 12 , 399 – 405 . Google Scholar Crossref Search ADS PubMed WorldCat Prochaska , J. O. , Redding , C. A., & Evers , K. E. ( 2002 ). The transtheoretical model and stages of change. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research, and practice (pp. 99 – 120 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rakowski , W. , Lipkus , I. M., Clark , M. A., Rimer , B. K., Ehrich , B., Lyna , P. R., et al. . ( 2003 ). Comparing a reminder letter, tailored stepped-care, and self-choice for repeat mammography . American Journal of Preventive Medicine , 25 , 308 – 314 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. , & Glassman , B. ( 1998 ). Tailoring communications for primary care settings . Methods of Information in Medicine , 37 ( 2 ), 171 – 177 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Rimer , B. K. , Halabi , S., Skinner , C. S., Kaplan , E. B., Crawford , Y., Samsa , G. P., et al. . ( 2001 ). The short-term impact of tailored mammography decision-making interventions . Patient Education and Counseling , 43 , 269 – 285 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. , Halabi , S., Skinner , C. S., Lipkus , I. M., Strigo , T. S., Kaplan , E. B., et al. . ( 2002 ) Effects of a mammography decision-making intervention at 12 and 24 months . American Journal of Preventive Medicine , 22 , 247 – 257 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. , Orleans , C. T., Fleischer , L., Cristinzio , S., Resch , N., Telepchak , J., et al. . ( 1994 ). Does tailoring matter? The impact of a tailored guide on ratings and short-term smoking-related outcomes for older smokers . Health Education Research, Theory and Practice , 9 ( 1 ), 69 – 84 . Google Scholar Crossref Search ADS WorldCat Robinson , R. G. , Orleans , C. T., James , D. A., & Sutton , C. D. ( 1992 ). Pathways to freedom: Winning the fight against tobacco . Philadelphia : Fox Chase Cancer Center . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rogers , E. M . ( 2003 ). Diffusion of innovations . New York : Free Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Salovey , P. , Schneider , T. R., & Bailey , C. ( 1999 ). Persuasion for the purpose of cancer risk reduction: A discussion . Journal of the National Cancer Institute Monographs , 25 , 119 – 122 . Google Scholar Crossref Search ADS WorldCat Samuelson , M . ( 1997 ). Changing unhealthy lifestyles: Who’s ready… who’s not?: An argument in support of the stages of change component of the transtheoretical model . American Journal of Health Promotion , 12 ( 1 ), 13 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat Saywell , R. M . Jr., Champion , V., Zollinger , T., Maraj , M., Skinner , C., Zoppi , K., et al. . ( 2003 ). The cost effectiveness of 5 interventions to increase mammography adherence in a managed care population . American Journal of Managed Care , 9 ( 1 ), 33 – 44 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Skinner , C. S. , Schildkraut , J. M., Berry , D., Calingaert , B., Marcom , B. K., Sugarman , J., et al. . ( 2002 ). Pre-counseling education materials for BRCA testing: Does tailoring make a difference? Genetic Testing , 6 ( 2 ), 93 – 105 . Google Scholar Crossref Search ADS PubMed WorldCat Skinner , C. S. , Strecher , V. J., & Hospers , H. ( 1994 ). Physicians’ recommendations for mammography: Do tailored messages make a difference? American Journal of Public Health , 84 ( 1 ), 43 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat Slater , M. , & Flora , J. ( 1991 ). Health lifestyles: Audience segmentation analysis for public health interventions . Health Education Quarterly , 18 , 221 – 233 . Google Scholar Crossref Search ADS PubMed WorldCat Strecher , V. J . ( 1999 ). Computer-tailored smoking cessation materials: A review and discussion . Patient Education and Counseling , 36 ( 2 ), 107 – 117 . Google Scholar Crossref Search ADS PubMed WorldCat Strecher , V. J. , Rimer , B. K., & Monaco , K. D. ( 1989 ). Development of a new self-help guide—Freedom from smoking for you and your family . Health Education Quarterly , 16 ( 1 ), 101 – 112 . Google Scholar Crossref Search ADS PubMed WorldCat Task Force on Community Preventive Services . ( 2005 ). Vaccine preventable diseases. In S. Zaza, P. Briss, & K. Harris (Eds.), The guide to community preventive services . New York : Oxford University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Wagner , E. H. , Beery , W. L., Schoenbach , V. J., & Graham , R. M. ( 1982 ). An assessment of health hazard/health risk appraisal . American Journal of Public Health , 72 , 347 – 352 . Google Scholar Crossref Search ADS PubMed WorldCat Weinstein , N. D. , & Sandman , P. M. ( 2002 ). The Precaution adoption process model. In K. Glanz, B. K. Rimer, & F. M. Lewis (Eds.), Health behavior and health education: Theory, research and practice (pp. 121 – 143 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Williams-Piehota , P. , Pizarr , J., Schneider , T. R., Mowad , L., & Salovey , P. ( 2005 ). Matching health messages to monitor-blunter coping styles to motivate screening mammography . Health Psychology , 24 ( 1 ), 58 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat © 2006 International Communication Association

Dillard, James, Price;Nabi, Robin, L.

doi: 10.1111/j.1460-2466.2006.00286.xpmid: N/A

Abstract This paper reviews and summarizes the literature on the relationship between emotion and persuasion as it bears on the production of cancer prevention and detection messages. A series of propositions are presented that serve to illustrate the intricacies of the emotion–persuasion relationship. These propositions deal with the necessary conditions for emotional arousal, individual differences in emotional reactivity to cancer messages, the potential for emotion-inducing messages to produce persuasive and counterpersuasive effects, the conditions that circumscribe the influence of emotions on persuasion, and the mechanisms by which that influence is achieved. To the extent that the literature permits, advice on message design is offered. Among its many responsibilities, the National Cancer Institute produces messages that encourage individuals to (a) behave in ways that lessen their risk of cancer and (b) recommend actions to detect cancer in its earliest stages. Given the inherently threatening nature of cancer, it may be impossible for such messages to achieve these goals without intentionally or unintentionally arousing their audiences’ emotions. Although emotions can overwhelm individuals, they may also motivate behaviors that are psychologically difficult to enact. The effective use of emotions as persuasive devices, however, requires an understanding of not only principles related to emotional arousal but also the processes that allow emotional arousal to be translated into an effective action. This article is intended to speak to these two overarching issues. Although there is scant literature linking persuasive effects of emotion and cancer-related issues, we illustrate our claims with cancer-related research whenever possible, including previously unpublished data on individuals’ reactions to an antitobacco public service announcement (PSA). However, we augment these data with hypothetical message scenarios to further illuminate our points, which we outline in the eight propositions that follow. Proposition 1: Cancer-related messages have the potential to arouse one or more emotions. Research supports the claim that messages intended to evoke a particular emotional state may arouse not only that emotion but others as well. Dillard, Plotnick, Godbold, Freimuth, and Edgar (1996) found that all but one of 31 AIDS prevention fear appeals evoked change in more than one emotional state. This is certainly not an isolated finding. For example, Pinto and Priest (1991) showed that guilt-based advertisements evoked anger as well as guilt; Nabi (2002a) found that social issue messages designed to evoke anger also aroused disgust; Bennett (1998) reported that messages intended to produce guilt also created feelings of shame. It is reasonable to assume that cancer prevention and detection messages are also likely to evoke multiple affects. The Victor Crawford PSA presents an ex-tobacco lobbyist, who is dying of cancer, admitting that he and the industry had lied to the public for years about the dangers of smoking (see Table 1). Research participants were shown the PSA, then asked to report their emotional reactions to it on a series of close-ended scales (0 =none of this emotion, 4 =a great deal of this emotion), their cognitive reactions to it using the thought-listing technique, as well as their perceptions of the message’s effectiveness, and their likelihood of smoking.1 After message exposure, the mean levels of six emotions were significantly different from 0 at p <.05 (even though happiness was only marginally present): surprise (0.71), fear (1.04), anger (1.10), sadness (1.40), happiness (0.03), and contentment (0.27). The implications of these results are twofold. First, as with any number of health threats, fear is likely to be the focus of cancer prevention or detection messages. However, other negative emotions, such as sadness, or even positive emotions, such as happiness, might prove to be the dominant affective outcome of such messages. Second, message creators should consider not only the emotions they intend to arouse but the unintentional ones as well as they may have implications for message effects (as will be discussed shortly). Table 1 Description of the Victor Crawford Public Service Announcement The words “The Truth” appear in white text on a black screen. The ad continues in black and white to show images of an old man, Victor Crawford. He begins by telling viewers that the tobacco company is targeting young children because “they don’t know better.”“They might get your sister or your brother.…” He then admits that he was a tobacco lobbyist for 20 years and knows how the tobacco industry works. Finally, he apologizes by saying, “I lied and I’m sorry.” The ad concludes with four individually presented black screens containing the following text: “Victor Crawford died. He died of lung cancer. Tobacco is addictive. Don’t smoke.” The words “The Truth” appear in white text on a black screen. The ad continues in black and white to show images of an old man, Victor Crawford. He begins by telling viewers that the tobacco company is targeting young children because “they don’t know better.”“They might get your sister or your brother.…” He then admits that he was a tobacco lobbyist for 20 years and knows how the tobacco industry works. Finally, he apologizes by saying, “I lied and I’m sorry.” The ad concludes with four individually presented black screens containing the following text: “Victor Crawford died. He died of lung cancer. Tobacco is addictive. Don’t smoke.” Open in new tab Table 1 Description of the Victor Crawford Public Service Announcement The words “The Truth” appear in white text on a black screen. The ad continues in black and white to show images of an old man, Victor Crawford. He begins by telling viewers that the tobacco company is targeting young children because “they don’t know better.”“They might get your sister or your brother.…” He then admits that he was a tobacco lobbyist for 20 years and knows how the tobacco industry works. Finally, he apologizes by saying, “I lied and I’m sorry.” The ad concludes with four individually presented black screens containing the following text: “Victor Crawford died. He died of lung cancer. Tobacco is addictive. Don’t smoke.” The words “The Truth” appear in white text on a black screen. The ad continues in black and white to show images of an old man, Victor Crawford. He begins by telling viewers that the tobacco company is targeting young children because “they don’t know better.”“They might get your sister or your brother.…” He then admits that he was a tobacco lobbyist for 20 years and knows how the tobacco industry works. Finally, he apologizes by saying, “I lied and I’m sorry.” The ad concludes with four individually presented black screens containing the following text: “Victor Crawford died. He died of lung cancer. Tobacco is addictive. Don’t smoke.” Open in new tab Proposition 2: The type and intensity of emotional reactions to cancer-related messages vary across individuals based on their cognitive appraisals. Individuals’ cognitive interpretations of message scenarios underlie not only the type of emotion experienced but also the degree to which it is felt. Research on appraisal theory (Scherer, Schorr, & Johnstone, 2001), a vein of inquiry that has been mined by both authors, supports this claim. In its most basic form, appraisal theory asserts that emotions arise from assessing the implications of events and situations relative to ones’ goals. These goals range from the mundane and immediate (e.g., buying a newspaper) to the abstract and distal (e.g., living a principled life). Regardless of the type of goal, when an individual judges the environment to be incongruent with his or her goals, negative emotions will result. Conversely, the perception of goal–environment compatibility yields positive emotional states. Appraisal theory generally, and Nabi’s (1999) cognitive-functional model (CFM) specifically, provide an understanding of emotions and their causes that take us well beyond the simple positive–negative distinction. Table 2 makes this point. The first column lists the emotions measured in the Victor Crawford study. The second column inventories the corresponding molar appraisals or, equivalently, core relational themes (Lazarus, 1991; Smith & Lazarus, 1993). For example, fear arises from the perception of danger. These molar appraisals are thought to summarize specific configurations of the more numerous molecular appraisals, which appear in the third column. We see that fear is predicated on the belief that the individual faces impending danger over which he or she may have little or no control. Table 2 Emotions and Their Cognitive Antecedents Emotions . Appraisals . Molar . Molecular . Surprise Novelty Sudden, unfamiliar Fear Danger High probability of serious harm Anger Offense Unwarranted obstruction of goal Sadness Loss Irrevocable failure to meet goal Happiness Progress Acute movement toward goal Contentment Satisfaction Aspirations have been met Emotions . Appraisals . Molar . Molecular . Surprise Novelty Sudden, unfamiliar Fear Danger High probability of serious harm Anger Offense Unwarranted obstruction of goal Sadness Loss Irrevocable failure to meet goal Happiness Progress Acute movement toward goal Contentment Satisfaction Aspirations have been met Open in new tab Table 2 Emotions and Their Cognitive Antecedents Emotions . Appraisals . Molar . Molecular . Surprise Novelty Sudden, unfamiliar Fear Danger High probability of serious harm Anger Offense Unwarranted obstruction of goal Sadness Loss Irrevocable failure to meet goal Happiness Progress Acute movement toward goal Contentment Satisfaction Aspirations have been met Emotions . Appraisals . Molar . Molecular . Surprise Novelty Sudden, unfamiliar Fear Danger High probability of serious harm Anger Offense Unwarranted obstruction of goal Sadness Loss Irrevocable failure to meet goal Happiness Progress Acute movement toward goal Contentment Satisfaction Aspirations have been met Open in new tab Table 2 provides a theoretical framework that can make understandable the range of emotional reactions people might have to cancer-related messages, such as the Victor Crawford PSA. Surprise likely resulted from seeing someone who devoted his life to lobbying for the tobacco companies admit to having been deceptive. In addition to the fright of viewing a man near death, the message stimulated fear by referencing powerful organizations that may harm loved ones. As intended, the PSA aroused anger by putting a face on the long-term mendacity of the tobacco companies. Sadness was also evoked by what appears to be a wasted life. Finally, the mean for contentment may reflect the satisfaction one might feel in seeing a man who may have caused harm to others suffering an unfortunate fate. We could make a similar argument about a hypothetical PSA promoting cancer detection behavior, like mammography screening. During exposure to such a message, a woman might be afraid if she is made to think about being diagnosed with a life-threatening disease, angry at being reminded of something so unpleasant to consider, sad at the thought of losing a breast to cancer, or even hopeful at the thought of taking action to defend her health status. Indeed, in their research on messages designed to motivate mammography screening, Williams-Piehota, Pizarro, Schneider, Mowad, and Salovey (2005) measured not only fear but also hope and reassurance, suggesting that cancer control messages might be appraised such that both positive and negative affective states may be evoked. In sum, the appraisals relevant to a range of emotions may be discernible in the content of cancer control PSAs. But it is the audience interpretation of the content that will influence which emotions are experienced. The stronger those perceptions, that is, the more connected one perceives the PSA content to one’s goals, the more intense the emotional experience is likely to be. Simply noting such variation in emotional response, however, does little to advance theory or practice. Rather, it is necessary to point to explanatory variables that magnify or minimize the intensity of emotional response. There are likely a range of individual differences that might impact the appraisals likely to dominate message interpretation, but unfortunately, such research is only in its nascent stages with minimal evidence on which we can draw. For this reason, we reference just one trait-based and one context-based difference—coping style and prior knowledge—that may serve as representative examples of the types of variables future research should investigate for their potential impact on appraisal processes. Coping styles In brief, monitoring and blunting coping styles are believed to be stable individual differences that reflect the degree to which people seek out or avoid information when confronted with threatening situations (e.g., Miller, 1987). In a health context, these coping styles are expected to influence people’s cognitive and emotional reactions to health threats that, in turn, may impact the amount of information desired and the impact that information might ultimately have on health-related behaviors (see Miller, 1991, 1995; Miller, Shoda, & Hurley, 1996). Despite substantial research in clinical settings, there is little attention to how coping style might interact with persuasive message design to impact emotional reaction and persuasive success. However, recently Williams-Piehota et al. (2005) offered evidence that messages matched to the needs of monitors versus blunters were more effective in promoting mammography screening, particularly for blunters. Of note, blunters exposed to messages designed to meet their information needs (i.e., simple and direct information) experienced the least amount of negative affect, whereas monitors exposed to messages designed for blunters experienced the most negative affect. In this case, negative affect appeared to serve as a barrier to persuasive success. Conversely, though along similar lines, Nabi (2003b) found that monitors experienced greater levels of fear and were more persuaded by a mild fear appeal regarding diabetes compared to blunters. Combined, these studies suggest that coping style impacts perception of information, which in turn influences the extent of emotional arousal and perhaps persuasive success, though the nature and target of the emotional arousal is surely important to consider in this process. As we consider how this research might inform cancer control message design, we might speculate that those with monitoring tendencies will be more emotionally reactive to cancer information generally. Thus, more subtle emotional appeals may be needed to prevent excessive, and perhaps counterproductive, emotional reactivity. Conversely, those with more blunting tendencies may be more likely to experience counterproductive emotional reactions, like reactance. Clearly, there is too little evidence in this area to make definitive claims. Instead, we merely hope to convey the point that research on coping style, and other trait-based individual differences, to the extent they impact appraisals of cancer-related information, might serve as useful frameworks for understanding variations in the appraisal processes underlying emotional response. Prior knowledge As to context-based variables, prior knowledge is surely relevant to perceptions of cancer-related message information and emotional response (Nabi, 2002b). Extant research has shown prior knowledge to influence persuasive outcome by promoting more systematic processing of the message (Wood & Kallgren, 1988; Wood, Kallgren, & Preisler, 1985), stronger resistance to attitude change (Wood, 1982), and greater attitude–behavior consistency (Kallgren & Wood, 1986). As novel stimuli are more likely to evoke emotion than familiar ones, prior knowledge is poised to offer resistance to emotional arousal. Nabi (2003b) found this to be the case, showing that prior knowledge about diabetes in a sample of 18- to 72-year-olds negatively correlated with fear arousal (r=−.13, p <.05). In a cancer-related context, Nabi, Roskos-Ewoldsen, and Dillman-Carpentier (in press) found that young adults higher in self-perceived knowledge of testicular cancer or breast cancer were less likely to experience fear in response to cancer-related information (r=−.15, p <.05). This effect of knowledge dampening emotional arousal may result from people either acquiring information concerning adaptive behavior, which likely diminishes the threat and its corresponding fear, or developing strategies for defensive message processing. In either case, topic-relevant knowledge may help to explain the extent of emotional arousal in response to cancer-related messages. In sum, both trait-based and situation-specific variables may partially govern the intensity of emotional response to a persuasive message. We offer the two above as potentially valuable means of segmenting audiences of emotionally based cancer prevention and detection messages, though we recognize the need for more systematic research on these and related variables. Proposition 3: To effectively arouse emotions, cancer prevention and detection messages will need to contain information reflecting the core relational theme of the desired emotional state. Nabi’s (1999) CFM can be used to guide the development of emotion-based persuasive appeals. First, the CFM suggests that message producers should determine in advance which emotion would be best suited to their persuasive goals. For example, fear might be well suited for encouraging behaviors that clearly protect one from the severe consequences of developing cancer or from having it detected in its later stages of development. However, concerns over fear “backfiring” may suggest that other negative emotions, like sadness, disgust, or guilt, or positive emotions, like pride or compassion, might be more appropriate. Once the key emotional state is selected, the message should be designed to reflect the core relational theme of that emotion. To evoke fear, the message should depict a severe threat to which the audience is susceptible. To evoke sadness, the message should convey the sense of irrevocable loss. To evoke guilt, the message should suggest violation of an accepted norm of behavior, and so forth (see Table 2). We recognize that our discussion thus far might allow the conclusion that the task of the message designer is a simple one: First, identify the emotion to be aroused, and then create a message based on the corresponding appraisal pattern. However, such a conclusion would be not only simple but also simplistic. In fact, message designers should probably not rely on their own intuitions about how best to instantiate appraisal patterns. Dillard et al. (1996) reported that about a third of the PSAs designed as fear appeals failed to arouse fear. In a similar vein, Henley and Donovan (2003) demonstrate that widely held beliefs, such as that young people are convinced of their immortality, may have no basis in empirical reality. As always, systematic formative research is needed prior to putting any message into the field. Having explored how message features might evoke particular emotions, we now turn to how that emotional arousal might relate to persuasive outcomes. Proposition 4: Emotions can enhance, inhibit, or be unrelated to the persuasive effectiveness of cancer-related messages. First, research supports the claim that different emotional states might enhance or impede persuasive success. Dillard and Peck (2000) make this point in modeling the impact of several emotional states on the persuasiveness of eight PSAs on a range of topics (e.g., drinking and driving, community involvement, charity donations). Their results suggested that fear, happiness, sadness, surprise, and guilt positively related to perceived effectiveness, whereas anger and contentment detracted from persuasive effectiveness. These findings were replicated in a cancer-related context with the Victor Crawford data. Using LISREL 8.0, we developed a structural equation model, treating individual emotional states and dominant cognitive response as predictors of perceived effectiveness and behavioral intention.2 Surprise and sadness were positively associated with perceived effectiveness, but contentment diminished perceptions of message effectiveness (see Figure 1). Figure 1 Open in new tabDownload slide Structural equation model derived from the Victor Crawford data. Note:χ2(4) = 4.49, ns; Root Mean Squared Error of Approximation (RMSEA) =.02; Goodness of Fit Index (GFI) =.99; Bayesian Information Criterion (BIC) =−18.65. Manifest variables, error terms, and associations among exogeneous variables are not shown. Figure 1 Open in new tabDownload slide Structural equation model derived from the Victor Crawford data. Note:χ2(4) = 4.49, ns; Root Mean Squared Error of Approximation (RMSEA) =.02; Goodness of Fit Index (GFI) =.99; Bayesian Information Criterion (BIC) =−18.65. Manifest variables, error terms, and associations among exogeneous variables are not shown. Second, it is also evident that the same emotional state might, under different circumstances, enhance or inhibit persuasive success. Most famous is the early fear appeal theorizing, which was stymied by the conflicting research in which fear at times advanced, and at other times blocked, persuasive goals (see Hovland, Janis, & Kelley, 1953; Leventhal, 1970). Third, emotional arousal might not relate to persuasive outcome at all. That is, though multiple emotions may be evoked, not every emotion will produce a persuasive effect. Again, the Victor Crawford data indicate that though there was a change in almost every emotional state including fear and anger, these latter two emotions were not associated with perceived message effectiveness or behavioral intention (Dillard & Anderson, 2004, provide another example). From the standpoint of message design, statements such as Proposition 4 lack utility because they do not specify the conditions under which emotions will exhibit one of the three anticipated outcomes (i.e., enhancement, inhibition, or the absence of an effect). Before turning to a discussion of those conditions, we briefly consider the joint effects of emotion and cognition. Proposition 5: For an emotional response to exert persuasive force, it must be perceived as both caused by the message and relevant to the advocacy. As suggested by Proposition 4, emotional arousal, in and of itself, is not sufficient to support an emotion–persuasion connection. What else is necessary? We suggest that for affect to influence the evaluation of a target, that affect must be perceived as a genuine response to the message (cf. Keltner, Locke, & Audrain, 1993; Schwarz & Clore, 1983). We further propose that the perceived relevance of the emotion to the judgment at hand may serve as a second necessary parameter (cf. Gasper & Clore, 2000). Consideration of previous research, which has varied the processing goals of research participants, can help to make this point clear. For example, Martin and Stoner (1996) manipulated the relevance of affect by instructing participants in their study either to keep reading until they had enough information or to keep reading until they no longer enjoyed it. Those subjects who received enjoyment instructions seemingly used their moods as input for the decision about when to stop reading. Good mood subjects read longer than bad mood subjects. Subjects who received the information instructions showed no such difference. In a similar vein, when individuals are told to evaluate a product in terms of how much enjoyment it will produce, their intentions to consume the product are influenced by their preexisting affective state such that those in a good mood report stronger intentions than those in a bad mood (Pham, 1998). Both of these studies show that mood influences evaluation when it is seen as relevant to the judgment at hand. Proposition 6: Whether an emotion enhances or inhibits persuasion depends on its relationship with the target of evaluation. Proposition 4 asserts that emotions may enhance, inhibit, or be unrelated to the persuasive impact of an appeal. Proposition 6 specifies the conditions under which an emotion can or cannot be expected to influence an individual’s evaluation of the message and its advocacy. However, even taken together, the two propositions do not specify whether enhancement or inhibition will occur. Consideration of the target of emotional arousal offers insight on this point (Nabi, 2002b). For example, fear associated with having breast or colon cancer may promote effective action, whereas fear associated with the detection processes of a self-exam or colonoscopy may inhibit the action. Sadness over potentially losing a loved one to cancer might facilitate desired goals, whereas the sadness associated with the lack of a cure may impede them. Anger that tobacco companies push a cancer-causing product might be productive, but anger at the message producer for raising an upsetting issue will not be. With this in mind, cancer prevention and detection messages should take care that the emotion evoked is targeted toward the agent most amenable to message acceptance. Assuring appropriate emotional arousal, however, also requires guarding against the evocation of unintentional, or collateral, emotions. In accordance with Proposition 1, Dillard et al. (1996) make clear that fear appeals evoke a number of emotions other than fear, and as Proposition 4 suggests, there are times when emotions, like anger, may work against persuasive goals. Together, these propositions suggest that unintentionally aroused affects have the potential to work against persuasive goals. For example, the research suggesting that guilt appeals are counterproductive (O’Keefe, 2002) should not necessarily be interpreted to mean that guilt itself is not effective (Cotte, Coulter, & Moore, 2005). Rather, attempts to evoke guilt are likely to also evoke anger or shame, which work against persuasive goals when the target is the message source (Bennett, 1998; Coulter & Pinto, 1995; Pinto & Priest, 1991). The potential for cancer prevention and detection messages to strike the wrong chord with audiences should not be underestimated. Those who smoke may resent reminders of their unhealthful habit. Those who decline to use sunscreen may feel uncomfortable when their behavior is pointed out to them. Of course, avoiding unintentional emotional arousal is in part dependent upon understanding a target audience’s likely interpretations. Finally, we note the potential importance of considering the type of behavior advocated, which may dictate the emotional approach most likely to be successful. In recent reviews, Salovey and his colleagues conclude that, generally speaking, gain frames, which emphasize benefits of adopting an alternative behavior, are more effective when used to encourage prevention health behaviors, such as regular exercise or sunscreen use (Rothman & Salovey, 1997; Rothman, Bartels, Wlaschin, & Salovey, 2006; Salovey, Schneider, & Apanovitch, 2002). Conversely, loss frames, which emphasize costs of not adopting the advocated behavior, appear to be more effective when used in the context of detection health behaviors, like breast self-examination and mammography use (but see O’Keefe & Jensen, in press). Further, it appears that loss frames are associated with more negative emotional arousal and gain frames with more positive affects. Although this research has yet to definitively identify emotional arousal as a moderator or mediator of effects, it seems evident that emotional arousal may play some role in the effectiveness of differently framed messages. As research in this area progresses, it should offer insight into how emotional arousal might best be applied in the context of gain- and loss-framed messages for cancer prevention and detection. Proposition 7: Whether an emotion enhances or inhibits persuasion depends on audience perceptions of efficacy. Once an emotion is aroused and appropriately targeted, its persuasive impact is likely contingent upon whether the audience believes there is a way to effectively address the source of the affect. The importance of perceived efficacy has been demonstrated in the fear appeal literature both theoretically (e.g., Rogers, 1975) and empirically (Witte & Allen, 2000). That is, it is widely accepted that for fear arousal to translate into effective action, the audience must not only believe there is a protective action that can divert the threat but also believe that they themselves can perform that action. So, for example, a PSA suggesting one is at risk for skin cancer is well advised to include information on the effectiveness and ease of sunscreen use or visiting a dermatologist to have suspect growths examined and removed. Although it is possible that a previously well-informed audience might be able to “fill in” or supply the efficacy information were a message to omit it, explicit inclusion of efficacy information is a recommended message design strategy. Although response and self-efficacy perceptions have been identified as critical for fear appeal effectiveness, comparable constructs have yet to be identified for emotional appeals based on other discrete emotions in part because of the minimal theorizing about their persuasive effects. Still, we are not without guidance in this regard. The CFM suggests that once a negative discrete emotion is aroused, the likelihood that it will result in persuasive success is in part a function of the audience’s expectation of receiving reassurance from the message. That is, will the message provide goal-relevant (i.e., reassuring) information that will help deal effectively with the undesirable situation and thus alleviate the negative affect? As Nabi (1999) asserted “if afraid, receivers seek information about protection; if angry, about retribution; if sad, about coping with loss; if disgusted, about avoidance of the noxious element; and if guilt-ridden, about proper reparation.” If this is the case, then, we might extrapolate that whereas fear appeals might require information suggesting an efficacious response to protect against threat, an anger appeal would be more effective if it suggested an efficacious response to retaliate against the offending agent, a guilt appeal would be more effective if it suggested an efficacious response to make amends to the injured party, and so forth. As research on these emotions is still sparse, we are unable to offer empirical support for these assertions (though see Turner, Wang, Yao, & Xie, 2005). However, they are well motivated theoretically and are worth considering for future research and practice. Thus, just as messages should contain the information likely to stimulate the appraisals underlying the desired emotional state, the message should also likely contain information on actions one might take to successfully resolve that state. In the context of cancer control, this might mean fighting back against cancer by persisting through difficult treatment, seeking out diagnostic procedures to avoid letting one’s family down, and the like. As promoting the advocated behavior represents the primary goal of the persuasive message, the emotion selected should be sensitive to the nature of that behavior—a seemingly obvious point but one that lacks supportive empirical research at this point. Indeed, as we have reached the end of the research that speaks most directly to message design, we address, with our final proposition, the path(s) through which emotions impact persuasive outcome. Proposition 8: Emotions influence persuasion via multiple pathways. Because emotions are complex states that implicate a host of perceptual, cognitive, physiological, and motivational systems, it should come as no surprise that emotions might influence the persuasion process through multiple means. Although space prohibits a comprehensive treatment of this topic, we now address some of the mechanisms proposed to explain how emotions bring about changes in evaluation. Motivated attention and motivated processing Drawing from the literature on emotion and information processing, Nabi’s (1999) CFM claims that emotions exert their persuasive impact by influencing message processing style. If message content reflects an emotion’s core relational theme and if the receiver recognizes that theme and its personal relevance, the resulting emotional response then triggers two simultaneous motivations: motivation to attend to or to avoid the message content, message topic, or both (i.e., motivated attention) and motivation to satisfy the emotion-induced goal (i.e., motivated processing). Based on the type of emotion experienced, motivated attention sets a baseline attention level that will either impede (for avoidance emotions, like fear) or facilitate (for approach emotions, like anger) subsequent information processing. Expectation of reassurance from the message will then further shape the style of processing likely to be engaged. For example, given anger’s nature, the CFM predicts it is likely to generally promote closer information processing, whereas fear is likely to promote closer information processing only when reassurance cues are not available (see Nabi, 1999, for specific predictions). The initial test of the CFM (Nabi, 2002a) showed that anger led to more systematic information processing than fear, and for both the anger and the fear groups, uncertainty of receiving reassurance from the message resulted in more thorough processing than certainty. Applied to cancer prevention and detection messages, the CFM suggests that attention to key information can be generated by emotion type coupled with uncertainty perceptions. For example, Ruiter, Kok, Verplanken, and Brug (2001) found that evoked fear (without reassurance cues) motivated more argument-based processing of a message promoting breast self-examination (see Meijnders, Midden, & Wilke, 2001, for related findings). In terms of message design, this suggests that evoking an emotion and offering strong arguments in support of the desired behavior without obvious reassuring cues (e.g., summary headlines) may be an effective way to harness the persuasive power of typical avoidance-oriented emotions. This may be especially true for relatively novel cancer-related issues, like vaccines to prevent cervical cancer or the causes of lung cancer in nonsmokers. If, however, a message promotes prevention or detection behaviors commonly known and understood by its target audience (e.g., sunscreen use to avoid skin cancer), perhaps emotional arousal coupled with reassuring cues can be an effective strategy, especially if one is particularly worried about message-based reactance. In essence, the degree and direction of message processing naturally motivated by various emotional states can be harnessed by understanding those natural predispositions and altering message features not only to evoke those states but also to maximize their effectiveness given the unique nature of each prevention or detection context. Emotions as frames According to Nabi (2003a), emotions, once evoked, engage the same selective cognitive processes found in framing research. That is, the information made accessible from memory, the information we seek out or avoid, and the decisions that we make will be guided by the particular emotion (i.e., frame) experienced, particularly for more relevant or familiar topics. In an initial test, Nabi found general support for this model, though in the context of public policy rather than health. Applied to cancer prevention and detection messages, the emotion-as-frame hypothesis suggests that emotions evoked in those contexts (e.g., fear of diagnosis or hope of effective treatment) will guide both the information respondents will likely have accessible during message processing (e.g., the costs associated with the illness or the success rates of various treatments) and how the message itself is processed (e.g., focus on how to avoid the threat of illness or on the likelihood of survival). The effects of emotional arousal, though, are expected to be influenced by prior knowledge, coping style, and other individual differences that shape responses to message content. For example, past research suggests that more subjectively knowledgeable audience members may experience weaker affective reactions to emotional appeals. But if emotions serve a framing function, the effects of that (perhaps dampened) emotional arousal may be equally or even more likely to translate into a persuasive effect than more intense emotions aroused in those with a less developed knowledge base, which limits the information made accessible and available to guide behavior. In the latter case, credible message information relevant to the particular emotional state (e.g., protection information for fear, retributive information for anger) may prove especially important to promote persuasive success. The implications of these ideas are that to be effective, cancer-related persuasive appeals need to be responsive to audience factors such that at times, less information or less emotional evocation may be more effective than messages that are completely articulated (see Nabi et al., in press). For example, to the extent that the idea of cancer is fear inducing, it may not be necessary to point out that it poses a severe threat to health. To do so may, in fact, be counterproductive. Similarly, to point out known responses, such as smoking cessation to avoid lung cancer, may be equally unnecessary. However, for those audiences who either have no particular emotional association with cancer, related prevention, or detection behaviors, or have counterproductive associations (e.g., fear associated with treatment), emotional appeals may be important to create such associations, which in turn may facilitate desired behavior. Discrete emotions activate behavior Several theorists contend that the function of emotions is to direct behavior (Frijda, Kuipers, & ter Schure, 1989; Roseman, Wiest, & Swartz, 1994). Thus, each emotion has an associated motivation known as an action tendency. At the general level, action tendencies can be characterized in terms of engagement and withdrawal. However, because emotions are evolutionarily designed to solve specific problems, the action tendency associated with each emotion is (a) a more specific variation on one, the other, or both of those two broad themes and (b) responsive to context. For example, anger is generally viewed as an engagement emotion (Harmon-Jones & Allen, 1998). Yet, it can invite persuasive attack (i.e., a form of engagement) or simple rejection when it encourages dismissal of a message on the basis of topic or message position (i.e., a form of withdrawal) (Dillard & Peck, 2001). Presumably, action tendencies have direct persuasive effects in that they map directly on to evaluations. This point stands in distinction to the other mechanisms discussed above that suggest the effects of emotion on evaluation are mediated (or at least moderated) by some type of cognitive change (e.g., deeper or more selective processing). The direct effects view of emotion is also distinct from the way in which cognitive response models of persuasion, which have dominated persuasion research in recent years, consider the role of affect. The dual-process models of attitude change (Briñol & Petty, 2006; Shen & Chaiken, 1999), which suggest that persuasion is the result of cognitive responses to the message, message topic, or persuasive cues present during message exposure, consider the effects of affect (i.e., mood and emotions), but only insofar as they influence cognition (e.g., DeSteno, Petty, Rucker, Wegener, & Braverman, 2004). In essence, none of these views addresses the direct effect of emotions on attitudes. In contrast, our theorizing and data cast cognition and emotion as concurrent phenomena that both possess the potential for persuasive impact. Dillard et al. (1996) and Dillard and Peck (2000) support this point. Through regression analyses and structural equation modeling, these studies demonstrate that emotional states explained variance in perceived message effectiveness, beyond that accounted for by cognitive responses. The Victor Crawford data extend this finding to a cancer-related context (see Figure 1). Despite the influence of dominant cognitive response, multiple affective states predicted perceived message effectiveness that, in turn, predicted behavioral intentions. Thus, although emotions may surely impact persuasive outcome via cognitive routes, they also appear to have a direct path to persuasion, making them all the more attractive to consider in trying to promote prevention and detection health behaviors. Future research that considers how discrete emotional states operate to directly affect attitudes and behavioral intentions in different cancer-related contexts (e.g., fear and self-exam behavior or disgust and antismoking sentiment) would be most welcome. Summary and conclusions Our goal in this paper was to offer insight into how persuasive messages might be designed to evoke emotions that can effectively promote cancer prevention and detection behaviors. We argued that the first task for those designing emotion-based cancer prevention and/or detection messages is to assess what emotion might be most amenable to the persuasive goals. For example, fear might be a productive means of alerting audiences to their susceptibility to risks based on their family history, environment, or previous behaviors. Sadness can be useful for encouraging reflection on the consequences of failure to take preventive action. Disgust may help create negative associations with risky behaviors, whereas evoking hope may spur action that would otherwise be seen as impossible. Matching persuasive goals with the essence of each unique emotional state will increase the likelihood of productive emotional arousal. After selecting an emotional strategy, appraisal theory can be used to guide the message content chosen for emotional evocation. For example, to evoke fear, the message should contain information depicting a severe threat to which the audience is susceptible. To evoke sadness, the message should demonstrate irrevocable loss. Further, we assert that to enhance the likelihood of persuasive success, the affect should be appropriately targeted, and efficacy information for how to divert threat (fear) or loss (sadness) or achieve retribution (anger) or restitution (guilt) should be made evident. Moreover, though there is little research in this domain, we argue that sensitivity to audience characteristics, like knowledge and coping style, in message design is also important. To the best of our ability, we have summarized existing theory and research on the role of emotion in persuasion. And, we have attempted to illuminate both the potentials and the pitfalls of applying emotion/persuasion theory to cancer prevention and detection. To the extent that we have succeeded in these aims, then we have also, by implication, made clear the enormity of what remains unknown. In particular, under what conditions should emotional appeals be used rather than more rational appeals? What is the impact of intermingling emotional and rational messages? How should campaigns sequence the presentation of different types of appeals? These queries serve to remind us that the use of emotional persuasion in cancer prevention and detection is but one component of the much larger question of how to use communication to improve public health. Notes 1 " Two hundred thirty-eight University of Wisconsin–Madison undergraduates viewed the PSA, then indicated their emotional reactions on multi-item scales designed to measure fear, anger, surprise, sadness, happiness, and contentment. Information on the reliability and validity of the scales, as well as the items themselves, can be found in Dillard et al. (1996), Dillard and Peck (2000, 2001), and Dillard and Anderson (2004). 2 " A measure of dominant cognitive response was created by coding open-ended responses into three categories—supporting thoughts, neutral thoughts, and counterarguments—and then subtracting the number of counterarguments from the number of supporting thoughts. Perceived effectiveness was measured with a series of 7-point semantic differential items (persuasive/not persuasive, convincing/not convincing, effective/ineffective, compelling/not compelling). Behavioral intention was tapped by three 5-point Likert-type items (e.g., “I intend to behave in ways that are consistent with the message,”“I plan to act in ways that are compatible with the position promoted by the message,” and “I am going to make an effort to do what the message urged me to do”). Higher values indicate higher perceived effectiveness and stronger intention to behave. References Bennett , R . ( 1998 ). Shame, guilt & responses to non profit and public sector ads . International Journal of Advertising , 17 , 483 – 499 . Google Scholar Crossref Search ADS WorldCat Briñol , P. , & Petty , R. E. ( 2006 ). Fundamental processes leading to attitude change: Implications for cancer prevention communication . Journal of Communication , 56 , S81 – S104 . Google Scholar Crossref Search ADS WorldCat Cotte , J. , Coulter , R. A., & Moore , M. ( 2005 ). Enhancing or disrupting guilt: The role of ad credibility and perceived manipulative intent . Journal of Business Research , 58 , 361 – 368 . Google Scholar Crossref Search ADS WorldCat Coulter , R. H. , & Pinto , M. B. ( 1995 ). Guilt appeals in advertising: What are their effects? Journal of Applied Psychology , 80 , 697 – 705 . Google Scholar Crossref Search ADS PubMed WorldCat DeSteno , D. , Petty , R. E., Rucker , D. D., Wegener , D. T., & Braverman , J. ( 2004 ). Discrete emotions and persuasion: The role of emotion-induced expectancies . Journal of Personality and Social Psychology , 86 , 43 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat Dillard , J. P. , & Anderson , J. W. ( 2004 ). The role of fear in persuasion . Psychology & Marketing , 21 , 909 – 926 . Google Scholar Crossref Search ADS WorldCat Dillard , J. P. , & Peck , E. ( 2000 ). Affect and persuasion: Emotional responses to public service announcements . Communication Research , 27 , 461 – 495 . Google Scholar Crossref Search ADS WorldCat Dillard , J. P. , & Peck , E. ( 2001 ). Persuasion and the structure of affect: Dual-systems and discrete emotions as complementary models . Human Communication Research , 27 , 38 – 68 . Google Scholar Crossref Search ADS WorldCat Dillard , J. P. , Plotnick , C. A., Godbold , L. C., Freimuth , V. S., & Edgar , T. ( 1996 ). The multiple affective outcomes of AIDS PSAs: Fear appeals do more than scare people . Communication Research , 23 , 44 – 72 . Google Scholar Crossref Search ADS WorldCat Frijda , N. , Kuipers , P., & Ter Schure , E. ( 1989 ). Relations among emotion, appraisal, and emotional action readiness . Journal of Personality and Social Psychology , 57 , 212 – 228 . Google Scholar Crossref Search ADS WorldCat Gasper , K. , & Clore , G. L. ( 2000 ). Do you have to pay attention to your feelings to be influenced by them? Personality and Social Psychology Bulletin , 26 , 698 – 711 . Google Scholar Crossref Search ADS WorldCat Harmon-Jones , E. , & Allen , J. B. ( 1998 ). Anger and frontal brain activity: EEG asymmetry consistent with approach motivation despite negative valence . Journal of Personality and Social Psychology , 74 , 1310 – 1316 . Google Scholar Crossref Search ADS PubMed WorldCat Henley , N. , & Donovan , R. J. ( 2003 ). Young people’s response to death threat appeals: Do they really feel immortal? Health Education Research , 18 , 1 – 14 . Google Scholar Crossref Search ADS PubMed WorldCat Hovland , C. I. , Janis , I. L., & Kelley , H. H. ( 1953 ). Communication and persuasion . New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kallgren , C. A. , & Wood , W. ( 1986 ). Access to attitude-relevant information in memory as a determinant of attitude-behavior consistency . Journal of Experimental Social Psychology , 22 , 328 – 338 . Google Scholar Crossref Search ADS WorldCat Keltner , D. , Locke , K. D., & Audrain , P. C. ( 1993 ). The influence of attributions on the relevance of negative feelings to personal satisfaction . Personality and Social Psychology Bulletin , 19 , 21 – 29 . Google Scholar Crossref Search ADS WorldCat Lazarus , R. S . ( 1991 ). Emotion and adaptation . New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Leventhal , H . ( 1970 ). Findings and theory in the study of fear communications. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 5 , pp. 119 – 186 ). New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Martin , L. L. , & Stoner , P. ( 1996 ). Mood as input: What we think about how we feel determines how we think. In L. L. Martin & A. Tesser (Eds.), Striving and feeling: Interactions among goals, affect, and self-regulation (pp. 279 – 301 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Meijnders , A. L. , Midden , C. J. H., & Wilke , H. A. M. ( 2001 ). Role of negative emotion in communication about CO2 risks . Risk Analysis , 21 , 955 – 966 . Google Scholar Crossref Search ADS PubMed WorldCat Miller , S. M . ( 1987 ). Monitoring and blunting: Validation of a questionnaire to assess styles of information seeking under threat . Journal of Personality and Social Psychology , 52 , 345 – 353 . Google Scholar Crossref Search ADS PubMed WorldCat Miller , S. M . ( 1991 ). Monitoring and blunting in the face of threat: Implications for adaptation and health. In L. Montada, S. H. Filipp, & M. J. Lerner (Eds.), Life crises and experiences of loss in adulthood . Englewood Cliffs, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Miller , S. M . ( 1995 ). Monitoring and blunting styles of coping with cancer influence the information patients want and need about their disease . Cancer , 76 , 167 – 177 . Google Scholar Crossref Search ADS PubMed WorldCat Miller , S. M. , Shoda , Y., & Hurley , K. ( 1996 ). Applying cognitive-social theory to health-protective behavior: Breast self-examination in cancer screening . Psychological Bulletin , 119 , 70 – 94 . Google Scholar Crossref Search ADS PubMed WorldCat Nabi , R. L . ( 1999 ). A cognitive-functional model for the effects of discrete negative emotions on information processing, attitude change, and recall . Communication Theory , 9 , 292 – 320 . Google Scholar Crossref Search ADS WorldCat Nabi , R. L . ( 2002a ). Anger, fear, uncertainty, and attitudes: A test of the cognitive-functional model . Communication Monographs , 69 , 204 – 216 . Google Scholar Crossref Search ADS WorldCat Nabi , R. L . ( 2002b ). Discrete emotions and persuasion. In J. Dillard & M. Pfau (Eds.), Handbook of persuasion (pp. 289 – 308 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nabi , R. L . ( 2003a ). The framing effects of emotion: Can discrete emotions influence information recall and policy preference? Communication Research , 30 , 224 – 247 . Google Scholar Crossref Search ADS WorldCat Nabi , R. L . ( 2003b ). Monitoring and blunting coping styles as moderators of fear appeal effects . Paper presented at the 53rd Annual Conference of the International Communication Association , San Diego, CA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nabi , R. L. , Roskos-Ewoldsen , D., & Dillman-Carpentier , F. ( in press ). Subjective knowledge and fear appeal effectiveness: Implications for message design . Health Communication . OpenURL Placeholder Text WorldCat O’Keefe , D. J . ( 2002 ). Guilt as a mechanism of persuasion. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 329 – 344 ). Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC O’Keefe , D. J. , & Jensen , J. D. ( in press ). The advantages of compliance or the disadvantages of noncompliance? A meta-analytic review of the relative persuasive effectiveness of gain-framed and loss-framed messages. In C. S. Beck (Ed.), Communication yearbook 30 . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pham , M. T . ( 1998 ). Representativeness, relevance, and the use of feelings in decision making . Journal of Consumer Research , 25 , 144 – 159 . Google Scholar Crossref Search ADS WorldCat Pinto , M. B. , & Priest , S. ( 1991 ). Guilt appeals in advertising: An exploratory study . Psychological Reports , 69 , 375 – 385 . Google Scholar Crossref Search ADS PubMed WorldCat Rogers , R. W . ( 1975 ). A protection motivation theory of fear appeals and attitude change . The Journal of Psychology , 91 , 93 – 114 . Google Scholar Crossref Search ADS PubMed WorldCat Roseman , I. J. , Wiest , C., & Swartz , T. S. ( 1994 ). Phenomenology, behaviors, and goals differentiate discrete emotions . Journal of Personality and Social Psychology , 67 , 206 – 221 . Google Scholar Crossref Search ADS WorldCat Rothman , A. J. , Bartels , R. D., Wlaschin , J., & Salovey , P. ( 2006 ). The strategic use of gain and loss framed messages to promote healthy behavior: How theory can inform practice . Journal of Communication , 56 , S202 – S220 . Google Scholar Crossref Search ADS WorldCat Rothman , A. J. , & Salovey , P. ( 1997 ). Shaping perceptions to motivate healthy behavior: The role of message framing . Psychological Bulletin , 121 , 3 – 19 . Google Scholar Crossref Search ADS PubMed WorldCat Ruiter , R. A. C. , Kok , F., Verplanken , B., & Brug , J. ( 2001 ). Evoked fear and effects of appeals on attitudes to performing breast self-examination: An information-processing perspective . Health Education Research , 16 , 307 – 319 . Google Scholar Crossref Search ADS PubMed WorldCat Salovey , P. , Schneider , T. R., & Apanovitch , A. M. ( 2002 ). Message framing in the prevention and early detection of illness. In J. P. Dillard & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 391 – 406 ). Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Scherer , K. R. , Schorr , A., & Johnstone , T. (Eds.). ( 2001 ). Appraisal processes in emotion . New York : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schwarz , N. , & Clore , G. L. ( 1983 ). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states . Journal of Personality and Social Psychology , 45 , 513 – 523 . Google Scholar Crossref Search ADS WorldCat Shen , S. , & Chaiken , S. ( 1999 ). The heuristic-systematic model in broader context. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 73 – 96 ). New York : Guilford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Smith , C. A. , & Lazarus , R. S. ( 1993 ). Appraisal components, core relational themes, and the emotions . Cognition and Emotion , 7 , 233 – 269 . Google Scholar Crossref Search ADS WorldCat Turner , M. M. , Wang , Q., Yao , S., & Xie , X. ( 2005, May ). The targeted anger appeals processing model: A new approach to persuasive message design . Paper presented at the 55th Annual Conference of the International Communication Association , New York . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Williams-Piehota , P. , Pizarro , J., Schneider , T. R., Mowad , L., & Salovey , P. ( 2005 ). Matching health messages to monitor-blunter coping styles to motivate screening mammography . Health Psychology , 24 , 58 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat Witte , K. , & Allen , M. ( 2000 ). A meta-analysis of fear appeals: Implications for effective public health campaigns . Health Education & Behavior , 27 , 591 – 615 . Google Scholar Crossref Search ADS WorldCat Wood , W . ( 1982 ). Retrieval of attitude-relevant information from memory: Effects on susceptibility to persuasion and on intrinsic motivation . Journal of Personality and Social Psychology , 42 , 798 – 810 . Google Scholar Crossref Search ADS WorldCat Wood , W. , & Kallgren , C. A. ( 1988 ). Communicator attributes and persuasion: Recipients’ access to attitude-relevant information in memory . Personality and Social Psychology Bulletin , 14 , 172 – 182 . Google Scholar Crossref Search ADS WorldCat Wood , W. , Kallgren , C. A., & Preisler , R. M. ( 1985 ). Access to attitude-relevant information in memory as a determinant of persuasion: The role of message attributes . Journal of Experimental Social Psychology , 21 , 73 – 85 . Google Scholar Crossref Search ADS WorldCat © 2006 International Communication Association

Briñol,, Pablo;Petty, Richard, E.

doi: 10.1111/j.1460-2466.2006.00284.xpmid: N/A

Abstract Developing appropriate intervention strategies for cancer-related communications depends on understanding the basic mechanisms underlying persuasion. By examining the psychological processes through which attitudes change, cancer researchers can understand and predict further changes in behavior and maximize the chances of designing effective research and interventions. In the present article, we group the main persuasion processes into meaningful categories to provide a useful guide to organize the reviewed findings. One of the most common findings on cancer communication has been that matching persuasive messages to people’s characteristics increases persuasion. The present review provides a detailed examination of the different psychological mechanisms through with such persuasive matching effects and exceptions might occur. Given that most people want to stay healthy and avoid personal injury, one might expect that changing health-relevant attitudes and behaviors would be relatively easy. Yet, it is not. For example, although adolescents are well aware of the health risks associated with cigarette smoking, a substantial minority continue to choose to smoke (e.g., Leventhal & Cleary, 1980). Similarly, despise the fact that most people are fully aware of the risks involved in carrying extra poundage, most of the people who lose weight by dieting will gain it all back (and more). In addition, despite the fact that people increasingly appreciate the risk of melanoma associated with tanning, tanning salons continue in business. As noted by Salovey and Rothman (2003), despite the fact that early detection of breast cancer allows women to have many more alternatives for effective treatment, only about half of the women who should obtain a screening mammogram actually do so. As might be expected by these examples, numerous studies of the effectiveness of media and direct health interventions have been disappointing in terms of producing or sustaining attitude and behavior change (see Devos-Comby & Salovey, 2002; Rothman, 2000). Indeed, different investigators have pointed out that simply providing information and increasing knowledge about a topic is not sufficient to lead to attitude and behavior change (Helweg-Larsen & Collins, 1997; Petty, Baker, & Gleicher, 1991). Our view is that persuasive communications and behavioral interventions can be made more effective to the extent that they are designed with relevant psychological theory in mind. Along with other social psychologists (see Salovey & Rothman, 2003), we believe that developing and selecting appropriate intervention strategies for social influence depends on understanding the basic mechanisms underlying persuasion.1 Moreover, we believe that understanding why a particular intervention technique is effective increases the probability that this procedure can be integrated into interventions across diverse health domains. The primary goal of this article is to explicate the primary psychological processes responsible for attitude change. We focus on psychological processes because of their pivotal role in understanding and determining health behavior. In particular, certain mechanisms of change are more likely to produce attitudes that are strong (i.e., persistent over time, resistant to change, and predictive of behavior). The article is divided into five sections. We describe (a) the key psychological processes underlying attitude change; (b) the relationship between communication variables and attitude change processes and their implications for attitude strength; (c) the influence of one highly studied variable, message matching (i.e., targeting, tailoring), on attitude change and strength; (d) the explicit and implicit psychological processes underlying the persuasive effects of matching; and (e) the implications of understanding attitude change processes for behavioral change. Fundamental processes underlying attitude change In this section, we describe the fundamental processes by which any communication variable can influence attitude change. By the term communication variable we refer to any aspect of the source (e.g., credibility), message (e.g., number of arguments), recipient (e.g., mood), or context (e.g., presence of distraction) that can vary in a given persuasion situation. The number of variables of potential relevance to persuasion is endless, so we will focus on some that have received research attention. Consistent with the Elaboration Likelihood Model (ELM) of persuasion (Petty & Cacioppo, 1986) we argue that any variable (i.e., whether related to source, message, recipient, or context) can influence attitude change by affecting the amount of thinking, the direction of thinking, structural features of thoughts, and by serving as arguments (evidence) or as cues (see Petty & Wegener, 1998a). Amount of thinking One of the most fundamental influences that a variable can exert on attitudes is to affect the amount of thinking people do about a persuasive communication. Increasing the amount of thinking can get people to carefully process the relevant information presented and therefore be influenced by it. The more motivated and able people are to think about a message, the more their attitudes are determined by their issue-relevant thoughts in response to the message (Petty, Ostrom, & Brock, 1981). In a persuasion context, issue-relevant elaboration typically involves accessing relevant information from both external and internal sources; scrutinizing, making inferences, generating new arguments, and drawing new conclusions about the merits of the attitude object (Petty & Cacioppo, 1986).2 These mental activities are likely to lead to the integration of all relevant information into the underlying structure (schema) for the attitude object, therefore making the adopted evaluation stable and coherent. Thus, attitudes based on high amounts of thinking are postulated to be stronger than attitudes based on little thought. That is, they are proposed to be more accessible, stable, resistant to countermessages, and predictive of behavior (see Petty, Haugtvedt, & Smith, 1995; see also Petty & Briñol, in press). Considering the amount of thinking underlying attitude change is fundamental for cancer-related communications because the overall goal of these messages is to produce long-term behavior change. The difficulties of creating thoughtful attitude change are familiar to health promotion researchers. For example, there are great challenges in engaging young adults in cancer prevention topics such as dieting simply because they often do not see them as personally relevant or important to their lives. Because of these challenges, it is essential for cancer-related researchers to understand what variables successfully engage the thoughtful processing of cancer-related communications in each population. A variety of different variables have been examined that can influence attitude change by affecting people’s general motivation and ability to think about a message. Perhaps the most important determinant of interest and motivation to process a message is the perceived personal relevance of the communication. The basic notion is that there are some situations in which some aspect of the persuasive message or issue can be linked to some aspect of the message recipient’s “self,” making the message personally relevant. Linking the message to almost any aspect of the self, such as one’s values, one’s outcomes, one’s self-conception, one’s identity, and so forth, can enhance self-relevance (Fleming & Petty, 2000; Petty & Cacioppo, 1990). For example, if a woman has a family history of breast cancer, then she might be motivated to think about a persuasive message about breast self-examination based on perceived self-relevance (Rothman & Schwarz, 1998). In a powerful demonstration of the power of perceived self-relevance, Burnkrant and Unnava (1989) found that simply changing the pronouns in a message from the third person (e.g., “one” or “he and she”) to the second person (“you”) was sufficient to increase personal involvement and processing of the message arguments. Similarly, Rothman, Salovey, Turvey, and Fishkin (1993) demonstrated that a persuasive communication emphasizing a woman’s own responsibility for getting a mammogram (“Eight out of 10 lumps that you might find will not be breast cancer.”) had more impact in the use of screening mammography than a communication emphasizing external responsibility for detecting breast cancer (e.g., “Eight out of 10 lumps that a doctor might find will not be breast cancer”). Notably, by increasing the personal relevance of a message, people scrutinize the evidence more carefully such that if the evidence is found to be strong, more attitude change results, but if the evidence is found to be weak, less attitude change occurs (Petty & Cacioppo, 1979, 1990). In addition to risk status, there are additional factors that might be relevant to increase perceived personal relevance of a cancer-related communication. One of potential importance is early diagnosis. Early diagnosis of possible cancer either through the discovery of elevated prostate specific antigen (PSA), colon polyps, or some other result that could possibly be benign would likely elevate relevance. Unfortunately, people often tend to be overconfident in their optimistic views regarding perceived risk, making it difficult to lead them to preventive behaviors (e.g., Dunning, Heath, & Suls, 2004). For example, smokers underestimate the likelihood of lung cancer death rates compared to nonsmokers, and differences hold even between light and heavy smokers (e.g., Leventhal & Cleary, 1980). Because unrealistic optimism is based in part on a need to defend the self against possible threats (Weinstein, 2003), persuasive treatments that allow people to self-affirm themselves prior to receipt of a message might be particularly useful for health communications (e.g., Sherman, Nelson, & Steele, 2000; see also Briñol, Petty, Gallardo, & DeMarree, in press). Although increasing the perceived personal relevance of a message is an effective way to increase thinking, it is not the only one. For example, several studies have shown that when a person is not normally motivated to think about the message arguments, more thinking can be provoked by summarizing the major arguments as questions rather than as assertions (Howard, 1990; Petty, Cacioppo, & Heesacker, 1981; Swasy & Munch, 1985), by having the individual arguments presented by multiple independent sources rather than just one (Harkins & Petty, 1981; Moore & Reardon, 1987), and by making some aspect of the message surprising or unexpected (Baker & Petty, 1994; Smith & Petty, 1996). Having the necessary motivation to process a message is not sufficient for thoughtful central route persuasion to occur, however. In addition, to create strong, consequential attitudes, people must also have the ability to process the message. For example, a complex or long message might require more than one exposure for maximal processing, even if the recipient was highly motivated to think about it (Cacioppo & Petty, 1989). Of course, repetition is just one variable that has an impact on a person’s ability to think about a message. For example, if a message is accompanied by distraction (Petty, Wells, & Brock, 1976) or if the speaker talks too fast (Briñol & Petty, 2003; Smith & Shaffer, 1991), thinking about the message will be disrupted. Anxiety (e.g., about a possible positive test result) could also reduce the ability to process cancer communications. Different media sources also have an impact on peoples’ ability to think about the message. Specifically, people are generally better able to process messages that appear in the print media than those that are controlled externally (e.g., radio and television; Chaiken & Eagly, 1976; Wright, 1981). When strong arguments are presented, disrupting thinking should diminish persuasion, but when weak arguments are presented, disrupting thinking should actually enhance persuasion by reducing the counterarguing that would have occurred. Type or direction of thinking When motivation and ability to think are high, people will be engaged in careful thought about a message, but that thinking can be biased by other variables in the persuasion setting. Most importantly, variables can motivate or enable people to either support or derogate the content of the information provided. The success of information in changing attitudes is a function of the number and valence of thoughts that come to mind when elaboration is high (see reviews by Eagly & Chaiken, 1993; Petty & Cacioppo, 1986). Some features of the persuasion situation increase the likelihood of favorable thoughts being elicited, but others increase the likelihood of unfavorable thoughts coming to mind. Although the subjective cogency of the arguments used in a message is a prime determinant of whether favorable or unfavorable thoughts are elicited when message thinking is high, other variables can also be influential in determining whether favorable or unfavorable thoughts predominate (Petty & Cacioppo, 1990). For example, instilling “reactance” in message recipients by telling them that they have no choice but to be persuaded on an important issue motivates counterarguing even when the arguments used are strong (Brehm, 1966; Petty & Cacioppo, 1979). On the other hand, if people are put in a good mood prior to hearing an involving message, their thoughts will be biased in favor of the message (Petty, Schumann, Richman, & Strathman, 1993). Biased thinking, whether favorable or unfavorable, often reduces the impact of message quality on persuasion compared to objective thinking (Petty & Cacioppo, 1986). Ability factors can also produce bias. For example, people who possess accessible attitudes bolstered by considerable attitude-congruent knowledge are better able to defend their attitudes than those who have inaccessible attitudes or attitudes with a minimal underlying foundation (Fazio & Williams, 1986; Wood, 1982). In general, anytime a message takes a position opposed to one’s attitudes, or disfavors one’s values, outcomes, or groups, people will be biased against it. And, when a message takes a position in favor of one’s attitudes, or is supportive of one’s values, outcomes, or groups, people will be biased in favor of it. To be clear, as noted earlier, when a message is framed as simply relevant to one’s attitudes, outcomes, values, or identity, the extent of information processing is affected, but when a message takes a particular position with respect to one’s attitudes, outcomes, values, or identity, the valence of the processing can be affected (Petty & Cacioppo, 1990). Structural features of thoughts The structural features of thoughts refer to dimensions other than direction (favorable or unfavorable) and amount (high or low). In particular, in this section we refer to metacognitive aspects of thinking, or thoughts about thoughts (for a review on this topic, see Petty, Briñol, Tormala, & Wegener, in press), though other structural features such as accessibility of thoughts are also important. When the amount of thinking is high, variables can affect metacognitive features of the thoughts that are generated such as how much confidence people have in their thoughts or how biasing they seem. Confidence in thoughts is important because when thoughts are held with greater confidence, people are more likely to use them in forming their judgments (Petty, Briñol, & Tormala, 2002). On the other hand, if people doubt the validity of their thoughts, the thoughts will not have an impact on judgments. This may be one reason why some communication campaigns are unsuccessful. That is, they may produce the appropriate favorable thoughts, but these thoughts may not be held with sufficient confidence to affect judgments. If people believe that their thoughts are biasing in some way, they can adjust their judgments in a direction opposite to the implication of the thoughts (correction processes; Petty & Wegener, 1993; Wegener & Petty, 1997; Wilson & Brekke, 1994). These metacognitive processes have the greatest impact when the amount of thinking is high because it is largely in such situations that people have a substantial number of issue-relevant thoughts with the potential to shape attitudes. Thus, individual and situational differences in the extent of thinking moderate these metacognitive processes. The most studied metacognitive process regarding thoughts is referred to as self-validation (Petty, Briñol, et al., 2002), or the process by which people come to have confidence or doubt in their thoughts. Thought confidence varies as a function of many factors (for a review, see Briñol & Petty, 2004). The factors affecting confidence range from individual variables, such as a person’s current mood state and overt behavior (Briñol & Petty, 2003), to situational factors such as the credibility of the source associated with the message (Briñol, Petty, & Tormala, 2004; Tormala, Briñol, & Petty, in press), the number of other people who share one’s thoughts (Petty, Briñol, et al., 2002), and the ease with which their thoughts come to mind (Tormala, Petty, & Briñol, 2002). Apanovitch, McCarthy, and Salovey (2003) proposed that the certainty associated with the likelihood of a test’s outcome moderates the traditional effects of message framing. These authors found that only among participants who reported being certain of an HIV test outcome was a gain-framed video more successful than a loss-framed message. Among women who perceived the outcome of HIV testing as relatively uncertain, gain- and loss-framed videos produced similar effects. This research suggests that assessing the confidence with which people hold their thoughts in response to prevention communications (and manipulating the sources of that confidence) might contribute to the design of effective campaigns. Serving as arguments When the amount of thinking is high, people assess the relevance of all of the information in the context and that comes to mind in order to determine the merits of the attitude object under consideration, that is, people examine source, message, recipient, contextual and internally generated information, and feelings—as possible arguments or reasons for favoring or disfavoring the attitude object. Individuals vary in what type of information serves as persuasive evidence for any given attitude object. For example, when thinking carefully, source attractiveness can serve as an argument for a skin protection product (i.e., “If I use the product, I’ll look like that”). As explained shortly, for other products, attractiveness might serve as a simple cue. Serving as cues Variables can influence attitudes by serving as simple cues. That is, under low-thinking conditions, attitudes are influenced by a variety of low-effort processes such as mere association (Cacioppo, Marshall-Goodell, Tassinary, & Petty, 1992) or reliance on simple heuristics (Chaiken, 1987). This is important because it suggests that attitude change does not always require effortful evaluation of the information for its merits. Instead, when a person’s motivation or ability to process the issue-relevant information is low, persuasion can occur by a peripheral route in which processes invoked by simple cues in the persuasion context influence attitudes. For example, various features of a communication (e.g., pleasant scenery in a TV commercial or an attractive model) can elicit an affective state (e.g., a good mood) that becomes associated with the advocated position (as in classical conditioning, Staats & Staats, 1958). Or, the source of a message can trigger a relatively simple inference or heuristic such as “experts are correct,” (Chaiken) that a person can use to judge the message. Similarly, a person might simply count the arguments and reason that “if there are so many arguments it must be good” (Petty & Cacioppo, 1984a). Peripheral ways to change attitudes can be very powerful in the short term. The problem is that over time, moods dissipate, peoples’ feelings about sources can change, and the cues can become dissociated from the message. These factors would then undermine the initial attitude change. Research has shown that attitude changes based on peripheral cues tend to be less accessible, enduring, and resistant to subsequent attacking messages than attitudes based on careful processing of message arguments (see Petty et al., 1995, for a review).3 The influence of communication variables on persuasion The Elaboration Likelihood Model of persuasion (ELM; Petty & Cacioppo, 1981, 1986) identifies the key processes just reviewed and highlights their role in producing attitude changes that are consequential or not (see Petty, Priester, & Briñol, 2002; Petty & Wegener, 1999). As depicted in Figure 1, the ELM postulates that any communication variable (i.e., whether source, message, recipient, or context) influences attitudes by affecting one of the key processes of persuasion. That is, any given feature of the persuasive setting (e.g., a recipient’s mood, the attractiveness of the message source) can serve as an issue-relevant argument, or a peripheral cue, or affect the motivation or ability to think about the message, bias the nature of the thoughts that come to mind, or affect structural properties of the thoughts such as how much confidence people have in them. Figure 1 Open in new tabDownload slide Fundamental processes in attitude change: antecedents and consequences. Figure 1 Open in new tabDownload slide Fundamental processes in attitude change: antecedents and consequences. If any one variable can influence persuasion by several means, it becomes critical to identify the general conditions under which the variable acts in each of the different roles or the ELM becomes descriptive rather than predictive (cf., Stiff, 1986). The ELM holds that when the elaboration likelihood is high (such as when perceived personal relevance is high), any given feature of the persuasive setting can serve as an argument if it is relevant to the merits of the issue. In addition, when thinking is high, variables can influence the valence of the thoughts that come to mind, thereby biasing the ongoing thinking. Or, variables can influence the structural properties of the cognitive responses that occur. The first two processes tend to occur when the variable is introduced either prior to or during the message (i.e., before or during message processing). The latter process (self-validation) tends to occur largely after thoughts are generated. On the other hand, when the elaboration likelihood is low (e.g., low personal relevance), evaluations are likely to be the result of relatively simple associations or inferences based on salient cues. Thus, variables can determine attitudes under low elaboration to the extent that they can function as simple cues. Finally, when the elaboration likelihood is moderate (e.g., uncertain personal relevance), people may examine the persuasion context for indications (e.g., is the source credible?) of whether or not they are interested in or should process the message. A few examples should help to clarify the multiple roles that any given feature of the persuasive setting can have in different situations. As one example, consider the mere number of arguments that a message contains. This variable serves as a simple peripheral cue when people are either unmotivated or unable to think about the information (Petty & Cacioppo, 1984a). That is, people might simply count the arguments in a message and agree more, the more information that is presented—regardless of the cogency of that information. When motivation and ability are high, however, the informational items in a message are not simply counted and used in a simple inference (more is better), but instead the information is processed for its quality. Thus, when the number of items in a message serves as a cue (low-elaboration conditions), adding weak reasons in support of a position enhances persuasion, but when the items in a message are processed in a qualitatively different way (for merit), adding weak reasons reduces persuasion (Alba & Marmorstein, 1987; Friedrich, Fetherstonhaugh, Casey, & Gallagher, 1996; Petty & Cacioppo). The mere number of arguments is only one of the message factors that can influence persuasion through different processes in different situations. To take one more example, consider the complexity of the message (e.g., difficult vocabulary, sentence structure, etc.) that is characteristic of cancer communications that include technical terminology. Such complexity could serve as a simple cue when the elaboration likelihood is low. For example, a person might use the heuristic, “this pamphlet seems to include a lot of complex information about sun blocking, therefore using it might be good.” When the elaboration likelihood is not constrained to be high or low, complexity might affect the amount of thinking that occurs. That is, some people (e.g., those high in need for cognition; Cacioppo & Petty, 1982) might be challenged by a message that seems complex, but other individuals (e.g., those low in need for cognition) might eschew processing a message that is perceived as difficult (Evans & Petty, 2003). Finally, under high-elaboration conditions, other roles for message complexity are possible. In one study, for instance, it was shown that under high-elaboration conditions, complex information undermined people’s confidence in their thoughts (cf. Petty & Briñol, 2002). Source factors, such as expertise or attractiveness, have been also found to affect attitude change through different processes depending on the situation (see Petty & Cacioppo, 1984b). Depending on the elaboration likelihood and other factors (e.g., relevance to the attitude object, placement before or after the message), source factors have been found to influence persuasion by serving as a peripheral cue (Petty, Cacioppo, & Goldman, 1981; see also Chaiken, 1980), by affecting how much thinking people did about the message (Heesacker, Petty, & Cacioppo, 1983; Moore, Hausknecht, & Thamodaran, 1986; Puckett, Petty, Cacioppo, & Fisher, 1983), by biasing the direction of the thoughts (Chaiken & Maheswaran, 1994), by serving as persuasive arguments (Petty & Cacioppo), by affecting thought confidence (Briñol, Petty, et al., 2004; Tormala et al., in press), and by leading to correction processes (Petty, Wegener, & White, 1998; Wegener & Petty, 1995). Similarly, recipient factors can serve in the same multiple roles as source and message factors. For example, depending on elaboration and other conditions, a person’s mood has been found to influence persuasion by (a) affecting the amount of information processing, (b) biasing the thoughts that are generated, (c) influencing confidence in one’s thoughts, (d) serving as a persuasive argument, or (e) affecting the selection and use of simple cues and heuristics (for a review, see Briñol, Petty, & Barden, 2006; Petty, Fabrigar, & Wegener, 2003). Note that both simple (e.g., source attractiveness) and complex (e.g., 8 item attributes) information can be processed as cues or arguments or serve in other roles depending on the elaboration likelihood. Next, we apply this multiple-roles framework to examine the variable that perhaps has generated more research in the domain of health persuasion than any other: message tailoring (see also Petty, Barden, & Wheeler, 2002). Persuasive effects of message matching One strategy that can increase the effectiveness of a health communication in changing attitudes and behavior consists of altering the arguments contained in the message to match the particular concerns of the message recipient. Matching or tailoring health communications to different aspects of an individual’s personal characteristics can increase its persuasive effect for a number of cancer prevention and detection behaviors, including decreasing dietary fat intake (e.g., Brug, Glanz, Van Assema, Kok, & Van Breukelen, 1998), increasing fruit and vegetable consumption (e.g., Campbell et al., 1994), promoting smoking cessation (Dijkstra, De Vries, Roijackers, & van Breukelen, 1998), and motivating mammography utilization (Rakowski et al., 1998; Skinner, Strecher, & Hospers, 1994). There are a variety of ways in which a message can be matched, including the use of an individualized message (matching at the individual level) and targeted messages (matching at the group level).4 Persuasive communications have been personalized by including recipients’ demographic characteristics (e.g., gender, ethnicity) or other identifying characteristics (e.g., the person’s name or occupation) in an effort to promote various health behaviors (Devos-Comby & Salovey, 2002). Matching procedures can be used to match a large variety of needs, interests, and concerns of a recipient or a group to which the recipient belongs (Kreuter, Farrell, Olevitch, & Brennan, 2000). For example, a communication can match the individual’s stage of behavior change. According to stages of change models (e.g., Prochaska, DiClemente, & Norcross, 1992), individuals pass through five distinct stages (precontemplation, contemplation, preparation, action, and maintenance) on the path to behavioral change. Educational materials that match the individual’s stage of change have been found to be more effective in increasing mammography utilization than standard materials or than not providing materials (e.g., Rakowski et al., 1998). Research conducted on the stage of change model has shown that matching different therapeutic treatments (e.g., self-reevaluation, conditioning, stimulus control) with the specific change stage (e.g., contemplation, preparation) can improve the efficacy of those treatments in a variety of cancer-relevant domains (e.g., Prochaska et al., 1992). Similarly, Herek et al. (1998) conducted an investigation on cultural matching in which it was found that messages matching a participant’s specific ethnic identity were more effective in affecting the evaluation of a HIV-related message than were multicultural messages. Highly individualized messages matching several variables at once have also proven successful (e.g., combining information from questionnaires, medical records, and other sources; Kreuter, Strecher, & Glassman, 1999). Communications have also been matched on the basis of more general individual differences. For example, Brock, Brannon, and Bridgwater (1990) found that matching messages about losing weight to self-schemas of the message recipients (measured with an adjective-rating task in which participants indicated their overall personality types) resulted in more persuasion than presenting the same information without matching to schema set. In other research, Bakker (1999) matched messages to individuals who differed in their need for cognition (Cacioppo & Petty, 1982). This individual difference refers to a person’s propensity to engage in and enjoy effortful cognitive activities. Those high in need for cognition tend to follow the central route to persuasion, whereas those low tend to follow the peripheral route (see Cacioppo, Petty, Feinstein, & Jarvis, 1986, for a review). Bakker found that those low in need for cognition were more persuaded by health-related communications when they were in a format that seemed easy (rather than difficult) to process (matching their processing style). Extending this research, Williams-Piehota, Schneider, Pizarro, Mowad, and Salovey (2003) demonstrated that messages matched to an individuals’ need for cognition were better at motivating mammography 6 months later as compared to mismatched messages. In a similar vein, Epstein used his Cognitive-Experiential Self-Theory (CEST) (see Epstein, 2003, for a review) to match messages to recipients. The CEST argues that there are two independent information-processing systems that operate in parallel (see also Strack & Deutsch, 2004). The experiential system is driven by emotion, is associative, rapid, and primarily nonverbal. Thus, it shares some things in common with the peripheral route. In contrast, the rational system is analytic, logical, and slower in information processing. Thus, it shares some things in common with the central route. Rosental and Epstein (2000; see Epstein, 2003) found matching effects for persuasion in a cancer prevention program. That is, in a study in which a rational message (emphasizing objective information) and an experiential message (including vivid individual cases) in favor of breast self-examination were presented, Rosental and Epstein found more persuasion when the message matched participants’ thinking style, as assessed by the Rational-Experiential Inventory (e.g., Pacini & Epstein, 1999). In this inventory, those high in the rational mode also score high in need for cognition (Cacioppo & Petty, 1982). In addition to these, matching messages to a number of other individual differences has affected persuasion. These variables include sensation seeking (Palmgreen, Stephenson, Everett, Baseheart, & Francies, 2002), optimism (Geers, Handley, & McLarney, 2003), uncertainty orientation (Sorrentino, Bobocel, Gitta, Olson, & Hewitt, 1988), ideal versus ought self-guides (e.g., Evans & Petty, 2003), independent versus interdependent self-construals (Lee, Aaker, & Gardner, 2000), dominance versus submission (Moon, 2002), introversion versus extraversion (Chang, 2002; Wheeler, Petty, & Bizer, 2005), sensitizers versus repressors (DeBono & Snyder, 1992), high versus low consideration of future consequences (Orbell, Perugini, & Rakow, 2004; Strathman, Gleicher, Boninger, & Edwards, 1994), locus of control (Williams-Piehota, Schneider, Pizarro, Mowad, & Salovey, 2004), and monitor–blunter coping styles (e.g., Williams-Piehota, Pizarro, Schneider, Mowad, & Salovey, 2005; for an extensive review of these and other variables, see Briñol & Petty, 2005). Psychological mechanism underlying message matching Although the effectiveness of matching seems to be a well-established phenomenon, most of the researchers did not focus their attention on the psychological mechanisms by which matching leads to persuasion. Understanding such psychological processes is essential to predict the impact of attitude change on further behavioral change. This section outlines the different processes by which matching a message to some characteristic of the recipient can influence attitudes. Perhaps the variable that has been studied most with respect to matching message to person is self-monitoring (Snyder, 1974). This individual difference makes a distinction between high self-monitors, who are oriented toward social approval, and low self-monitors, who are more motivated to be consistent with their internal beliefs and values. Much research on self-monitoring has shown that messages can be made more effective by matching the message to a person’s self-monitoring status. For example, in one study, Snyder and DeBono (1985) exposed high and low self-monitors to advertisements for a variety of products that contained arguments appealing either to the social adjustment function (i.e., describing the social image that consumers could gain from the use of the product) or to the value-expressive function (i.e., presenting content regarding the intrinsic quality or merit of the product). They found that high self-monitors were more influenced by ads with image content than ads with quality content. In contrast, the attitudes of low self-monitors were more vulnerable to messages that made appeals to values or quality (see also DeBono, 1987; Lavine & Snyder, 1996; Snyder & DeBono, 1989). But what are the mechanisms underlying the effectiveness of message matching? This is important to understand because of the strength properties that follow from different processes of persuasion (see Petty & Krosnick, 1995). As noted earlier, research indicates that attitude changes based on high amounts of issue-relevant thought tend to show more persistence over time, resistance, and influence in guiding behavior than changes based on little thought. Unfortunately, much of the available research on matching was not designed to examine underlying mechanisms, so it was not always clear whether matching affected attitudes by serving as a simple cue or an argument, or by affecting the amount or direction of thinking, or influencing confidence in thoughts. The ELM, of course, holds that all of these provide possible mechanisms by which matching can influence attitudes. For example, when thinking is set at a high level (e.g., a topic of high personal interest, high accountability for a decision), then matching could bias the direction of thinking. Indeed, some research suggests that high self-monitors are more motivated to generate favorable thoughts to involving messages that make an appeal to image rather than an appeal to values (e.g., Lavine & Snyder, 1996). In contrast, when the circumstances constrain the likelihood of elaboration to be very low, a match of message to person is more likely to influence attitudes by serving as a simple cue (e.g., DeBono, 1987). That is, even when the content of the message is not processed, if a source simply asserted that the arguments are consistent with a person’s values, a low self-monitor might be more inclined to agree than a high self-monitor by reasoning, “if it links to my values, it must be good.” Furthermore, when thinking is not already constrained by other variables to be high or low, matching a message to a person could increase thinking about the message. This interpretation would be consistent with results obtained by Kreuter et al. (1999) in which participants generated more thoughts in response to messages designed to match the recipients. Also in accord with the interpretation that matching can increase thinking, matched information has been found to be more likely to be read, remembered, and perceived as more relevant than mismatched health communications (Brug et al., 1998; Skinner et al., 1994). Research that has manipulated the quality of the message arguments along with a matching manipulation and shown that matching can increase persuasion when the message is strong but decrease it when it is weak provides especially cogent evidence for the view that matching can affect the extent of thinking. For example, in one study, Petty and Wegener (1998b) matched or mismatched messages that were strong or weak to individuals who differed in their self-monitoring. In this research, high and low self-monitors read image (e.g., how good a product makes you look) or quality (e.g., how efficient a product is) appeals that contained either strong (e.g., beauty or efficacy that last) or weak arguments (e.g., momentary beauty or efficacy). The cogency of the arguments had a larger effect on attitudes when the message matched rather than mismatched the person’s self-monitoring status indicating that matching increased attention to message quality (see also DeBono & Harnish, 1988; Wheeler et al., 2005). In summary, the accumulated research suggests that matching of a message to some characteristics of the recipient can influence attitudes by serving as a peripheral cue when elaboration is low, by biasing thoughts when elaboration is high, and by enhancing the amount of information processing when elaboration is moderate. Additionally, it is worth noting that matching message contents and/or frames with personality types might influence attitude change by other mechanisms under other circumstances. For example, another possibility is that when a message is matched to the person, people might come to accept the message position simply because the message “feels right” (Cesario, Grant, & Higgins, 2004) or is easier to process (e.g., Lee & Aaker, 2004). These simple fluency experiences might influence attitudes under relatively low thinking conditions. Or, the processing fluency and/or the “feeling right” experience might affect persuasion by influencing thought confidence (Cesario et al., 2004; Tormala et al., 2002) if this feeling seems to follow the thought generation. As described for other variables, this metacognitive role would be more likely to occur under relatively high elaboration conditions. Unconscious processes in message matching All the individual differences relevant to attitudes and persuasion described so far (e.g., need for cognition, self-monitoring) have been measured by directly asking people about their self-views. However, just as people can hold conscious, easily reportable self-conceptions, there can be less consciously held aspects of the self-concept as well (McClelland, 1985; Wilson, Lindsey, & Schooler, 2000). Early on, these types of dimensions were assessed with projective tests (Proshansky, 1943) and other indirect measures. More recently, investigators have begun to assess these self-conceptions with more contemporary implicit measures based on reaction times. The most popular of these measures, the Implicit Association Test (Greenwald, McGhee, & Schwartz, 1998) and the Evaluative Priming Task (Fazio, Jackson, Dunton, & Williams, 1995) are designed to assess automatic reactions to various objects, including the self. Automatic aspects of personality, motivations, and attitudes are important because they can influence information processing and behavior in certain contexts independent of more deliberative aspects of one’s self (see Fazio & Olson, 2003; Greenwald & Banaji, 1995). The importance of the distinction between explicit and implicit individual differences is especially apparent when there is a discrepancy between them. Briñol, Petty, and Wheeler (in press) have suggested that such discrepancies can have important consequences for information processing and attitude change. For example, because internal inconsistencies that are explicit are often associated with aversive feelings (e.g., Abelson et al., 1968) and enhanced information processing (e.g., Maio, Bell, & Esses, 1996), individuals with discrepancies between their automatic and deliberative self-conceptions or their more automatic and deliberative evaluations might similarly be (implicitly) motivated to reduce this ambivalence by seeking and processing discrepancy-relevant information (see also Petty, Tormala, Briñol, & Jarvis, 2006). Extensive elaboration of such information might reduce discrepancies by changing either the deliberative or the automatic dimension of the self-concept or attitude object (Briñol, Petty, Horcajo, & Barden, 2006). To examine the consequences of implicit–explicit discrepancies, Briñol, Petty, and Wheeler (in press) conducted a study in which both deliberative and automatic self-dimensions (e.g., self-esteem) were measured. Results showed that as discrepancies between the two increased, participants engaged in more thinking about a persuasive message that was framed as self-relevant. In this research, message processing was assessed by the impact of strong versus weak arguments on attitudes and valenced thoughts related to the proposal about dietary change (i.e., increasing vegetable consumption). These findings suggest that discrepancies between deliberative and automatic self-conceptions are important to understand because such discrepancies can influence attitudes by affecting the extent of information processing. Attitudes–behavior link We have argued that it is important to understand the process by which any variable has influenced a person’s attitude. For example, if matching a message to a person produces persuasion by serving as a simple cue under low-elaboration conditions, the attitude induced will be less predictive of behavior than if matching produced the same amount of persuasion, but worked by increasing positive thoughts to the message arguments under high-elaboration conditions. Thus, predicting behavioral changes depend on understanding of the different processes by which attitude change occurs (see Figure 1). In addition to examining the mechanisms responsible for attitude change, understanding attitude–behavior correspondence depends on a number of other relevant factors (see Fishbein & Cappella, 2006). Several models of the process by which attitudes guide behavior have achieved widespread acceptance, such as Fishbein and Ajzen’s (1975),theory of reasoned action, and Ajzen’s (1991)theory of planned behavior. In contrast to the thoughtful processing highlighted by the theories of reasoned action and planned behavior, Fazio (1990) has proposed that much behavior is rather spontaneous and that attitudes guide behavior by a relatively automatic process. That is, if the relevant attitude comes to mind, consistent behavior is likely to follow. Fazio argued that attitudes can guide behavior without any deliberate reflection or reasoning if (a) the attitude is accessed spontaneously by the mere presence of the attitude object and (b) the attitude colors perception of the object so that if the attitude is favorable (or unfavorable), the qualities of the object appear favorable (or unfavorable). For some cancer communications, attitude change, though an important first step, may still be insufficient to produce the desired behavioral responses even if appropriate new attitudes were formed by the central route. People may also need to rehearse the new attitude sufficiently so that it overcomes and replaces past attitudes (e.g., Petty et al., 2006), or they may need to acquire new skills and self-perceptions of confidence that allow newly acquired attitudes and intentions to be translated into action. Bandura’s (1977, 1986) social cognitive theory provides a framework to understand these processes. Summary and conclusions A fundamental understanding of the basic mechanisms of persuasion can be useful in a wide variety of social contexts. In this article, those mechanisms include persuasion variables that can (a) affect the amount of information processing, (b) bias the thoughts that are generated or (c) one’s confidence in those thoughts (or other structural features), (d) serve as persuasive arguments or evidence, or (e) affect attitudes by serving as simple cues and heuristics. Notably, any given variable, whether part of the source, message, recipient, or context, is capable of serving in these roles. By grouping the persuasion processes into meaningful categories, we aimed to provide a useful guide to organize and facilitate access to key findings in this literature and to maximize the chances of designing effective research and interventions related to cancer prevention. Perhaps the most common finding in the literature on cancer communication has been that matching persuasive messages to people’s characteristics increases persuasion. Consistent with the multiple-roles notion of the ELM, matching messages with personality has been found to influence persuasion by different processes depending on the likelihood of thinking. This review also notes that the same basic human characteristics, such as motives, personality, and attitudes, are often assessed by reliance on what people consciously and deliberately report about their self-concept. However, there might be other more automatically accessible individual differences relevant to attitude change. Matching persuasive messages to automatic aspects of the self-concept and studying the combinatory effects associated with both explicit and implicit individual differences constitute important avenues for future research. Notes 1 " Attitude change simply means that a person’s evaluation is modified from one value to another. There are many ways in which such changes can occur. Although persuasion is often seen as one of the particular ways in which attitudes are changed (i.e., as a consequence of exposing the recipient to some communication delivered by a source), we do not draw a sharp distinction between the terms persuasion and attitude change because the fundamental processes underlying change (with or without a persuasive communication) are the same. 2 " More specifically, elaboration refers to the process by which people add something of their own to the information available. For ease of presentation, however, we will use the terms elaboration, thinking, and information processing similarly. 3 " For expository purposes, we have emphasized the distinction between the central and the peripheral routes to persuasion. That is, we have focused on the prototypical processes at the end points of the elaboration likelihood continuum. In most persuasion situations (which fall somewhere along this continuum), some combinations of central and peripheral processes are likely to have an impact on attitudes. 4 " In the domain of health communication, this matching approach also has been referred to as tailoring (at the individual level), targeting (at the group level), and other similar terms. In this article, we will use the more general term, matching. Although some researchers (e.g., Kreuter & Rimer, 2006) draw a distinction between personalization and targeting, we believe that these types of matching share important underlying conceptual similarities. After reviewing matching effects in a variety of social–psychological domains, Petty, Wheeler, and Bizer (2000) suggested that the most important of these underlying factors has to do with establishing a link to the self. References Abelson , R. P. , Aronson , E., McGuire , W. J., Newcomb , Th. M., Rosenberg , M. J., & Tannenbaum , P. H. ( 1968 ). Theories of cognitive consistency: A sourcebook . Chicago : Rand McNally . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ajzen , I . ( 1991 ). The theory of planned behavior . Organizational Behavior and Human Decision Processes , 50 ( 2 ), 179 – 210 . Google Scholar Crossref Search ADS WorldCat Alba , J. W. , & Marmorstein , H. ( 1987 ). The effects of frequency knowledge on consumer decision making . Journal of Consumer Research , 13 , 411 – 454 . Google Scholar Crossref Search ADS WorldCat Apanovitch , A. M. , McCarthy , D., & Salovey , P. ( 2003 ). Using message framing to motivate HIV testing among low-income, ethnic minority women . Health Psychology , 22 ( 1 ), 60 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat Baker , S. M. , & Petty , R. E. ( 1994 ). Majority and minority influence: Source advocacy as a determinant of message scrutiny . Journal of Personality and Social Psychology , 67 ( 1 ), 5 – 19 . Google Scholar Crossref Search ADS WorldCat Bakker , A. B . ( 1999 ). Persuasive communication about AIDS prevention: Need for cognition determines the impact of message format . AIDS Education and Prevention , 11 ( 2 ), 150 – 162 . Google Scholar PubMed OpenURL Placeholder Text WorldCat Bandura , A . ( 1977 ). Social learning theory . Englewood Cliffs, NJ : Prentice-Hall . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bandura , A . ( 1986 ). Social foundations of thought and action . Englewood Cliffs, NJ : Prentice-Hall . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brehm , J. W . ( 1966 ). A theory of psychological reactance . New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Briñol , P. , & Petty , R. E. ( 2003 ). Overt head movements and persuasion: A self-validation analysis . Journal of Personality and Social Psychology , 84 , 1123 – 1139 . Google Scholar Crossref Search ADS PubMed WorldCat Briñol , P. , & Petty , R. E. ( 2004 ). Self-validation processes: The role of thought confidence in persuasion. In G. Haddock & G. Maio (Eds.), Contemporary perspectives on the psychology of attitudes (pp. 205 – 226 ). Philadelphia : Psychology Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Briñol , P. , & Petty , R. E. ( 2005 ). Individual differences in persuasion. In D. Albarracín, B. T. Johnson, & M. P. Zanna (Eds.), The handbook of attitudes and attitude change (pp. 575 – 616 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Briñol , P. , Petty , R. E., Gallardo , I., & DeMarree , K. G. ( in press ). The role of self-affirmation in consumer persuasion . Advances in Consumer Research . OpenURL Placeholder Text WorldCat Briñol , P. , Petty , R. E., Horcajo , J., & Barden , J. ( 2006 ). The role of elaboration in the malleability of implicit attitudes . Manuscript submitted for publication. Briñol , P. , Petty , R. E., & Tormala , Z. L. ( 2004 ). The self-validation of cognitive responses to advertisements . Journal of Consumer Research , 31 , 559 – 573 . Google Scholar Crossref Search ADS WorldCat Briñol , P. , Petty , R. E., & Wheeler , S. C. ( in press ). Discrepancies between explicit and implicit self-concepts: Consequences for information processing . Journal of Personality and Social Psychology . OpenURL Placeholder Text WorldCat Brock , T. C. , Brannon , L. A., & Bridgwater , C. ( 1990 ). Message effectiveness can be increased by matching appeals to recipients’ self-schemas: Laboratory demonstrations and a national field experiment. In S. J. Agres & J. A. Edell (Eds.), Emotion in advertising: Theoretical and practical explorations (pp. 285 – 315 ). Westport, CT : Quorum Books . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brug , J. , Glanz , K., Van Assema , P., Kok , G., & Van Breukelen , G. J. P. ( 1998 ). The impact of computer-tailored feedback and iterative feedback on fat, fruit, and vegetable intake . Health Education and Behavior , 25 , 517 – 531 . Google Scholar Crossref Search ADS PubMed WorldCat Burnkrant , R. , & Unnava , R. ( 1989 ). Self-referencing: A strategy for increasing processing of message content . Personality and Social Psychology Bulletin , 15 , 628 – 638 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , Marshall-Goodell , B. S., Tassinary , L. G., & Petty , R. E. ( 1992 ). Rudimentary determinants of attitudes: Classical conditioning is more effective when prior knowledge about the attitude stimulus is low than high . Journal of Experimental Social Psychology , 28 , 207 – 233 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , & Petty , R. E. ( 1982 ). The need for cognition . Journal of Personality and Social Psychology , 42 ( 1 ), 116 – 131 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , & Petty , R. E. ( 1989 ). Effects of message repetition on argument processing, recall, and persuasion . Basic and Applied Social Psychology , 10 ( 1 ), 3 – 12 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , Petty , R. E., Feinstein , J., & Jarvis , W. B. G. ( 1996 ). Dispositional differences in cognitive motivation: The life and times of individuals varying in need for cognition . Psychological Bulletin , 119 ( 2 ), 197 – 253 . Google Scholar Crossref Search ADS WorldCat Campbell , M. K. , DeVellis , B. M., Strecher , V. J., Ammerman , A. S., DeVellis , R. R., & Sandler , R. S. ( 1994 ). Improving dietary behavior: The effectiveness of tailored messages in primary care settings . American Journal of Public Health , 84 , 783 – 787 . Google Scholar Crossref Search ADS PubMed WorldCat Cesario , J. , Grant , H., & Higgins , E. T. ( 2004 ). Regulatory fit and persuasion: Transfer from “feeling right.” Journal of Personality and Social Psychology , 86 , 388 – 404 . Google Scholar Crossref Search ADS PubMed WorldCat Chaiken , S . ( 1980 ). Heuristic versus systematic information processing and the use of source versus message cues in persuasion . Journal of Personality and Social Psychology , 39 , 752 – 756 . Google Scholar Crossref Search ADS WorldCat Chaiken , S . ( 1987 ). The heuristic model of persuasion. In M. P. Zanna, J. Olson, & C. P. Herman (Eds.), Social influence: The Ontario symposium (Vol. 5 , pp. 3 – 39 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chaiken , S. , & Eagly , A. H. ( 1976 ). Communication modality as a determinant of message persuasiveness and message comprehensibility . Journal of Personality and Social Psychology , 34 , 605 – 614 . Google Scholar Crossref Search ADS WorldCat Chaiken , S. , & Maheswaran , D. ( 1994 ). Heuristic processing can bias systematic processing: Effects of source credibility, argument ambiguity, and task importance on attitude judgment . Journal of Personality and Social Psychology , 66 , 460 – 473 . Google Scholar Crossref Search ADS PubMed WorldCat Chang , C . ( 2002 ). Self-congruency as a cue in different advertising-processing contexts . Communication Research , 29 , 503 – 536 . Google Scholar Crossref Search ADS WorldCat DeBono , K. G . ( 1987 ). Investigating the social-adjustive and value-expressive functions of attitudes: Implications for persuasion processes . Journal of Personality and Social Psychology , 52 , 279 – 287 . Google Scholar Crossref Search ADS WorldCat DeBono , K. G. , & Harnish , R. J. ( 1988 ). Source expertise, source attractiveness, and processing or persuasive information: A functional approach . Journal of Personality and Social Psychology , 55 , 541 – 546 . Google Scholar Crossref Search ADS WorldCat DeBono , K. G. , & Snyder , A. ( 1992 ). Repressors, sensitizers, source expertise, and persuasion . Social Behavior and Personality , 20 , 263 – 272 . Google Scholar Crossref Search ADS WorldCat Devos-Comby , L. , & Salovey , P. ( 2002 ). Applying persuasion strategies to alter HIV-relevant thoughts and behaviors . Review of General Psychology , 6 , 287 – 304 . Google Scholar Crossref Search ADS WorldCat Dijkstra , S. , De Vries , J., Roijackers , J., & Van Breukelen , G. ( 1998 ). Tailored interventions to communicate stage-matched information to smokers in different motivational stages . Journal of Consulting and Clinical Psychology , 66 , 549 – 557 . Google Scholar Crossref Search ADS PubMed WorldCat Dunning , D. , Heath , C., & Suls , J. M. ( 2004 ). Flawed self-assessment: Implications for health, education, and the workplace . Psychological Science in the Public Interest , 5 ( 3 ), 69 – 106 . Google Scholar Crossref Search ADS PubMed WorldCat Eagly A. H. , & Chaiken , S. ( 1993 ). The psychology of attitudes . Fort Worth, TX : Harcourt, Brace, Jovanovich . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Epstein , S . ( 2003 ). Cognitive-experiential self-theory of personality. In T. Millon & M. J. Lerner (Eds.), Handbook of psychology (Vol. 5 , pp. 159 – 184 ). Hoboken, NJ : John Wiley & Sons . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Evans , L. M. , & Petty , R. E. ( 2003 ). Self-guide framing and persuasion: Responsibly increasing message processing to ideal levels . Personality and Social Psychology Bulletin , 29 , 313 – 324 . Google Scholar Crossref Search ADS PubMed WorldCat Fazio , R. H . ( 1990 ). Multiple processes by which attitudes guide behavior: The MODE model as an integrative framework. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 23 , pp. 75 – 109 ). New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fazio , R. H. , Jackson , J. R., Dunton , B. C., & Williams , C. J. ( 1995 ). Variability in automatic activation as an unobtrusive measure of racial attitudes: A bona fide pipeline? Journal of Personality and Social Psychology , 69 , 1013 – 1027 . Google Scholar Crossref Search ADS PubMed WorldCat Fazio , R. H. , & Olson , M. ( 2003 ). Implicit measures in social cognition research: Their meaning and uses . Annual Review of Psychology , 54 , 297 – 327 . Google Scholar Crossref Search ADS PubMed WorldCat Fazio , R. H. , & Williams , C. J. ( 1986 ). Attitude accessibility as a moderator of the attitude-perception and attitude-behavior relations: An investigation of the 1984 presidential election . Journal of Personality and Social Psychology , 51 , 505 – 514 . Google Scholar Crossref Search ADS PubMed WorldCat Fishbein , M. , & Ajzen , I. ( 1975 ). Belief, attitude, intention, and behavior: An introduction to theory and research . Menlo Park, CA : Addison-Wesley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fishbein , M. , & Cappella , J. N. ( 2006 ). The role of theory in developing effective health communications . Journal of Communication , 56 , S1 – S17 . Google Scholar Crossref Search ADS WorldCat Fleming , M. A. , & Petty , R. E. ( 2000 ). Identity and persuasion: An elaboration likelihood approach. In M. A. Hogg & D. J. Terry (Eds.), Attitudes, behavior, and social context: The role of norms and group membership (pp. 171 – 199 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Friedrich , J. , Fetherstonhaugh , D., Casey , S., & Gallagher , D. ( 1996 ). Argument integration and attitude change: Suppression effects in the integration of one-sided arguments that vary in persuasiveness . Personality and Social Psychology Bulletin , 22 ( 2 ), 179 – 191 . Google Scholar Crossref Search ADS WorldCat Geers , A. L. , Handley , I. M., & McLarney , A. R. ( 2003 ). Discerning the role of optimism in persuasion: The valence-enhancement hypothesis . Journal of Personality and Social Psychology , 85 , 554 – 565 . Google Scholar Crossref Search ADS PubMed WorldCat Greenwald , A. G. , & Banaji , M. ( 1995 ). Implicit social cognition: Attitudes, self-esteem, and stereotypes . Psychological Review , 102 ( 1 ), 4 – 27 . Google Scholar Crossref Search ADS PubMed WorldCat Greenwald , A. G. , McGhee , D. E., & Schwartz , J. L. K. ( 1998 ). Measuring individual differences in implicit cognition: The implicit association task . Journal of Personality and Social Psychology , 74 , 1464 – 1480 . Google Scholar Crossref Search ADS PubMed WorldCat Harkins , S. G. , & Petty , R. E. ( 1981 ). The effects of source magnification cognitive effort on attitudes: An information processing view . Journal of Personality and Social Psychology , 40 , 401 – 413 . Google Scholar Crossref Search ADS WorldCat Heesacker , M. , Petty , R. E., & Cacioppo , J. T. ( 1983 ). Field dependence and attitude change: Source credibility can alter persuasion by affecting message-relevant thinking . Journal of Personality , 51 , 653 – 666 . Google Scholar Crossref Search ADS WorldCat Helweg-Larse , M. , & Collins , B. E. ( 1997 ). A social psychological perspective on the role of knowledge about AIDS and AIDS prevention . Current Directions in Psychological Science , 6 ( 1 ), 23 – 26 . Google Scholar Crossref Search ADS WorldCat Herek , G. M. , Gillis , J. R., Glunt , E. K., Lewis , J., Welton , D., & Capitanio , J. P. ( 1998 ). Culturally sensitive AIDS educational videos for African American audiences: Effects of source, message, receiver, and context . American Journal of Community Psychology , 26 , 705 – 743 . Google Scholar Crossref Search ADS PubMed WorldCat Howard , D. J . ( 1990 ). Rhetorical question effects on message processing and persuasion: The role of information availability and the elicitation of judgment . Journal of Experimental Social Psychology , 26 , 217 – 239 . Google Scholar Crossref Search ADS WorldCat Kreuter , M. W. , Farrell , D., Olevitch , L., & Brennan , L. ( 2000 ). What is tailored communication? In J. Bryant & D. Zillmann (Eds.), Tailoring health messages: Customizing communication with computer technology (pp. 1 – 23 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kreuter , M. W. , Strecher , V. J., & Glassman , B. ( 1999 ). One size does not fit all: The case for tailoring print materials . Annals of Behavioral Medicine , 21 , 276 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Lavine , H. , & Snyder , M. ( 1996 ). Cognitive processing and the functional matching effect in persuasion. The mediating role of subjective perceptions of message quality . Journal of Experimental Social Psychology , 32 , 580 – 604 . Google Scholar Crossref Search ADS PubMed WorldCat Lee , A. Y. , & Aaker , J. L. ( 2004 ). Bringing the frame into focus: The influence of regulatory fit on processing fluency and persuasion . Journal of Personality and Social Psychology , 86 ( 2 ), 205 – 218 . Google Scholar Crossref Search ADS PubMed WorldCat Lee , A. Y. , Aaker , J. L., & Gardner , W. L. ( 2000 ). The pleasures and pains of distinct self-construals: The role of interdependence in regulatory focus . Journal of Personality and Social Psychology , 78 , 1122 – 1134 . Google Scholar Crossref Search ADS PubMed WorldCat Leventhal , H. , & Cleary , P. D. ( 1980 ). The smoking problem: A review of the research and theory in behavioral risk modifications . Psychological Bulleting , 88 , 370 – 405 . Google Scholar Crossref Search ADS WorldCat Maio , G. R. , Bell , D. E., & Esses , V. M. ( 1996 ). Ambivalence and persuasion: The processing of messages about immigrant groups . Journal of Experimental Social Psychology , 32 , 513 – 536 . Google Scholar Crossref Search ADS PubMed WorldCat McClelland , D. C . ( 1985 ). How motives, skills, and values determine what people do . American Psychologist , 40 , 812 – 825 . Google Scholar Crossref Search ADS WorldCat Moon , Y . ( 2002 ). Personalization and personality: Some effects of customizing message style based on consumer personality . Journal of Consumer Psychology , 12 , 313 – 326 . Google Scholar Crossref Search ADS WorldCat Moore , D. L. , Hausknecht , D., & Thamodaran , K. ( 1986 ). Time pressure, response opportunity, and persuasion . Journal of Consumer Research , 13 ( 1 ), 85 – 99 . Google Scholar Crossref Search ADS WorldCat Moore , D. L. , & Reardon , R. ( 1987 ). Source magnification: The role of multiple sources in processing of advertising appeals . Journal of Marketing Research , 24 , 412 – 417 . Google Scholar Crossref Search ADS WorldCat Orbell , S. Perugini , M., & Rakow , T. ( 2004 ). Individual differences in sensitivity to health communications: Consideration of future consequences . Heath Psychology , 23 , 388 – 396 . Google Scholar Crossref Search ADS WorldCat Pacini , R. , & Epstein , S. ( 1999 ). The relation of rational and experiential information processing styles to personality, basic beliefs, and the ration-bias phenomenon . Journal of Personality and Social Psychology , 76 , 972 – 987 . Google Scholar Crossref Search ADS PubMed WorldCat Palmgreen , P. , Stephenson , M. T., Everett , M. W., Baseheart , J. R., & Francies , R. ( 2002 ). Perceived message sensation value (PMSV) and the dimensions and validation of a PMSV scale . Health Communication , 14 , 403 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R. E. , Baker , S., & Gleicher , F. ( 1991 ). Attitudes and drug abuse prevention: Implications of the Elaboration Likelihood Model of persuasion. In L. Donohew, H. E. Sypher, & W. J. Bukoski (Eds.), Persuasive communication and drug abuse prevention (pp. 71 – 90 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Barden , J., & Wheeler , S. C. ( 2002 ). The elaboration likelihood model of persuasion. In R. J. DiClemente, R. A. Crosby, & M. Kegler (Eds.), Emerging theories in health promotion practice and research (pp. 71 – 99 ). San Francisco : Jossey-Bass . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Briñol , P. ( 2002 ). Attitude change: The Elaboration Likelihood Model of persuasion. In G. Bartels & W. Nielissen (Eds.), Marketing for sustainability: Towards transactional policy making (pp. 176 – 190 ). Amsterdam : IOS press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Briñol , P. ( in press ). A meta-cognitive approach to “implicit” and “explicit” evaluations: Comment on Gawronski and Bodenhausen ( 2006 ). Psychological Bulletin . OpenURL Placeholder Text WorldCat Petty , R. E. , Briñol , P., & Tormala , Z. L. ( 2002 ). Thought confidence as a determinant of persuasion: The self-validation hypothesis . Journal of Personality and Social Psychology , 82 , 722 – 741 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R. E. , Briñol , P., Tormala , Z. L., & Wegener , D. T. ( in press ). The role of meta-cognition in social psychology. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: A handbook of basic principles (2nd ed.). New York : Guilford Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Cacioppo , J. T. ( 1979 ). Effects of forewarning of persuasive intent on cognitive responses and persuasion . Personality and Social Psychology Bulletin , 5 ( 2 ), 173 – 176 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1981 ). Attitudes and persuasion: Classics and contemporary approaches . Dubuque, IA : Win. C. Brown . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Cacioppo , J. T. ( 1984a ). The effects of involvement on responses to argument quantity and quality: Central and peripheral routes to persuasion . Journal of Personality and Social Psychology , 46 ( 1 ), 69 – 81 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1984b ). Source factors and the elaboration likelihood model of persuasion . Advances in Consumer Research , 11 , 668 – 672 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1986 ). Communication and persuasion: Central and peripheral routes to attitude change . New York : Springer-Verlag . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Cacioppo , J. T. ( 1990 ). Involvement and persuasion: Tradition versus integration . Psychological Bulletin , 107 , 367 – 374 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Cacioppo , J. T., & Goldman , R. ( 1981 ). Personal involvement as a determinant of argument-based persuasion . Journal of Personality and Social Psychology , 41 , 847 – 855 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Cacioppo , J. T., & Heesacker , M. ( 1981 ). The use of rhetorical questions in persuasion: A cognitive response analysis . Journal of Personality and Social Psychology , 40 , 432 – 440 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Fabrigar , L. R., & Wegener , D. T. ( 2003 ). Emotional factors in attitudes and persuasion. In R. J. Davidson, K. R. Scherer, & H. H. Goldsmith (Eds.), Handbook of affective sciences (pp. 752 – 772 ). Oxford, UK : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Haugtvedt , C., & Smith , S. M. ( 1995 ). Elaboration as a determinant of attitude strength: Creating attitudes that are persistent, resistant, and predictive of behavior. In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (pp. 93 – 130 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Krosnick , J. A. (Eds.). ( 1995 ). Attitude strength: Antecedents and consequences . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Ostrom , T. M. & Brock , T. C. ( 1981 ). Cognitive responses in persuasion . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Priester , J. R., & Briñol , P. ( 2002 ). Mass media attitude change: Implications of the elaboration likelihood model of persuasion. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (2nd ed., pp. 155 – 199 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Schumann , D. W., Richman , S. A., & Strathman , A. J. ( 1993 ). Positive mood and persuasion: Different roles for affect under high and low elaboration conditions . Journal of Personality and Social Psychology , 64 ( 1 ), 5 – 20 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Tormala , Z. L., Briñol , P., & Jarvis , W. B. G. ( 2006 ). Implicit ambivalence from attitude change: An exploration of the PAST model . Journal of Personality and Social Psychology , 90 ( 1 ), 21 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R. E. , & Wegener , D. T. ( 1993 ). Flexible correction processes in social judgment: Correcting for context induced contrast . Journal of Experimental Social Psychology , 29 ( 2 ), 137 – 165 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Wegener , D. T. ( 1998a ). Attitude change: Multiple roles for persuasion variables. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed., Vol. 1 , pp. 323 – 390 ). New York : McGraw-Hill . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Wegener , D. T. ( 1998b ). Matching versus mismatching attitude functions: Implications for scrutiny of persuasive messages . Personality and Social Psychology Bulletin , 24 , 227 – 240 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Wegener , D. T. ( 1999 ). The Elaboration Likelihood Model: Current status and controversies. In S. Chaiken & Y. Trope (Eds.), Dual process theories in social psychology (pp. 41 – 72 ). New York : Guilford Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Wegener , D. T., & White , P. ( 1998 ). Flexible correction processes in social judgment: Implications for persuasion . Social Cognition , 16 ( 1 ), 93 – 113 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Wells , G. L., & Brock , T. C. ( 1976 ). Distraction can enhance or reduce yielding to propaganda: Thought disruption versus effort justification . Journal of Personality and Social Psychology , 34 , 874 – 884 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , Wheeler , S. C., & Bizer , G. ( 2000 ). Matching effects in persuasion: An elaboration likelihood analysis. In G. Maio & J. Olson (Eds.), Why we evaluate: Functions of attitudes (pp. 133 – 162 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Prochaska , J. O. , DiClemente , C. C., & Norcross , J. C. ( 1992 ). In search of how people change: Applications to addictive behavior . American Psychologist , 47 , 1102 – 1114 . Google Scholar Crossref Search ADS PubMed WorldCat Proshansky , H. M . ( 1943 ). A projective method for the study of attitudes . Journal of Abnormal and Social Psychology , 38 , 393 – 395 . Google Scholar Crossref Search ADS WorldCat Puckett , J. , Petty , R. E., Cacioppo , J. T., & Fisher , D. ( 1983 ). The relative impact of age and attractiveness stereotypes on persuasion . Journal of Gerontology , 38 , 340 – 343 . Google Scholar Crossref Search ADS PubMed WorldCat Rakowski , W. , Ehrich , B., Goldstein , M. G., Rimer , B. K., Pearlman , D. N., Clark , M. A., et al. . 1998 ). Increasing mammography among women aged 40–74 by use of stage-matched, tailored intervention . Preventive Medicine , 27 , 748 – 756 . Google Scholar Crossref Search ADS PubMed WorldCat Rimer , B. K. , & Kreuter , M. W. ( 2006 ). Advancing tailored health communication: A persuasion and message effects perspective . Journal of Communication , 56 , S184 – S201 . Google Scholar Crossref Search ADS WorldCat Rosenthal , L. , & Epstein , S. ( 2000 ). Rational and experiential thinking styles as related to receptivity to messages syntonic and dystonic with thinking style . Unpublished manuscript, University of Massachusetts at Amherst . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rothman , A. J . ( 2000 ). Toward a theory-based analysis of behavioral maintenance . Health Psychology , 19 ( Suppl. 1 ), 64 – 69 . Google Scholar Crossref Search ADS PubMed WorldCat Rothman , A. J. , & Schwarz , N. ( 1998 ). Constructing perceptions of vulnerability: Personal relevance and the use of experiential information in health judgments . Personality and Social Psychology Bulletin , 24 , 1053 – 1064 . Google Scholar Crossref Search ADS WorldCat Rothman , A. J. , Salovey , P., Antone , C., Keough , K., & Martin , C. D. ( 1993 ). The influence of message framing on intentions to perform health behaviors . Journal of Experimental Social Psychology , 29 , 408 – 433 . Google Scholar Crossref Search ADS WorldCat Rothman , A. J. , Salovey , P., Turvey , C., & Fishkins , S. A. ( 1993 ). Attributions of responsibility and persuasion: Increasing mammography utilization among women over 40 with an internally oriented message . Health Psychology , 12 ( 1 ), 39 – 47 . Google Scholar Crossref Search ADS PubMed WorldCat Salovey , P. , & Rothman , A. J. ( 2003 ). Social psychology of health: Key readings . New York : Psychology Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sherman , D. A. K. , Nelson , L. D., & Steele , C. M. ( 2000 ). Do messages about health risk threaten the self? Increasing the acceptance of threatening health messages via self-affirmation . Personality and Social Psychology Bulletin , 26 , 1046 – 1058 . Google Scholar Crossref Search ADS WorldCat Skinner , C. S. , Strecher , V. J., & Hospers , H. ( 1994 ). Physicians recommendations for mammography: Do tailored messages make a difference? American Journal of Public Health , 84 ( 1 ), 43 – 49 . Google Scholar Crossref Search ADS PubMed WorldCat Smith , S. M. , & Petty , R. E. ( 1996 ). Message framing and persuasion: A message processing analysis . Personality and Social Psychology Bulletin , 22 , 257 – 268 . Google Scholar Crossref Search ADS WorldCat Smith , S. M. , & Shaffer , D. R. ( 1991 ). Celebrity and cajolery: Rapid speech may promote or inhibit persuasion via its impact on message elaboration . Personality and Social Psychology Bulletin , 17 , 663 – 669 . Google Scholar Crossref Search ADS WorldCat Snyder , M . ( 1974 ). Self-monitoring of expressive behavior . Journal of Personality and Social Psychology , 30 , 526 – 537 . Google Scholar Crossref Search ADS WorldCat Snyder , M. , & DeBono , K. G. ( 1985 ). Appeals to image and claims about quality: Understanding the psychology of advertising . Journal of Personality and Social Psychology , 49 , 586 – 597 . Google Scholar Crossref Search ADS WorldCat Snyder , M. , & DeBono , K. G. ( 1989 ). Understanding the functions of attitudes: Lessons from personality and social behavior. In A. Pratkanis, S. Breckler, & A. Greenwald (Eds.), Attitude structure and function (pp. 339 – 359 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sorrentino , R. M. , Bobocel , D. R., Gitta , M. Z., Olson , J. M., & Hewitt , E. C. ( 1988 ). Uncertainty orientation and persuasion: Individual differences in the effects of personal relevance on social judgments . Journal of Personality and Social Psychology , 55 , 357 – 371 . Google Scholar Crossref Search ADS WorldCat Staats , A. W. , & Staats , C. ( 1958 ). Attitudes established by classical conditioning . Journal of Abnormal and Social Psychology , 67 ( 1 ), 159 – 167 . OpenURL Placeholder Text WorldCat Stiff , J. B . ( 1986 ). Cognitive processing of persuasive message cues: A meta-analytic review of the effects of supporting information on attitudes . Communication Monographs , 53 ( 1 ), 75 – 89 . Google Scholar Crossref Search ADS WorldCat Strack , F. , & Deutsch , R. ( 2004 ). Reflective and impulsive determinants of social behavior . Personality and Social Psychology Review , 8 , 220 – 247 . Google Scholar Crossref Search ADS PubMed WorldCat Strathman , D. , Gleicher , F., Boninger , D. S., & Edwards , C. S. ( 1994 ). The consideration of future consequences: Weighing outcomes of behavior . Journal of Personality and Social Psychology , 66 , 742 – 752 . Google Scholar Crossref Search ADS WorldCat Swasy , J. L. , & Munch , J. M. ( 1985 ). Examining the target of receiver elaborations: Rhetorical question effects on source processing and persuasion . Journal of Consumer Research , 11 , 877 – 886 . Google Scholar Crossref Search ADS WorldCat Tormala , Z. L. , Briñol , P., & Petty , R. E. ( in press ). When credibility attacks: The reverse impact of source credibility on persuasion . Journal of Experimental Social Psychology . OpenURL Placeholder Text WorldCat Tormala , Z. L. , Petty , R. E., & Briñol , P. ( 2002 ). Ease of retrieval effects in persuasion: The roles of elaboration and thought-confidence . Personality and Social Psychology Bulletin , 28 , 1700 – 1712 . Google Scholar Crossref Search ADS WorldCat Wegener , D. T. , & Petty , R. E. ( 1995 ). Flexible correction processes in social judgment: The role of naive theories in corrections for perceived bias . Journal of Personality and Social Psychology , 68 ( 1 ), 36 – 51 . Google Scholar Crossref Search ADS PubMed WorldCat Wegener , D. T. , & Petty , R. E. ( 1997 ). The flexible correction model: The role of naïve theories in bias correction. In M. P. Zanna (Ed.), Advances in experimental social psychology (Vol. 29 , pp. 141 – 208 ). San Diego, CA : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Weinstein , N. D . ( 2003 ). Exploring the links between risk perceptions and preventive health behavior. In J. Suls & K. Wallston (Eds.), Social psychological foundations of health and illness (pp. 22 – 53 ). Malden, MA : Blackwell . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Wheeler , S. C. , Petty , R. E., & Bizer , G. Y. ( 2005 ). Self-schema matching and attitude change: Situational and dispositional determinants of message elaboration . Journal of Consumer Research , 31 , 787 – 797 . Google Scholar Crossref Search ADS WorldCat Williams-Piehota , P. , Pizarro , J., Schneider , T. R., Mowad , L., & Salovey , P. ( 2005 ). Matching health messages to monitor-blunter coping styles to motivate screening mammography . Health Psychology , 24 ( 1 ), 58 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat Williams-Piehota , P. , Schneider , T. R., Pizarro , J., Mowad , L., & Salovey , P. ( 2003 ). Matching health messages to information-processing styles: Need for cognition and mammography utilization . Health Communication , 15 , 375 – 392 . Google Scholar Crossref Search ADS PubMed WorldCat Williams-Piehota , P. , Schneider , T. R., Pizarro , J., Mowad , L., & Salovey , P. ( 2004 ). Matching health messages to health locus of control beliefs for promoting mammography utilization . Psychology and Health , 19 , 407 – 423 . Google Scholar Crossref Search ADS WorldCat Wilson , T. D. , & Brekke , N. ( 1994 ). Mental contamination and mental correction: Unwanted influences on judgments and evaluations . Psychological Bulletin , 116 ( 1 ), 117 – 142 . Google Scholar Crossref Search ADS PubMed WorldCat Wilson , T. D. , Lindsey , S., & Schooler , T. Y. ( 2000 ). A model of dual attitudes . Psychological Review , 107 ( 1 ), 101 – 126 . Google Scholar Crossref Search ADS PubMed WorldCat Wood , W . ( 1982 ). Retrieval of attitude relevant information from memory: Effects on susceptibility to persuasion and on intrinsic motivation . Journal of Personality and Social Psychology , 42 , 798 – 810 . Google Scholar Crossref Search ADS WorldCat Wright , P. L . ( 1981 ). Cognitive responses to mass media advocacy. In R. E. Petty, T. M. Ostrom, & T. C. Brock (Eds.), Cognitive responses in persuasion (pp. 263 – 282 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association

Lang,, Annie

doi: 10.1111/j.1460-2466.2006.00283.xpmid: N/A

Abstract This paper applies the limited capacity model of motivated mediated messages (LC4MP) to the problem of creating effective messages about cancer. A general description of the model is presented and then applied specifically to the task of creating effective cancer communication messages by asking the following questions about cancer communication: (a) What is the goal of the message? (b) Who is in the target market? (c) What medium will carry the message? and (d) What is the motivational and personal relevance of the main information in the message for the majority of people in the target market? The paper concludes that cancer is a motivationally relevant topic that will elicit aversive activation. Target markets for various types of cancer-related messages (e.g., smokers or people of a certain age) will process mediated messages in predictably different ways making certain design decisions better for certain target markets. Both structural and content elements of messages interact with the limited capacity information processing system to impact resource allocation, which in turn determines how well messages are encoded, stored, and retrieved at a decision point. Individual differences in peoples’ motivational activation influence both their tendencies to engage in risky behaviors that increase the probabilities of getting cancer and their processing of health-related messages. Future research from this perspective should be done to optimize cancer messages for specific target audiences using specific media. Why do we study health communication specifically? Is health communication different from other types of communication? Is health communication about cancer different from health communication about other topics? Obviously, we study health communication because we want to be able to deliver effective messages about health-related information and behaviors to people who would benefit from those messages. Health communication shares a great deal with other forms of goal-directed communication (e.g., advertising, political communication, public relations, and educational messages), and we should take advantage of what is known to construct health-related messages that achieve specific goals for designated target audiences. The goals of the message may range from awareness to reinforcement to knowledge gain to persuasion to behavior change. The target market may range from every member of the public, young and old, one to a small well-defined group of people. In many ways, communication about cancer and cancer control is a restricted subset of health communication. Communications about cancer deal with a topic that is scary, negative, and may impact, in some way, virtually everyone. Communicating about cancer includes all of the goals extant in other forms of health communication: awareness (e.g., of new treatments and findings), reinforcement of behaviors that decrease cancer risk (e.g., NOT smoking), knowledge gain (e.g., creating messages that give people information about cancer, medication, treatment, etc.), persuasion (e.g., changing people’s minds about the health risks of common and uncommon behaviors such as getting a tan, drinking a glass of wine, or having yearly mammograms), and behavior change (e.g., getting screened, stopping smoking, eating more vegetables). Communication about cancer may be highly targeted (e.g., teens, women over 55, sufferers of a specific condition) or it may be aimed at the entire population (e.g., messages about healthy eating and lifestyles or the danger of too much sun). In other words, communication about health in general and cancer in particular is much like any other sort of goal-directed communication. You must determine the audience for your message, the goal of your message, what medium can be used to reach that audience, and then you must build an effective message. This paper is about this final step. How do you build effective messages about cancer? The theory presented here (the limited capacity model of motivated mediated message processing [LC4MP]; A. Lang, 2000, 2005) is a theory about how individuals process mediated messages. The theory is meant to be applicable to all contents, all media, and all goals. The medium, the content of the message, and the goal of the message are variables within the theory. Different media, contents, and goals will lead to different patterns of motivational and cognitive responses in viewers that, interacting with the structure and content of the messages and the individual differences of the media user, determine a great deal about how a message is processed, including which parts of the message are attended to, encoded, and stored and how the message is evaluated and liked. The goal of this paper is to apply LC4MP specifically to the problem of creating effective messages about cancer. To do so, this paper will introduce LC4MP in general and review and apply the findings of research done within the LC4MP framework to this question. The LC4MP This model has five major assumptions—the first about the nature of cognition, the second about the nature of motivation, the third about the nature of media, the fourth about the nature of time, and the fifth about the nature of communication. First, people are assumed to be limited capacity information processors (Basil, 1994; Schneider, Dumais, & Shiffrin, 1984; Shiffrin & Schneider, 1977). They have only a limited number of cognitive resources to expend on the tasks of perceiving, encoding, understanding, and remembering the world they live in. When there are insufficient resources available, processing suffers. Second, people have two underlying motivational systems, the appetitive (or approach) system and the aversive (or avoidance) system (Bradley, 1994; Cacioppo & Gardner, 1999; P. J. Lang, Bradley, & Cuthbert, 1997). These systems activate automatically in response to motivationally relevant stimuli in the environment and influence ongoing cognitive processing. Third, media are made up of variably redundant streams of information presented through multiple sensory channels (eyes, ears, touch) and formats (words, text, still pictures, moving pictures, etc.; Reeves, Thorson, & Schleuder, 1986; Reeves et al., 1985; Thorson, Reeves, & Schleuder, 1986). Fourth, all human behavior occurs over time and is constantly changing from one second to the next. Human behavior, and therefore human cognition, is a dynamic process (Thelen & Smith, 1994). Fifth, communication is the overtime interaction between the human motivated information processing system and the communication message (S. Geiger & Reeves, 1993; A. Lang, 2000; Rafaeli, 1988). This interaction is continuous and truly interactive. Aspects of the message influence the motivational and cognitive systems and aspects of those systems influence how the message is perceived, encoded, stored, and eventually retrieved. In other words, communication is a continuous, interactive, dynamic, embodied process all of which must be taken into account when designing effective messages about anything—including cancer. According to LC4MP, processing messages involves three major subprocesses: encoding, storage, and retrieval. These subprocesses occur constantly, continuously, and simultaneously. Encoding is the act of creating a mental representation of a stimulus. It is the process of selecting information from the environment for further processing. Encoding is not a veridical process. People do not make exact copies of the world in their heads. Rather, they automatically (and unconsciously) select the important aspects of a message and encode them. Information that is not encoded is lost. A major area of importance in LC4MP is understanding how to design messages to insure that important parts of a message are encoded. Processing resources must be allocated to a piece of information in order for it to be encoded. Resources can be allocated as a result of automatic or controlled processing mechanisms (A. Lang, Potter, & Bolls, 1999; Schneider, Dumais, & Shiffrin, 1984; Wickens, 1984). Controlled allocation is related to a person’s ongoing goals and interests. On the other hand, many aspects of the environment automatically elicit the allocation of resources. One way this is done is through the elicitation of an orienting response. An orienting response is an automatic attention response, sometimes called the “What is it?” response, which is elicited by novel and signal stimuli in the environment (Graham, 1979). Novel stimuli are things that represent a change in the environment. They are things that are new to this particular environment—not new in general. Therefore, a person walking into a room, a sudden noise, or a sudden change in the light level would all elicit an orienting response. Orienting responses are also elicited by signal stimuli, that is, stimuli that a person has learned signal important information. These include things such as your name. Novel stimuli are the same for all people; signal stimuli differ for different individuals. Another type of stimulus to which resources are automatically allocated is motivationally relevant stimuli (A. Lang, Bradley, & Sparks, 2004; A. Lang, Newhagen, & Reeves, 1996; A. Lang, Sparks, Bradley, Lee, & Wang, 2004; P. J. Lang, Bradley, & Cuthbert, 1990). These are stimuli that are related to survival. The appetitive motivational system evolved to help the organism get food and mates in order to ensure the survival of both the individual and the species. The aversive motivational system evolved to protect the individual from danger. Primary motivational stimuli include sex, food, and danger. Like novel stimuli, primary motivational stimuli are the same for everyone. However, individuals also learn that certain stimuli signal positive or negative consequences; therefore, stimuli can take on motivational relevance through learning (like Pavlov’s dogs salivating to the dinner bell), and these stimuli will often vary from individual to individual. In sum, elements of the environment or of a media message are encoded when limited cognitive resources are allocated to them. Resources can be allocated as a result of controlled or automatic mechanisms. Stimuli that are novel, signal, or motivationally relevant are encoded automatically. According to LC4MP, encoding can be indexed by measuring audio and video recognition. Audio recognition tends to be a more sensitive measure than video recognition (A. Lang, Bolls, Potter, & Kawahara, 1999).1 The second major subprocess in LC4MP is storage. Storage is conceived of as the linking of recently encoded information to previously stored information (Baddeley, 1990; Bradley, 1994; Christianson, 1992; Zechmeister & Nyberg, 1982). New and old information are linked when they are concurrently activated. Thus, once new information is encoded or old information is retrieved, an active mental representation exists. Being active simultaneously forges the link. In general, the more links a new piece of information has to old information, the better it is stored. Thus, in order for information to become part of an individual’s long-term memory, it must be encoded and it must be linked to already stored information. Although something must be encoded in order to be stored, many things that are encoded are only poorly stored, because few resources are allocated to storage. Thus, encoding does not necessarily predict storage (A. Lang, Bolls, Potter, & Kawahara, 1999a). LC4MP argues that motivational relevance leads to the automatic allocation of resources to storage. Storage, in LC4MP, is indexed by cued recall techniques. Finally, the third subprocess is retrieval (Craik & Lockhart, 1972; Zechmeister & Nyberg, 1982). This subprocess involves retrieving previously stored information. Again, resources (allocated through controlled or automatic mechanisms) are required in order for retrieval to occur. The primary automatic mechanism is some sort of spreading activation. Memory is loosely conceived of as bits of information that are linked to one another. When a bit is active, this activation is thought to spread through the links to activate closely related information. Thus, as information is encoded from a message, activation spreads to related information leading to the ongoing concurrent retrieval of information related to the topic of the message. Again, the amount of ongoing concurrent retrieval will be dependent on the resources allocated to it. LC4MP indexes retrieval using free recall measures. According to LC4MP, these three processes are simultaneously and continuously active during media use. Aspects of the individual’s goals, the message content, and the message structure are continuously resulting in automatic and controlled allocation and reallocation of resources to encoding, storage, and retrieval. Resources are allocated independently to the three subprocesses out of the same fixed pool of limited resources (Basil, 1994; A. Lang et al., 1999a). When the message requirements and the user’s goals result in more calls for resources than there are, cognitive overload is said to occur. This means that there are insufficient resources available to perform all three subprocesses to the level required. When this happens, performance on one, two, or all three subprocesses will deteriorate. When there are insufficient resources, some processes will receive sufficient and others will receive insufficient resources. According to LC4MP, where the resources go may depend on the time demands of the message. Thus, if the user cannot control the speed of the message (no stopping, rewinding, or pausing), then time-sensitive subprocesses (like encoding and to some extent concurrent retrieval) will automatically receive more resources and storage will be shorted. When this occurs you end up with a message that was attended to (all resources allocated), encoded (very good recognition memory), but cannot be retrieved (poorly stored; A. Lang & Basil, 1998). According to LC4MP, the pattern of resource allocation across the subprocesses, in addition to being a function of message structure, content, and viewer goals, is also affected by motivational activation. LC4MP incorporates a view of motivational activation that posits two independent motivational systems. The appetitive and aversive systems are thought to activate in four different ways: reciprocally, that is, one is active and one is inactive; coactively, that is, both are active; inactively, that is, neither is active; or in an uncoupled way, that is, there is no relationship between their activities (Berntson & Cacioppo, 2000; Cacioppo & Gardner, 1999; A. Lang, Shin, & Lee, 2005). Thus, any given set of environmental stimuli (or mediated stimuli) might automatically activate the appetitive system, the aversive system, both systems, or neither system. LC4MP theorizes that the level of activation in the motivational systems impacts how resources are allocated across the subprocesses (A. Lang, 2005). At low levels of activation (i.e., in a neutral environment), the appetitive system is thought to be slightly more active than the aversive system. This has the evolutionary advantage of getting the animal out of the nest to explore the environment. This greater appetitive activation is called the positivity offset. An increase in negative stimuli in the environment leads to increased activation in the aversive system and an increase in positive stimuli leads to an increase in appetitive system activation. The activation functions of the two systems are thought to differ in shape. The appetitive system is thought to increase in a relatively smooth, monotonic function. Approach can occur slowly and carefully and mistaking a neutral object for a positive object (e.g., not food for food), whereas disappointing is not overly consequential. The aversive system, on the other hand, is thought to activate much more quickly at lower levels of stimulation because it is functioning to protect the organism from danger. Mistaking a dangerous thing for a neutral object might mean death, which is very consequential. Thus, the system ramps up more quickly. This steeper slope is referred to as the negativity bias. Thus, it takes an extremely positive stimulus to achieve a high level of appetitive activation, whereas a moderately negative stimulus achieves an equivalent level of aversive activation. According to LC4MP a major goal of appetitive activation is information intake, which means as much information as possible about the stimulus being approached, and the surrounding environment, needs to be encoded. In addition, the more motivationally arousing the stimulus is, the more it activates the appetitive motivational system and the more evolutionary advantage there is to encoding and storing the stimulus and the surrounding environment so that it can be found again in the future. Hence, LC4MP argues that appetitive activation increases the automatic allocation of resources to both encoding and storage and that these allocations and the level of appetitive activation increase as the stimulus becomes more arousing (A. Lang, Bradley et al., 2004; A. Lang, Sparks et al., 2004; Yegiyan, Bradley, & Lang, 2005). The goal of aversive activation, however, is not information intake but rather protection. At low levels of aversive activation, potentially negative stimuli need to be identified; therefore, at low levels of activation, resources are also automatically allocated to encoding. However, as activation increases, the organism needs to turn its mental resources to the task of figuring out what to do (fight, flee, or freeze). This requires the retrieval of relevant already stored information about the environment (escape routes, hidey holes) and the stimulus itself (how fast, how dangerous). Thus, resources are shifted away from encoding—the danger is known—to retrieval. In addition, because the danger needs to be avoided in the future, some resources are automatically shifted to storage in order to remember where this danger is located and hopefully the successful strategy about to be used to avoid it (A. Lang, Bradley et al., 2004; A. Lang, Sparks et al., 2004; Yegiyan et al., 2005). Thus, according to LC4MP, during mediated message use, controlled and automatic mechanisms allocate processing resources continuously overtime to encoding, storage, and retrieval as a function of the structure, content, and motivational and personal relevance of the mediated message. One can apply this theoretical approach to the task of creating effective messages by answering the following questions for the message being created: (a) What is the message goal? (b) Who is in the target market? (c) What medium will carry the message? and (d) What is the motivational and personal relevance of the main information in the message for the majority of people in the target audience? The answer to each question tells you something important about how to construct your message. The first question asks about the goals of the message. If the goal is awareness, then you want to create a message that attracts attention and is well encoded. If the goal is knowledge gain, the message must also be well stored. If the goal is persuasion, then the message should elicit positive evaluations, be perceived as effective, and alter attitudes. If the goal is behavior change, then the behavior to be changed, the reasons to change, and how to change must all be encoded and stored and the appropriate level of motivational activation should be associated with each of these elements (i.e., appetitive activation with how to change and aversive activation with the reasons to change). The second question asks, to whom is the message directed? This is important because some of the variables that one might use to define a target market also define groups that have demonstrably different patterns of cognitive processing and motivational activation. For example, older people allocate resources somewhat differently than younger people, their orienting responses habituate more slowly, and less arousing stimuli elicit greater motivational activation compared to young people (Kane et al., 1994; Light, Prull, & Kennison, 2000; Madden, 1990; Maki, Zonderman, & Weingartner, 1999; Orchard, 1995; Tipper, 1991). Similarly, health messages about risky behaviors are often targeted at high-sensation seekers. High-sensation seekers allocate more resources to encoding than low-sensation seekers, they tend to have slower responding aversive systems, greater activation overall in their appetitive systems, and they have bigger responses to novelty and smaller responses to arousing content than low-sensation seekers (Brocke, Beauducel, & Tasche, 1999; Greene, Krcmar, Walters, Rubin, & Hale, 2000; A. Lang, Chung, Lee, Shin, & Schwartz, in press; Lukas, 1987). These types of differences should be considered when designing effective messages. The third question asks, what medium will convey the message? Because the structure of the message plays an extremely important role in the automatic allocation of resources, medium matters (Anderson & Lorch, 1983; S. F. Geiger & Reeves, 1991; A. Lang et al., 1996). Different media have different structural features. Determining, for each medium, what structural features elicit the automatic allocation of resources has been a major focus of LC4MP research (A. Lang, Borse, Wise, & David, 2002). In addition, how different elements of structure interact over time with the presentation of important information is a major determinant of how much information is encoded and stored (A. Lang, Geiger, Strickwerda, & Sumner, 1993). Further, some aspects of structure, in some media, have been shown to affect evaluation of the message arguments, overall attitude toward the message, and intent to engage in the behavior being advocated regardless of the informational content of the message (Bolls, Muehling, & Yoon, 2003; Hitchon, Thorson & Duckler, 1994). Finally, the fourth question asks, for most people, what is the motivational and personal relevance of the primary information in the message? The answer to this question will partially determine motivational activation, which will affect the pattern of resource allocation across the subprocesses and the level of controlled resource allocation that the message may expect to receive. The latter is of particular importance because one designs messages quite differently depending on whether you think the receiver is seeking out your message and is interested in its contents or is simply being exposed to your message during a commercial break. Designing effective cancer communication messages The remainder of this paper will ask these four questions specifically about cancer communication and review relevant studies that make suggestions about how effective messages can be designed in light of these answers. Motivational and personal relevance What is the motivational and personal relevance of the main information in the message for the majority of people in the target audience? We still start with this question because for all cancer communication messages, regardless of goal, target market, or medium, a central feature of the message is that it is about cancer and cancer is to some extent motivationally relevant. Cancer is bad. Cancer is scary. Cancer is life threatening. Cancer almost certainly activates the aversive motivational system. Experience, culture, and society impart motivational relevance to things. Our motivational systems do not respond only to the primary motivators of food, sex, and danger. They also respond automatically to things that we have learned are good and bad. Once this lesson has been learned for a thing, that thing will automatically activate the appropriate motivational system (appetitive for good and aversive for bad). Thus, we have learned that cancer is dangerous and bad, so it should automatically activate the aversive motivational system to some extent, though surely our own sense of risk will influence the level of activation (related to, e.g., age, family history, and experience). For example, A. Lang, Chung, Lee, and Zhao (2005) have argued that society has defined risky products as motivationally relevant—some in positive ways and some in negative ways. Therefore, when they appear in media messages, they should elicit increased motivational activation (which is measured by increased physiological arousal) and cause resources to be automatically allocated to storage. To test this, A. Lang et al. did a study where participants viewed 30 life-sized pictures of products that were either risky (can of beer, pack of cigarettes, etc.) or not risky (can of pop, box of chalk) while their heart rate (HR) and skin conductance were collected. As predicted, participants had higher skin conductance and greater free recall for risky compared to nonrisky pictures. There were no differences in ratings of positive and negative valence possibly because for some participants, these were positive stimuli and for some they were negative stimuli. A second study repeated the same procedures using risky (marijuana, cocaine, AIDS, etc.) and nonrisky words (Tylenol, antibiotics, asthma), with the same results suggesting that it is the thing—not the representation—that matters. It seems likely then that simply mentioning cancer will elicit some activation in the aversive motivational system. This is important because at low levels of aversive activation, few resources are allocated to encoding (due to positivity offset). Indeed, recent studies suggest that calm negative messages—that is, messages about negative topics that are not arousing—receive a baseline allocation of resources that is insufficient for thorough processing; therefore, they tend to be very poorly encoded (Park, Sanders-Jackson, Wilson, & Lang, 2005). However, if messages about cancer (like messages about risky products) automatically activate the aversive motivational system, then more resources should be automatically allocated to encoding and storing these messages compared to equivalent messages not about cancer. However, this also means that one must be cautious about adding other motivationally relevant items to the message because you might create a message that is more arousing than you intended, which could result in cognitive overload. Research examining the online processing of television drug abuse prevention messages suggests that this is a very real possibility. In this research, studies using the LC4MP theoretical framework have found that increasing the structural pacing of television messages increases attention to and memory for the messages up to a point (by increasing the resources allocated to encoding through the elicitation of orienting responses) but that when messages become too fast the encoding subprocess becomes overloaded and memory suffers (A. Lang, Schwartz, Chung, & Lee, 2004; A. Lang et al., in press). In addition, when the message topic is arousing, which causes additional resources to be automatically allocated to storage, cognitive overload occurs at a slower level of structural pacing. Of particular interest here is that this same research shows that when the topic of the message involves a risky activity or product (e.g., drug abuse or alcohol advertising) overload occurs at a lower level of structural pacing for both calm and arousing appeals. The authors argue that this is due to the additional requirement to allocate resources to encoding and storing the risky (motivationally relevant) stimulus, which pushes the system into overload more quickly. This may mean that when creating messages about cancer, one must remember that the very topic itself will elicit some level of aversive activation that will cause the automatic allocation of resources to encoding and storage. Therefore, one might produce a calm message and be less concerned that it is insufficiently arousing to elicit a reasonable baseline allocation of resources. If the message design calls for complex structure to elicit orienting responses or an emotional appeal to elicit motivational activation, one should be cautious and not overuse those strategies because the system might overload at a lower level than it would with a different topic. On the other hand, for some goals, messages that produce cognitive overload may still achieve the message goals. During cognitive overload, encoding is often quite good, which makes these ads successful at increasing awareness. Storage is usually quite poor, and little information from the messages can be retrieved so overload should be avoided when information gain is a primary goal. On the other hand, messages that result in cognitive overload are often rated as enjoyable and lead to positive attitude toward the ad and intent to purchase products so some overload may be acceptable if persuasion is a primary goal (Yoon, Bolls, & Lang, 1998). It is also worth noting that people’s personal level of involvement with messages about cancer varies, which also affects motivational activation. For example, if the target market consists of cancer survivors or the families of people with cancer, these people will likely experience greater aversive activation in response to messages about cancer and those messages will be more personally relevant. As a result, both automatic and controlled processes will allocate resources to processing the message. The use of structure and emotion to increase automatic resource allocation, on top of the controlled and motivational resource allocation, would likely be wasted and might indeed backfire because you might overactivate the aversive system and cause actual withdrawal or message avoidance. On the other hand, if your target market is young adults who, on average, have a lower level of personal involvement and experience with cancer, there will be less controlled and motivationally related resource allocation; thus, the use of structure and emotional appeals to increase resource allocation may be an excellent strategy. Finally, despite the fact that all cancer messages are about cancer, which is bad, some messages may be about positive developments, which are good—new treatments, success stories, and so forth. When this is the case, you may very well be creating a message that elicits coactivation of the appetitive and aversive systems. For example, messages about prevention likely elicit the least aversive activation because they are aimed at healthy, often low risk, people and can be produced using positive appeals to increase appetitive activation. In this case, although the mere mention of cancer will elicit some aversive activation, the overall positive tone of the rest of the message should elicit a greater amount of positive activation. Because positive activation and low levels of aversive activation both result in automatic allocation of resources to encoding and storage, these messages will have a memory advantage regardless of how else they are produced. On the other hand, messages about treatments or screening for high-risk groups will have greater aversive activation (associated with the greater risk of getting or the actuality of having cancer). When this is the case, increasing appetitive activation through the use of positive appeals is likely to be very important to combat the tendency to withdraw from negative stimuli. Again, coactivation may improve processing of the message but carries with it the increased possibility of cognitive overload. What medium? The next question asks what medium will be used to deliver the message? This is an important question within the framework of the LC4MP. Each medium (the Web, television, radio, print) has its own set of perceptual channels, temporal constraints, learned signals, and orienting eliciting structural features. To design a message that achieves an optimum balance between resources allocated and resources required (i.e., does not cause cognitive overload) but at the same time maximizes resource allocation to the message (ensuring high levels of attention) requires careful attention to the demands and opportunities of the medium. For example, television (a medium much studied within the LC4MP framework) has two channels of information (audio and video). The auditory channel can have verbal information, natural sound information, music, or sound effects. The visual channel can contain still pictures, moving pictures, text, live action images, animated images, or a combination of these. The audio and the visual channels can vary in terms of the redundancy of the information in the two channels from identical to completely unrelated. The television message is usually designed to be watched in its entirety—without stopping and starting—and it is temporally demanding so that information not encoded at exposure is lost. New parts of the message are continually demanding to be encoded, whereas the older parts of the message are being stored. Viewed in this way, the cognitive task of encoding and storing the information in a television message is extremely taxing. As a result, only a small subset of the information in a television message is ever encoded, less of it is stored, and even less of it is retrievable at a later date. The goal, when designing television messages, is to use structural features to increase the probability that important parts of the message are encoded and then to ensure that there are periods of time during the message when sufficient resources are allocated to storage and competing message demands for resources are low so that the encoded information can indeed be stored. Research has shown that many structural features of television elicit orienting responses including scene changes, camera changes, loud noises, sudden movements toward the camera, and the onset of videographics (A. Lang, 2000). Other structural features do not appear to elicit orienting responses like the onset of slow motion, pans, and zooms. This means that some structural features can be used to call resources to encoding. Research shows that the information presented immediately following an orienting eliciting structural feature is encoded better when the resource demands of the informational content are low and worse if the informational demands are high (Fox, Lang, Chung, Lee, & Potter, 2004; Thorson & Lang, 1992). Thus, when the message itself is difficult or unfamiliar—and therefore requires many resources—orienting responses will interfere with that processing and decrease encoding of the message. However, when messages are familiar or easy, the reverse occurs and the additional allocation of resources elicited by the orienting response increases encoding of the information. This means that when designing television messages, one should judge, for your target audience, the difficulty of the information you want them to understand and learn. If it is easy, use judiciously placed structural features to maintain a high level of encoding. If it is difficult, refrain from using structural features as they will interfere with ongoing processing. For example, if your message is simple and familiar (e.g., smoking causes lung cancer, stop smoking) you may want to include graphics (say animations of smoke turning pretty pink lungs black and tumor ridden) to increase resource allocation. On the other hand, if the message is new and complex (the steps needed to carry out a specific diet regimen), you might want to simplify your presentation and maximize audio/visual redundancy in order to increase storage. Further, judicious placing is an important part of the use of orienting eliciting structural features. Information placed 1–2 seconds after a structural feature is not encoded as well as information placed 3–5 seconds later (A. Lang, 1991). When designing television messages for maximum ease of processing, it is important to keep the temporal unfolding of events in mind. Television messages happen over time. When the message begins, the viewer’s mind is not necessarily attuned to the topic of the message. The moment the message begins, the images and words begin to activate links to stored information and the automatic retrieval of information needed to understand the message begins. Thus, the choice of what comes first, in a message, will influence the type of related information that is retrieved. When initial information does not elicit appropriate background information, then the processing task becomes much more difficult because newly encoded information must be maintained as an active mental representation without being linked to related content (A. Lang, Sias, Chantrill, & Burek, 1995). Research has shown, for example, that if news stories are presented in chronological order rather than in the traditional broadcast style of what’s new, followed by the causes and the consequences, this reduces the need to retrieve already stored information and increases encoding of the news stories by 15% (A. Lang, 1989). Similarly, if the goal of the message designer is to maximize encoding and storage, it makes sense to design a message that activates the appetitive motivational system. However, it is important to realize that motivational activation also occurs over time. It takes a certain amount of time to build emotion in the message in order to activate the associated motivational system. Although some images are hardwired and may elicit almost immediate activation (babies being born, sexual encounters), others require narrative consequences to gain their emotional punch. Therefore, you may want to begin a message with a measured emotional appeal designed to appropriately activate the motivational system. During this period, one would not want to introduce much information. However, once the emotional story has activated, the motivational system—then the presentation of factual information, which you want to be stored—will be facilitated by the motivational activation. Similarly, messages might vary in emotional tone in order to activate first one motivational system and then the other. For example, consider a behavior change message where the goal is to encourage women to go in for a yearly mammogram. One might begin the message with strong positive images linked to a positive story followed by the message “it’s easy, it’s important, here’s what you should do.” This part of the message, what to do and how, will then be well encoded and stored as a result of the appetitive activation. The next part of the message might tie the lack of action to aversive motivation. Thus, the message might change suddenly from positive to negative (the same people who were happy and cancer free now dying of cancer), followed by the message, “failure to act can kill.” This creates a message that functions as a conditioning stimulus. Positive emotions and appetitive activation are linked to the message of having a mammogram, whereas negative emotions and aversive activation are linked to the message that not having a mammogram is potentially deadly. Finally, it should also be noted that the use of words is very important in audio and audio/visual messages. Research shows that motivationally relevant words embedded in ongoing audio messages have the ability to elicit orienting responses and are remembered better than nonmotivationally relevant words (Lee & Potter, 2005; Potter et al., 1997). Thus, the use of motivationally relevant and emotional words will increase resources allocated to processing the message and focus those resources on a specific word. Care should be taken to make sure that the word is fundamentally related to the message or else the word will distract from the overall meaning of the message. Research using the LC4MP framework has also been done on other media to begin identifying the structural features that elicit orienting and the types of information presentation that are maximally effective. To date, both audio-only information (e.g., radio messages) and simple computer- or Web-based messages have been investigated (Borse & Lang, 2000; Chung, Lee, Lang, Borse, & Buchman, 2002; A. Lang, Lee, Chung, & Borse, 2001; A. Lang et al., 2002). Little research has been done on text messages (Wise & Lang, 2001). A brief summary of what this research tells us about these media follows. Audio-only, or radio, information, like television messages, is temporally constrained. It tends to begin and continue. Generally, these messages are designed to be heard in their entirety. Therefore, they too are subject to cognitive overload if too much information is presented too quickly or if structural elements are placed in such a way as to interfere with rather than aid the processing of verbal information (Angelini, Lee, Schwartz, Sparks, & Lang, 2003; Lee, Angelini, Schwartz, & Lang, 2003). Unlike television, there is only a single channel of information presentation—audio—but there may be several strands of audio contained in that channel. Thus, a message might contain a verbal narrative, a natural sound background, and sound effects punctuating the high points of the message. Thus, one strand of audio could interfere with or compliment a second strand of audio. Many audio structural features have been shown to elicit orienting responses in attentive viewers including motivationally relevant words, sound effects, voice changes (from one speaker to another), and music onsets (Potter et al., 1997). As with television, research has shown that increasing the pacing of audio structural features initially improves memory for the message but eventually overloads the system and decreases memory for messages (Potter, 2000; Potter & Callison, 2000). Research examining audio drug prevention messages strongly suggests that audio messages about arousing topics are processed much more thoroughly at low compared to high levels of audio structural complexity (A. Lang, Schwartz, Lee, & Angelini, 2006,A. Lang, Schwartz, Lee, & Shin, 2003). With audio messages, it is likely that either emotional content or structural complexity improves processing but the combination does not. Similarly, the use of concrete language, chronological presentation of information, and multiple voices are good combinations for increasing encoding of the message. Research examining computer presentation of information and simple Web stimuli from the LC4MP perspective is still in its early stages. Several early studies demonstrated convincingly and surprisingly that the simple appearance of information on a computer screen does not elicit an orienting response. Instead, research suggests that the content of the information must be either personally or motivationally relevant for an orienting response to occur. This means that a great deal of computer-presented information must rely on controlled resource allocation (i.e., the viewer wants to pay attention to it) in order to be encoded. Of particular interest here is that studies suggest that the element of control may not play a particularly large role in the automatic processing of computer-presented messages. In one study (A. Lang et al., 2002), three different types of text-based messages were presented on a computer screen, a headline, a headline in a box, or a personally relevant warning. Participants clicked to bring up half of the stimuli and the other half were presented by the computer. There was no difference in orienting behavior between the two groups. The headlines and the boxed headlines did not elicit orienting but the personally relevant warnings did, regardless of how onset was initiated. Several other studies have examined the effects of animation on orienting behavior (Chung, 2005). These studies suggest that initially animation does elicit orienting responses but that the orienting response quickly habituates to repeated use of animation. In addition, the resources allocated as a function of the orienting response appear to be allocated to the object that is moving and not to the rest of the information on the screen. Thus, unless the goal is to have people remember the animated object, animation may interfere with the processing of the central information in a message. Studies on educational multimedia have also suggested that sound effects elicit orienting in computer users and increase encoding for the sound. For example, in one study, 9- to 13-year-old participants encoded the most information from multimedia presentations that included both animation and sound effects compared to one or the other or none (Schwartz, 2005). Of particular interest here, however, was that the increase in encoding that occurs as a result of the orienting responses is not generalized to the message. Instead, sound effects increase memory for the sound and animations increase memory for the animated object. This conjunction of orienting eliciting feature and encoding specificity appears to be somewhat unique to computer and multimedia messages perhaps because users have control over the time of presentation. In television and radio messages, which continue rather relentlessly onward following an orienting eliciting structural feature, the additional resources allocated seem to be less focused on the feature that elicited them, which has already become history, and more generally used to process the parts of the message that follow. Target market? The next question asks, who is in the target market? Obviously this needs to be considered first when considering what medium to use. Is the target market media savvy? Are they online, connected, or dialed in? Or is the target market more traditional, watching TV, listening to the radio, reading the newspaper. This aspect of targeting will play an important role in determining what medium to use that will then affect message design as discussed in the previous section. However, the aspect of this question on which this section is focused is whether there is something about the target that will make this group of people process media differently. In general, will this group respond differently to structural or content aspects of the message? Is there something about the before group that makes their cognitive and/or motivational systems function differently from the “norm”? This is an area in need of much research. Much of what is known is new, speculative, or borrowed and not yet tested in the types of mediated environments under consideration here. Yet, it is a critically important when designing messages to teach, persuade, inform, reinforce, and even change behavior. Research on drug prevention messages provides an excellent example of a target market that is presumed to process mediated messages differently, namely, sensation seekers. Sensation seeking is a personality trait that has been shown to be associated with drug experimentation, drug use, drug abuse, smoking, and a myriad of other risky behaviors like gambling, unsafe sex, and reckless driving, some of which are related to increased cancer risk such as smoking and risky sexual behaviors (Donohew et al., 2000; Greene et al., 2000; Jonah, 1997; Parent & Newman, 1999; Sheer & Cline, 1995; Wood, Cochran, Pfefferbaum, & Arneklev, 1995). Research in this area suggests that high-sensation seekers seek out novel and arousing content and that they require more arousing content to experience the same level of arousal as low-sensation seekers (Aluja Fabregat & Torrubia Beltri, 1998; Donohew, Palmgreen, & Lorch, 1994; Krcmar & Greene, 1999; Lorch, Palmgreen, Donohew, & Helm, 1994; Palmgreen et al., 1991). For these reasons, drug prevention messages targeted at high-sensation seekers, who are at the most risk of using drugs, need to be placed within media that appeal to high-sensation seekers and the messages themselves need to have high sensation value (i.e., they should be complex, fast paced, emotional, etc.). Within the framework of the LC4MP, individual differences are thought to operate at almost every level of interaction between a mediated message and a media user. Although certain things will elicit orienting responses in all people, there is also a series of idiosyncratic things that will elicit orienting just for a single individual. For some target markets, specific words or images might elicit orienting though they would not for the general population. Similarly, although positive messages should elicit appetitive activation and negative messages should elicit aversive activation, what is positive and what is negative will have both cultural and individual variation. For example, research has shown that pictures of food activate the aversive system in people with eating disorders and the appetitive system in those without eating disorders. Similarly, pictures of people smoking activate the aversive system in nonsmokers and the appetitive system in smokers (Geir, Mucha, & Pauli, 2000). Obviously, this sort of group difference will have an effect on designing messages to stop people smoking because for one group, smoking-related imagery is positive, whereas for the other it is negative and the resulting motivational activation will affect message processing. Another area in which individual variation may play a role is related to both sensation seeking and individual differences in motivational activation. As discussed previously, the motivational activation functions are characterized by positivity offset (a greater level of appetitive compared to aversive activation at low levels of activation) and negativity bias (a steeper rate of increase in aversive compared to appetitive activation). Recent research suggests that these two things—positivity offset and negativity bias—also exist as individual difference variables (A. Lang, Shin et al., 2005). Thus, some people have a particularly high positivity offset and are more driven to seek out new activities and new experiences, whereas others have a low positivity offset and are quite cautious about new experiences and activities. Similarly, those with high negativity bias respond very quickly and strongly to stimuli that are only weakly negative, whereas those with a low negativity bias require extremely strong negative stimuli to achieve the same level of aversive activation. Recent work on the processing of drug prevention messages has demonstrated that sensation seeking is related to positivity offset and negativity bias and that an individual’s level of positivity offset and negativity bias influences their online processing of emotional health-related messages (Yegiyan et al., 2005). In one study, participants viewed public service announcements about smoking cessation, drug prevention, and safe sex that were positive or negative and arousing or calm. During viewing, participant’s HR, skin conductance, evoked startle response, facial muscle activity (smiling, zygomatic and frowning, corrugator), and self-reported emotional experience were measured. Results showed that individuals with a high positivity offset paid more attention to and felt more positive during all types of messages as was demonstrated by inhibited startles (indicative of appetitive activation), lower HR (indicative of greater attention), higher skin conductance (indicative of more arousal), less corrugator activity (indicative of less negative emotion), and more zygomatic activity (indicative or positive emotion) and greater self-reported positive feelings. However, they did not remember more of everything—despite their evident attention and enjoyment—positivity offset did not have an impact on encoding. Instead, arousing content (positive or negative) was encoded best regardless of positivity offset. As a result, messages aimed at those high in positivity offset (who tend to use and experiment more with drugs) should be arousing because they attend to and enjoy all sorts of messages but learn the most from arousing messages. Further, results showed that participants with a high compared to those with a low negativity bias responded more strongly and more negatively to negative messages. This increase in negative responses to negative messages was demonstrated by increased startle magnitude, greater corrugator activity, higher ratings of experienced negativity, increased skin conductance, and decreased HR indicative of disengagement. Surprisingly, however, despite this dislike and disengagement, they encoded the messages better. Thus, it seems likely that high negativity bias individuals, who tend to experiment with and use drugs less than other groups, will avoid negative public service announcements, but if they are exposed to them, they will convey information and reinforce their existing tendencies to avoid drug use. Another way in which target market might influence message design is when target markets are determined by age. Research shows that differences exist in how children, adolescents, adults, and the elderly process mediated messages. For example, research on structural complexity or pacing suggests that younger viewers like fast-paced messages more than older viewers (A. Lang & Schwartz, 2002). However, children and older viewers are overloaded by complex structure at a lower level of complexity than are adolescent and young adult viewers (A. Lang, Schwartz et al., 2004). Similarly, children and adolescents appear to be somewhat more affected by structural and concrete elements of messages; in other words, they are a little bit more closely bound by the stimulus, whereas adults and older adults appear to respond somewhat more to the abstract or emotional elements of messages. There is also a great deal of research examining the effects of aging on cognitive processing though there is not yet a great deal of research on how these cognitive effects of aging might alter mediated message processing. Quite a bit of research suggests that although the elicitation of orienting responses is not effected by age, the habituation of orienting responses is (Kane et al., 1994; Light et al., 2000; Madden, 1990; Maki et al., 1999). Thus, for example, the research discussed earlier that showed orienting responses to computer animation habituating fairly quickly was done on younger adults. It is possible that orienting responses to animations on the web might habituate more slowly or not at all with older media users. Similarly, research also suggests that arousing messages and negative messages may elicit greater aversive activation in older adults, especially women. In any case, once the target market is determined, care should be taken to understand whether the characteristics that define the target market will also have a generalized effect on some aspect of mediated message processing. Message goal The final question asks, what is the goal of the message? Is it to maintain attention, to create awareness of something, to convey complex information, to persuade, to change behavior, to associate an emotional state with a behavior, or something else? In order to apply LC4MP to designing effective messages, one must know what part of processing needs to be optimized because the system’s limited capacity will not allow all aspects of processing to be performed equally well. Thus, for example, local broadcast news consultants have devised a news format that maximizes attentional inertia and repeat viewing (Grabe, Lang, & Zhao, 2003; Grabe, Lang, Zhou, & Bolls, 2000; Grabe, Zhou, Lang, & Bolls, 2000; A. Lang, Potter, & Grabe, 2003). To keep people watching, they use repeated automatic calls for attention using both structural and motivational elements. To bring them back, they end happy in order to leave viewers with a good feeling about watching. Thus, fast pacing is the primary tool because people tend to like fast-paced presentation and it results in frequent orienting eliciting structural features. Similarly, news maxims like “if it bleeds it leads” epitomize the use of motivationally relevant stimuli to gain and maintain attention. However, this same format, which keeps eyeballs on screen, leads to extremely low levels of encoding and storage, which is why so many studies show a negative relationship between watching television news and level of news information holding. Another broadcast format people are well acquainted with is Sesame Street. Again, one of its goals is to bring eyes to the screen (Anderson & Levin, 1976; Ball, Palmer, & Millward, 1986; Lorch & Castle, 1997). Like broadcast news, it uses short bits and lots of structural features to automatically call wandering attention back to the message. Unlike broadcast news, however, it also has the goal of maximizing learning (i.e., encoding and storage); therefore, despite the somewhat complex structure, the message content is kept simple and is repeated frequently. This is an A, an A, an A, this is the letter A. Thus, the structural features do increase resource allocation to encoding, but there is less information to encode because the message is simple and cognitive overload is less likely to occur. A great deal of research from the LC4MP perspective has dealt with how to design messages that increase attention and arousal without pushing the system into cognitive overload. The reason for this is that most message designers want their messages to be attended to and many want them to be remembered. The keys to attention are eliciting orienting responses and using motivationally relevant elements in your message. The keys to good memory are controlled resource allocation (e.g., cognitive effort), the automatic allocation of resources to encoding, and the creation of arousal in viewers through the use of either arousing content or fast-paced structure. The correct combination of these things leads to high levels of attention and memory. Other message goals include persuasion, evaluative responses, and behavior change. Less work using the LC4MP exists in these areas. However, some research with television and radio has looked at attitudes and liking for messages, evaluation of argument effectiveness, and intent to change behavior. This research has generally shown that fast pacing (at least with young adults) leads to positive attitude toward the messages, increased liking for the messages, increased intent to adopt the product/change advocated by the message, and increased evaluation of argument effectiveness (Bolls et al., 2003; Potter & Choi, in press; Yoon et al., 1998). However, evaluation also interacts with the emotional content of the message. For example, recent research has compared the judged effectiveness of claims appearing in drug prevention PSAs that vary in terms of valence (positive or negative), arousing content (arousing or calm), and structural pacing with that of the same claims separated from the executional elements of the message (i.e., with the claims presented only in text). This research shows that the claims are judged to be most effective when messages are arousing and negative. Claims in calm positive messages are judged to be the least effective. However, if one controls for the effectiveness of the claims (judged as text sentences), an interaction appears. Claims in negative arousing messages that are judged to be effective outside the message environment are judged as the most effective. On the other hand, negative arousing messages that contain claims judged to be ineffective outside the message environment are judged to be the least effective. Thus, arousing negative emotional appeals only improve the evaluation of PSAs effectiveness if the actual information in the PSA is information that would be judged to be effective outside of the PSA (A. Lang & Yegiyan, 2005). Conclusions This paper argues that using the LC4MP perspective to design effective health communication messages, and specifically effective health messages about cancer, can be valuable. The LC4MP would suggest that cancer is a motivationally relevant topic that will elicit some level of aversive activation. Target markets for various types of cancer-related messages (e.g., smokers or people of a certain age) will process mediated messages in predictably different ways making certain design decisions better for certain target markets. Both structural and content elements of messages interact with the limited capacity information processing system to impact the level of resources allocated independently to encoding, storage, and retrieval. This pattern of allocation, in turn, will determine if messages are encoded and stored well and can therefore be retrieved at a later decision point. Finally, individual difference in people’s motivational activation may influence both their tendencies to engage in certain risky behaviors that increase the probabilities of getting cancer and their processing of health-related messages. Future research in this perspective should be done to learn how to optimize cancer messages for specific target audiences using specific media to convey that information. Acknowledgment This research supported by the National Institute of Drug Abuse R01-NIDA-12359. Note 1 " In general, video encoding requires fewer resources than audio encoding. Therefore, changes in message design that impact resource levels impact audio encoding before they impact video encoding. Because most of the actual information in messages is contained in the audio or textual aspects of the message, the impact of resource depletion on audio encoding tells us more about changes in design that are affecting message encoding. References Aluja Fabregat , A. , & Torrubia Beltri , R. ( 1998 ). Viewing of mass media violence, perception of violence, personality and academic achievement . Personality and Individual Differences , 25 , 973 – 989 . Google Scholar Crossref Search ADS WorldCat Anderson , D. R. , & Levin , S. R. ( 1976 ). Young children’s attention to “Sesame Street.” Child Development , 47 , 806 – 811 . Google Scholar Crossref Search ADS WorldCat Anderson , D. R. , & Lorch , E. ( 1983 ). Looking at television: Action or reaction. In J. Bryant & D. R. Anderson (Eds.), Children’s understanding of television (pp. 1 – 34 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Angelini , J. R. , Lee , S., Schwartz , N., Sparks , J. V., & Lang , A. ( 2003 ). Processing radio public service announcements: Arousing content, production pacing, and children . Psychophysiology , 40 ( Suppl. 1 ), S23 . OpenURL Placeholder Text WorldCat Baddeley , A. D . ( 1990 ). Human memory: Theory and practice . Needham Heights, MA : Allyn & Bacon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ball , S. , Palmer , P., & Millward , E. ( 1986 ). Television and its educational impact: A reconsideration. In J. Bryant & D. Zillmann (Eds.), Perspectives on media effects (pp. 129 – 142 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Basil , M . ( 1994 ). Multiple resource theory I: Application to television viewing . Communication Research , 21 , 177 – 207 . Google Scholar Crossref Search ADS WorldCat Berntson , G. G. , & Cacioppo , J. T. ( 2000 ). Psychobiology and social psychology: Past, present, and future . Personality and Social Psychology Review , 4 , 3 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Bolls , P. , Muehling , D. D., & Yoon , K. ( 2003 ). The effects of television commercial pacing on viewers’ attention and memory . Journal of Marketing Communication , 9 , 17 – 28 . Google Scholar Crossref Search ADS WorldCat Borse , J. , & Lang , A. ( 2000 ). The effects of Web banner advertisements: A study of the impact of animation and interactivity on memory, click-through, attention, arousal, and affect . Paper presented at the International Communication Association , Acapulco, Mexico . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bradley , M. M . ( 1994 ). Emotional memory: A dimensional analysis. In S. H. M. V. Goozen, N. E. Van de Poll & J. A. Sergeant (Eds.), Emotions: Essays on emotion theory . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Brocke , B. , Beauducel , A., & Tasche , K. G. ( 1999 ). Biopsychological bases and behavioral correlates of sensation seeking: Contributions to a multilevel validation . Personality and Individual Differences , 26 , 1103 – 1123 . Google Scholar Crossref Search ADS WorldCat Cacioppo , J. T. , & Gardner , W. L. ( 1999 ). Emotion . Annual Reviews: Psychology , 50 , 191 – 214 . Google Scholar Crossref Search ADS WorldCat Christianson , S . ( 1992 ). The handbook of emotion and memory: Research and theory . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chung , Y . ( 2005 ). Processing advertisements on the Web: The effect of animation and arousing pictures on orienting, arousal, encoding, and storage . Unpublished dissertation, Indiana University , Bloomington . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Chung , Y. , Lee , S., Lang , A., Borse , J., & Buchman , J. ( 2002 , July). Orienting to text on screen: Task or medium? Paper presented at the International Communication Association , Seoul, South Korea . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Craik , F. I. M. , & Lockhart , R. S. ( 1972 ). Levels of processing: A framework for memory research . Journal of Verbal Learning and Verbal Behavior , 11 , 671 – 684 . Google Scholar Crossref Search ADS WorldCat Donohew , L. , Palmgreen , P., & Lorch , E. P. ( 1994 ). Attention, need for sensation, and health communication campaigns . American Behavioral Scientist , 38 , 310 – 322 . Google Scholar Crossref Search ADS WorldCat Donohew , L. , Zimmerman , R., Cupp , P. S., Novak , S., Colon , S., & Abell , R. ( 2000 ). Sensation seeking, impulsive decision-making, and risky sex: Implications for risk-taking and design of interventions . Personality and Individual Differences , 28 , 1079 – 1091 . Google Scholar Crossref Search ADS WorldCat Fox , J. , Lang , A., Chung , Y., Lee , S., & Potter , D. ( 2004 ). Picture this: Effects of graphics on the processing of television news . Journal of Broadcasting and Electronic Media , 48 , 646 – 674 . Google Scholar Crossref Search ADS WorldCat Geiger , S. , & Reeves , B. ( 1993 ). The effects of scene changes and semantic relatedness on attention to television . Communication Research , 20 , 155 – 175 . Google Scholar Crossref Search ADS WorldCat Geiger , S. F. , & Reeves , B. ( 1991 ). The effects of visual structure and content emphasis on the evaluation and memory for political candidates. In F. Biocca (Ed.), Television and political advertising. Volume 1: Psychological processes (pp. 125 – 143 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Geir , A. , Mucha , R. F., & Pauli , P. ( 2000 ). Appetitive nature of drug cues confirmed with physiological measures in a model using pictures of smoking . Psychopharmacology , 150 , 283 – 291 . Google Scholar Crossref Search ADS PubMed WorldCat Grabe , M. E. , Lang , A., & Zhao , X. Q. ( 2003 ). News content and form—Implications for memory and audience evaluations . Communication Research , 3 , 387 – 413 . Google Scholar Crossref Search ADS WorldCat Grabe , M. E. , Lang , A., Zhou , S. H., & Bolls , P. D. ( 2000 ). Cognitive access to negatively arousing news—An experimental investigation of the knowledge gap . Communication Research , 27 , 3 – 26 . Google Scholar Crossref Search ADS WorldCat Grabe , M. E. , Zhou , S. H., Lang , A., & Bolls , P. D. ( 2000 ). Packaging television news: The effects of tabloid on information processing and evaluative responses . Journal of Broadcasting and Electronic Media , 44 , 581 – 598 . Google Scholar Crossref Search ADS WorldCat Graham , F. K . ( 1979 ). Distinguishing among orienting, defense, and startle reflexes. In H. D. Kimmel, E. H. Van Olst, & J. F. Orlebeke (Eds.), The orienting reflex in humans (pp. 137 – 167 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Greene , K. , Krcmar , M., Walters , L. H., Rubin , D. L., & Hale , J. L. ( 2000 ). Targeting adolescent risk-taking behaviors: The contribution of egocentrism and sensation-seeking . Journal of Adolescence , 23 , 439 – 461 . Google Scholar Crossref Search ADS PubMed WorldCat Hitchon , J. , Thorson , E., & Duckler , P. ( 1994 ). Effects of ambiguity and complexity on consumer response to music video commercials . Journal of Broadcasting and Electronic Media , 38 , 289 – 306 . Google Scholar Crossref Search ADS WorldCat Jonah , B. A . ( 1997 ). Sensation seeking and risky driving: A review and synthesis of the literature . Accident Analysis and Prevention , 29 , 651 – 665 . Google Scholar Crossref Search ADS PubMed WorldCat Kane , M. J. , Hasher , L., Stoltzfus , E. R., Zacks , R. T., Connelly , S. L. ( 1994 ). Inhibitory attentional mechanisms and aging . Psychology and Aging , 9 , 103 – 112 . Google Scholar Crossref Search ADS PubMed WorldCat Krcmar , M. , & Greene , K. ( 1999 ). Predicting exposure to and uses of television violence . Journal of Communication , 49 , 24 – 45 . Google Scholar Crossref Search ADS WorldCat Lang , A . ( 1989 ). Effects of chronological presentation of information on processing and memory for broadcast news . Journal of Broadcasting and Electronic Media , 33 , 441 – 452 . OpenURL Placeholder Text WorldCat Lang , A . ( 1991 ). Emotion, formal features, and memory for televised political advertisements. In F. Biocca (Ed.), Television and political advertising, Vol. 1: Psychological processes; Vol. 2: Signs, codes, and images (pp. 221 – 243 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A . ( 2000 ). The limited capacity model of mediated message processing . Journal of Communication , 50 ( 1 ), 46 – 70 . Google Scholar Crossref Search ADS WorldCat Lang , A . ( 2006 ). Motivated cognition (LC4MP): The influence of appetitive and aversive activation on the processing of video games. In P. Messarsis & L. Humphries (Eds.), Digital media: Transformation in human communication (pp. 237 – 256 ). New York : Peter Lang . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , & Basil , M. ( 1998 ). What do secondary task reaction times measure anyway? In M. Roloff (Ed.), Communication yearbook (Vol. 21 , pp. 443 – 470 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , Bolls , P., Potter , R. F., & Kawahara , K. ( 1999 ). The effects of production pacing and arousing content on the information processing of television messages . Journal of Broadcasting and Electronic Media , 43 , 451 – 475 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Borse , J., Wise , K., & David , P. ( 2002 ). Captured by the World Wide Web—Orienting to structural and content features of computer-presented information . Communication Research , 29 , 215 – 245 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Bradley , S. D., & Sparks , J. V. ( 2004 ). Processing arousing information: Psychophysiological predictors of motivated attention, sensation seeking, and substance use . Paper presented to the Information Systems division of the International Communication Association , New Orleans, LA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , Chung , Y., Lee , S., Shin , M., & Schwartz , N. ( 2005 ). It’s an arousing, fast-paced kind of the world: The effects of age and sensation seeking on the information processing of substance abuse PSAs . Media Psychology , 7 , 421 – 454 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Chung , Y., Lee , S., & Zhao , X. ( 2005 ). It’s the product: Do risky products compel attention and elicit arousal in media users? Health Communication , 17 , 283 – 300 . Google Scholar Crossref Search ADS PubMed WorldCat Lang , A. , Geiger , S., Strickwerda , M., & Sumner , J. ( 1993 ). The effects of related and unrelated cuts on viewers’ memory for television: A limited capacity theory of television viewing . Communication Research , 20 , 4 – 29 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Lee , S., Chung , Y., & Borse , J. ( 2001 ). Task not medium: Orienting to Web banner advertisements . Unpublished manuscript, Department of Telecommunications, Indiana University . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , Newhagen , J., & Reeves , B. ( 1996 ). Negative video as structure: Emotion, attention, capacity, and memory . Journal of Broadcasting and Electronic Media , 40 , 460 – 477 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Potter , D., & Grabe , M. E. ( 2003 ). Making news memorable: Applying theory to the production of local television news . Journal of Broadcasting and Electronic Media , 47 , 113 – 123 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Potter , R. F., & Bolls , P. D. ( 1999 ). Something for nothing: Is visual encoding automatic? Media Psychology , 1 , 145 – 163 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Schwartz , N., Chung , Y., & Lee , S. ( 2004 ). Processing substance abuse messages: Production pacing, arousing content, and age . Journal of Broadcasting and Electronic Media , 48 , 61 – 88 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Schwartz , N., Lee , S., & Angelini , J. ( 2006 ). Processing radio PSAs: Production pacing, arousing content, and age . Manuscript submitted for publication. Lang , A. , Schwartz , N., Lee , S., & Shin , M. ( 2003 ). Processing radio PSAs: Production pacing, arousing content, and age . Presented to the Information Systems Division of the International Communication Association , San Diego, CA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , Schwartz , N., & Snyder , J. F. ( 1999, August ). Slow down, you’re moving to fast: Pacing, arousing content, and those aging boomers . Presented to the Theory and Methodology Division of the Association for Education in Journalism and Mass Communication , New Orleans, LA . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , A. , Shin , M., & Lee , S. ( 2005 ). Sensation seeking, motivation, and substance use: A dual system approach . Media Psychology , 7 , 1 – 29 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Sias , P. M., Chantrill , P., & Burek , J. A. ( 1995 ). Tell me a story: Narrative elaboration and memory for television . Communication Reports , 8 , 102 – 110 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Sparks , J. V., Bradley , S. D., Lee , S., & Wang , Z. ( 2004 ). Processing arousing information: Psychophysiological predictors of motivated attention . Psychophysiology , 41 ( Suppl. 1 ), S61 . OpenURL Placeholder Text WorldCat Lang , A. , & Yegiyan , N. ( 2005 ). The effects of argument strength, emotion, and production pacing on judgments of claim effectiveness . Unpublished manuscript. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lang , P. J. , Bradley , M. M., & Cuthbert , B. N. ( 1990 ). Emotion, attention, and the startle reflex . Psychological Review , 97 , 377 – 395 . Google Scholar Crossref Search ADS PubMed WorldCat Lang , P. J. , Bradley , M. M., & Cuthbert , B. N. ( 1997 ). Motivated attention: Affect, activation, and action. In P. J. Lang, R. F. Simons, & M. T. Balaban (Eds.), Attention and orienting: Sensory and motivational processes . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Lee , S. , Angelini , J. R., Schwartz , N., & Lang , A. ( 2003 ). Processing radio public service announcements: Arousing content, production pacing, and college students . Psychophysiology , 40 ( Suppl. 1 ), S54 . OpenURL Placeholder Text WorldCat Lee , S. , & Potter , R. F. ( 2005 ). Do words matter?: Effects of emotional words in radio messages on orienting, physiological arousal, and facial EMG response . Paper presented to the International Communications Association at their annual conference in New York . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Light , L. L. , Prull , M. W., & Kennison , R. F. ( 2000 ). Divided attention, aging, and priming in exemplar generation and category verification . Memory and Cognition , 28 , 856 – 872 . Google Scholar Crossref Search ADS PubMed WorldCat Lorch , E. P. , & Castle , V. J. ( 1997 ). Preschool children’s attention to television: Visual attention and probe response times . Journal of Experimental Child Psychology , 66 , 111 – 127 . Google Scholar Crossref Search ADS PubMed WorldCat Lorch , E. P. , Palmgreen , P., Donohew , L., & Helm , D. ( 1994 ). Program context, sensation seeking, and attention to televised anti-drug public service announcements . Human Communication Research , 20 , 390 – 412 . Google Scholar Crossref Search ADS WorldCat Lukas , J. H . ( 1987 ). Visual evoked potential augmenting-reducing and personality: The vertex augmenter is a sensation seeker . Personality and Individual Differences , 8 , 385 – 395 . Google Scholar Crossref Search ADS WorldCat Madden , D. J . ( 1990 ). Adult age differences in attentional selectivity and capacity . European Journal of Cognitive Psychology , 2 , 229 – 252 . Google Scholar Crossref Search ADS WorldCat Maki , P. M. , Zonderman , A. B., & Weingartner , H. ( 1999 ). Age differences in implicit memory: Fragmented object identification and category exemplar generation . Psychology and Aging , 14 , 284 – 294 . Google Scholar Crossref Search ADS PubMed WorldCat Orchard , R. J . ( 1995 ). Memory and attention in the healthy elderly . Dissertation Abstracts International: Section B: The Sciences and Engineering , 55 ( 9-B ), 4165 . OpenURL Placeholder Text WorldCat Palmgreen , P. , Donohew , L., Lorch , E. P., Rogus , M., Helm , D., & Grant , N. ( 1991 ). Sensation seeking, message sensation value, and drug use as mediators of PSA effectiveness . Health Communication , 3 , 217 – 227 . Google Scholar Crossref Search ADS WorldCat Parent , E. C. , & Newman , D. L. ( 1999 ). The role of sensation-seeking in alcohol use and risk-taking behavior among college women . Journal of Alcohol and Drug Education , 44 , 12 – 28 . OpenURL Placeholder Text WorldCat Park , B. , Sanders-Jackson , A., Wilson , B. D., & Lang , A. ( 2005 , May). Separating speed from load: Understanding how pacing and information contribute to variation in STRTs . Paper presented to the Information Systems Division of the International Communication Association , New York . Potter , R. F . ( 2000 ). The effects of voice changes on orienting and immediate cognitive overload in radio listeners . Media Psychology , 2 , 147 – 178 . Google Scholar Crossref Search ADS WorldCat Potter , R. F. , Bolls , P., Lang , A., Zhou , S., Schwartz , N., Borse , J., et al. . ( 1997 , August). What is it? Orienting to structural features of radio messages . Paper presented to the Theory and Methodology Division of the Association for Education in Journalism and Mass Communication , Chicago, IL . Potter , R. F. , & Callison , C. ( 2000 ). Sounds exciting!: The effects of audio complexity on listeners’ attitudes and memory for radio promotional announcements . Journal of Radio Studies , 7 , 29 – 51 . Google Scholar Crossref Search ADS WorldCat Potter , R. F. , & Choi , J. ( in press ). The effects of auditory structural complexity on attitudes, attention, arousal, and memory . Media Psychology . OpenURL Placeholder Text WorldCat Rafaeli , S . ( 1988 ). Interactivity from new media to communication. In R. P. Hawkins, J. M. Wiemann & S. Pingree (Eds.), Advancing communication science: Merging mass and interpersonal processes (pp. 110 – 134 ). Newbury Park, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Reeves , B. , Thorson , E., Rothschild , M., McDonald , D., Hirsch , J., & Goldstein , R. ( 1985 ). Attention to television: Intrastimulus effects of movement and scene changes on alpha variation over time . International Journal of Neuroscience , 25 , 241 – 255 . Google Scholar Crossref Search ADS WorldCat Reeves , B. , Thorson , E., & Schleuder , J. ( 1986 ). Attention to television: Psychological theories and chronometric measures. In J. Bryant & D. Zillmann (Eds.), Perspectives on media effects (pp. 251 – 279 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schneider , W. , Dumais , S. T., & Shiffrin , R. M. ( 1984 ). Automatic and control processing and attention. In R. Parasuraman & D. R. Davies (Eds.), Varieties of attention (pp. 1 – 25 ). Orlando, FL : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schwartz , N . ( 2005 ). The impact of animation and sound effects on attention and memory for computer mediated messages . Unpublished dissertation, Indiana University , Bloomington . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sheer , V. C. , & Cline , R. J. W. ( 1995 ). Individual differences in sensation seeking and sexual behavior: Implications for communication intervention for HIV/AIDS prevention among college students . Health Communication , 7 , 205 – 223 . Google Scholar Crossref Search ADS WorldCat Shiffrin , R. M. , & Schneider , W. ( 1977 ). Controlled and automatic human information processing: II. Perceptual learning, automatic attending and a general theory . Psychological Review , 84 , 127 – 189 . Google Scholar Crossref Search ADS WorldCat Thelen , E. , & Smith , L. B. ( 1994 ). A dynamic systems approach to the development of cognition and action . Cambridge, MA : MIT Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Thorson , E. , & Lang , A. ( 1992 ). The effects of television videographics and lecture familiarity on adult cardiac orienting responses and memory . Communication Research , 19 , 346 – 369 . Google Scholar Crossref Search ADS WorldCat Thorson , E. , Reeves , B., & Schleuder , J. ( 1986 ). Attention to local and global complexity in television messages. In M. L. McLaughlin (Ed.), Communication yearbook 10 (pp. 366 – 383 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Tipper , S. P . ( 1991 ). Less attentional selectivity as a result of declining inhibition in older adults . Bulletin of the Psychonomic Society , 29 ( 1 ), 45 – 47 . Google Scholar Crossref Search ADS WorldCat Wickens , C. D . ( 1984 ). Processing resources in attention. In R. Parasuraman & D. R. Davies (Eds.), Varieties of attention (pp. 63 – 99 ). Orlando, FL : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Wise , K. , & Lang , A. ( 2001 ). Cardiac orienting and recognition for text appearing on a computer screen . Paper presented at the International Communication Association , Washington, DC . Wood , P. B. , Cochran , J. K., Pfefferbaum , B., & Arneklev , B. J. ( 1995 ). Sensation-seeking and delinquent substance use: An extension of learning theory . Journal of Drug Issues , 25 ( 1 ), 173 – 193 . Google Scholar Crossref Search ADS WorldCat Yegiyan , N. , Bradley , S. D., & Lang , A. ( 2005 ). Approach or avoid: How motivation type affects the processing of risky information . Paper presented to the Information Systems Division of the International Communication Association , New York . Yoon , K. , Bolls , P., & Lang , A. ( 1998 ). The effects of arousal on liking and believability of commercials . Journal of Marketing Communication , 4 , 101 – 114 . Google Scholar Crossref Search ADS WorldCat Zechmeister , E. B. , & Nyberg , S. E. ( 1982 ). Human memory: An introduction to research and theory . Monterey, CA : Brooks/Cole . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association

Stephenson, Michael, T.;Southwell, Brian, G.

doi: 10.1111/j.1460-2466.2006.00282.xpmid: N/A

Abstract The Activation Model of Information Exposure highlights the potential for individual differences in arousal in response to information, as well as the consequences of these patterns for information processing and seeking. Over the past 2 decades, the theoretical approach has generated considerable research in health communication. Most applications, however, have focused on substance use among adolescents and young adults. In this article, we assess the relevance of the activation approach for cancer communication. Although a wide range of communication efforts related to cancer prevention and treatment stand to benefit from acknowledgement of individual differences in optimal levels of arousal, we also acknowledge issues and challenges that remain for work on the Activation Model and sensation seeking. In reaching this conclusion, we explore some limitations of the Activation Model in its current form and point to new directions for future research. Although a number of theoretical models provide a foundation for explaining and predicting human behavior, the majority are cognitive models that assume a reasoning decision maker (Freimuth, 1992). These models explain behavior in terms of an individual’s beliefs, perceptions of threat and vulnerability, normative influences, or response efficacy (see Fishbein & Cappella, 2006). Such models, however, generally do not serve the communication researcher who has specific questions about the influence of the vast array of structural message features (i.e., audio and visual components) on human behavior. One theoretical approach that does offer help in this regard, however, is the Activation Model of Information Exposure (Donohew, Lorch, & Palmgreen, 1998; Donohew, Palmgreen, & Duncan, 1980). The Activation Model focuses on the relationship between one’s need for stimulation and the likelihood that a message will attract and maintain one’s attention. Recent communication research using the Activation Model has focused largely on sensation seeking—a personality trait with biological origins that we discuss in detail later—and on determining how television messages can be designed in a way that attracts and maintains the attention of those with high–sensation seeking tendencies. Additionally, much of the research based on the Activation Model to date focuses on adolescent drug prevention (and, to a lesser extent, risky sexual behaviors). Only recently have researchers begun to consider the concepts of the Activation Model as relevant to cancer communication more broadly, leaving open many fruitful paths to explore. In this article, we intend to facilitate use of the Activation Model and sensation seeking for cancer prevention and control. Our goals are fourfold: to review the fundamental assumptions of the Activation Model, to review extant research on mediated messages and sensation seeking on which the Activation Model is based, to discuss how the Activation Model and the sensation-seeking concept are relevant to cancer prevention and control, and to offer suggestions for further development of the Activation Model. The Activation Model for Information Exposure The Activation Model for Information Exposure focuses squarely on explaining the relationship between one’s need for stimulation and one’s attention to a mediated message. Maintaining that an individual’s need for stimulation is a function of his or her biological composition (Donohew et al., 1998, for discussion), the deductive-nomological Activation Model (Donohew & Palmgreen, 2003) is driven by two primary assumptions. First, “human beings have individual levels of need for activation or arousal at which they are most comfortable” (Donohew et al., 1998, p. 457; Donohew, Palmgreen, Zimmerman, Harrington, & Lane, 2003). Second, if this ideal level of stimulation is not maintained, individuals may (consciously or unconsciously) seek alternate forms of activation. These two assumptions imply a process illustrated in Figure 1. Figure 1 Open in new tabDownload slide The Activation Model of Information Exposure. Figure 1 Open in new tabDownload slide The Activation Model of Information Exposure. According to the model, when an individual is exposed to a message, the individual will be attracted to it and continue attending if the arousal generated by the message is consistent with the arousal desired by the individual. When arousal generated and arousal desired are consistent, the individual experiences positive affect and continues exposure to the message (e.g., Zillmann, 1991, 2000; Zillmann & Bryant, 1985). In a different situation, a message may be overly stimulating and exceed one’s optimal level of arousal. According to the Activation Model, the individual will experience negative affect, choose not to attend to the overstimulating message, and seek less arousing content. Alternatively, a message may fail to be stimulating enough such that the individual’s arousal level is not achieved or maintained. This individual will also not attend to the message, but instead search for content that is more stimulating and thereby consistent with the individual’s arousal needs, all else being equal. The Activation Model’s assumption, that individuals have levels of arousal at which they are most comfortable, stems from broader theories on optimal stimulation levels (Hebb, 1955), sources that provide stimulation (Berlyne, 1960), and individual differences in the response to stimuli (Zuckerman, 1969, 1979). These perspectives suggest that individual differences in one’s biological composition explain why some individuals prefer more stimulation than others do. Zuckerman (1994), for example, has argued that “the appetite for arousing stimulation and experience, whether direct or vicarious, is based in significant part on biological mechanisms, and individual differences in this appetite are based on variations in the underlying biological mechanisms as well as the outcomes of experience associated with such stimuli” (p. 174). Zuckerman’s (1979, 1984) emphasis on underlying biological mechanisms is also empirically supported. Zuckerman claims, for example, that one source of reward for those who need higher levels of stimulation and activation is the arousal of subcortical limbic systems, particularly involving the catecholamine neurotransmitters dopamine and norepinephrine (see also Bardo, Donohew, & Harrington, 1996; Bardo & Mueller, 1991; Zuckerman, 1994). This is consistent with the finding that dopamine, the brain’s pleasure neurotransmitter, regulates different aspects of activation (Le Moal & Simon, 1991), with the mesolimbocortical dopamine neurons actively regulating incentive reward motivation and avoidance responses (Rammsayer, 1997, 2004). Generally speaking, dopamine level is associated with one’s level of arousal from stimuli in that the mesolimbic dopamine pathway, when active, chemically reinforces arousal-producing behavior (Netter, Hennig, & Roed, 1996; Netter & Rammsayer, 1991). Although the Activation Model is focused on the match between one’s optimal arousal level and stimulation provided by media content, of course this match is not the only reason individuals will attend to a message. A reasonable critique of the Activation Model, in fact, is that by focusing only on need for activation, the model often does not explain the majority of the variance in attention to a message. Indeed, Donohew, Finn, and Christ (1988) acknowledge that other factors may also draw attention to a message, including the fact that “individuals may have a cognitive reason for exposing themselves to a particular stimulus, such as a source of information or a game” (p. 311). Still, they maintain that automatic processes that lead one to gravitate to their optimal arousal level—in search of chemical reward (Bardo et al., 1996)—justify acknowledgement of the Activation Model as an important part of the array of explanations for media use.1 A related problem with this model lies in the difficulty of pinpointing precisely what “optimal” means, in an absolute sense, for each individual. Assuming that “optimal” can be defined in absolute terms, how does one go about identifying study participants’ optimal level in an efficient and accurate manner? Moreover, for communication scholars, what combination of structural and content features of a message would be necessary to meet or exceed that optimal level? Would researchers seeking to optimize arousal need a unique combination of structural and message features for each and every study participant in each and every conceivable context? Practically speaking, these issues are likely to remain unresolved for the near future. Despite these problems, communication researchers nonetheless have profitably investigated differences in arousal need by looking at relative comparisons. To date, the most commonly employed approach to studying optimal arousal levels, as they relate to engagement with media content, is to measure the aforementioned (and later discussed) variable of sensation seeking and then to categorize individuals broadly as relatively higher or lower in sensation-seeking tendencies. As we will discuss, individuals with higher sensation-seeking tendencies both desire and tolerate a host of behaviors (so as to satisfy their relatively greater need for arousal) than those low in sensation seeking find to be overstimulating. What exactly is sensation seeking, then, and how does it provide a tool for health communication? Sensation seeking Sensation seeking can be defined as “the seeking of varied, novel, complex, and intense sensations and experiences, and the willingness to take physical, social, legal, and financial risks for the sake of such experiences” (Zuckerman, 1994, p. 27). In light of the connection between such behavioral tendencies and the documented biochemical dynamics outlined earlier, “nearly all” researchers of sensation seeking have chosen to “take a robustly biological approach” (Furnham, 2004, p. 170), believing that these biological influences on our personality are mediated primarily by dopamine (Netter & Rammsayer, 1991; Rammsayer, 2004). Greater tendency to seek sensation, simply stated, is often associated with lower default levels of dopamine. This tendency can be linked to thinking about appetitive and aversive response systems more broadly. Following Cacioppo (Cacioppo & Bernston, 1994; Ito, Cacioppo, & Lang, 1998), Lang, Shin, and Lee (2005) have extended the conceptualization of sensation seeking to note that those high in sensation seeking appear to have an overactive appetitive system (i.e., a positivity offset), and a weakly active aversive system (i.e., a negativity bias), and that other combinations of appetite and aversion may exist (see Lang, 2006). These biochemical differences are believed to be at the root of one’s motivation to engage in stimulating and novel behavior as engagement in novel behaviors or with exciting stimuli elicit a “transient and rapid surge” of dopamine and “points to an enhanced functioning of the mesolimbic DA [dopamine] reward pathway in the brain of high novelty responders” (Donohew, Bardo, & Zimmerman, 2004, p. 227). Engaging in these thrilling or novel experiences activates the release of dopamine and, along with serotonin and norepinephrine, provides the chemical reward that produces a positive emotional response. Based on adolescent twin studies, some scholars now believe that sensation seeking, as a trait, also is at least partially heritable (Fulker, Eysenck, & Zuckerman, 1980; Hur & Bouchard, 1997; Koopmans, Boomsma, Heath, & van Doornen, 1995; Koopmans, van Doornen, & Boomsma, 1997), implying that offspring may show similar risk-taking behaviors as their parents, all else being equal. This possibility is consistent with genetic work on dopamine. The recent genetic encoding of the dopamine D4 receptors and the connection of the 7-repeat D4 receptor allele to the related concept of novelty seeking (Benjamin et al., 1996; Ebstein et al., 1996; but see Kluger, Siegfried, & Ebstein, 2002; > Schinka, Letsch, & Crawford, 2002), for example, underscores the possibility of a link between generations. This connection suggests that need for activation and sensation seeking may be, at least in part, genetically driven to produce underlying biological differences in need for activation. Sensation seeking was originally measured using four underlying subscales—disinhibition, thrill and adventure seeking, experience seeking, and boredom susceptibility (Zuckerman, 1979, 1994)—that underpinned Zuckerman’s 40-item SSS-Form V (Zuckerman, Eysenck, & Eysenck, 1978). More recent conceptualizations offer more brevity (Arnett, 1994; Hoyle, Stephenson, Palmgreen, Lorch, & Donohew, 2002; Stephenson, Hoyle, Slater, & Palmgreen, 2003) and address possible age-related variance in sensation seeking (Russo & Beidel, 1993; Russo et al., 1991). Some scholars have argued recently that sensation seeking can be reconceptualized as a super trait called impulsive sensation seeking,2 defined as “a tendency to act quickly without planning and a general need for novelty, thrills and excitement” (Joireman & Kuhlman, 2004, p. 63; see also Zuckerman & Kuhlman, 2000; Zuckerman, Kuhlman, Joireman, Teta, & Kraft, 1993). The new connection to impulsivity reflects, to some degree, the earlier conceptualization of disinhibition. Whether one is discussing impulsive sensation seeking or the more conventional construct measured with a handful of items focused on, for example, one’s tendency to have unpredictable friends or new and exciting experiences, the core emphasis on one’s tendency to seek stimulation, even in the face of challenges, nonetheless remains. The divergent experiences sought by high- and low-sensation seekers are well documented. Those reporting high-sensation seeking tendencies are more likely to engage in thrill-seeking activities such as bungee jumping, parachuting, hang gliding, rock climbing, fast and risky driving (including driving while intoxicated), gambling, and attending X-rated movies and horror films (Roberti, 2004; Zuckerman, 1994; Zuckerman & Kuhlman, 2000). These activities apparently provide stimulation that can help a person to achieve a preferred level of arousal (Bardo et al., 1996; > Donohew et al., 2004). Need for sensation also translates into increased levels of risky behaviors directly linked to health. Specifically, sensation seeking is positively related to the use of drugs, both in amount and frequency of use (Donohew, 1990; Martin et al., 2004; Newcomb & Felix-Ortiz, 1992); alcohol use (Stein, Newcomb, & Bentler, 1994); heavy alcohol consumption (Zuckerman & Kuhlman, 2000); and high-risk sexual behavior such as unprotected sex with multiple partners (Ball & Schottenfeld, 1997; Donohew et al., 2000; Xiaoming et al., 2000; Zuckerman, 1994). Across the spectrum of risky behaviors, “it seems that the common denominator of the sensations attractive to sensation seekers is that they all produce transient spurts of physiological arousal” (Gomà-i-Friexanet, 2004, p. 187). This link not only signals the utility of sensation seeking as a tool for refining antidrug efforts, as it has been used in the past, but also, as we will discuss later, informs the application of sensation seeking to cancer communication efforts. Targeting sensation seekers with mediated messages The same motivation that drives sensation seekers to search for and participate in thrill-seeking behaviors also extends to their preferences for media and message exposure, with significant implications for the prevention of risky behaviors. Palmgreen and Donohew (2002) discussed these issues in detail and outlined a four-pronged media strategy called SENTAR (sensation-seeking targeting) for reaching sensation seekers effectively with mediated messages. At the heart of their suggestion is the belief that if we know that sensation seekers are likely to engage in a risky behavior, and we also know what type of media content these sensation seekers enjoy, then we have a significant advantage in both producing arousing stimuli and in deciding when and where to place communication campaign materials developed with sensation seekers’ preferences in mind. On this plane, Palmgreen and his colleagues (Everett & Palmgreen, 1995; Palmgreen, Stephenson, Everett, Basehart, & Francies, 2002) developed the concept message sensation value (MSV) to identify the characteristics of video-based messages that have the greatest potential for meeting the optimal stimulation needs of high-sensation seekers. A host of structural message features, for example, may contribute to the sensation value of a message. These features include, among others, multiple and quick camera cuts, special visual effects like morphing or unusual camera angles, zooms, close-ups, and suspenseful or intense sound saturation (Donohew, Lorch, & Palmgreen, 1991; Morgan, Palmgreen, Stephenson, Lorch, & Hoyle, 2003). Ads classified by judges as high in MSV typically contain more of these message features. High-MSV ads gain attention from, and are preferred by, high-sensation seekers because, in line with the Activation Model, these message features provide a level of stimulation that addresses sensation seekers’ need for arousal. Low-MSV ads, in contrast, provide less than the optimal arousal level for high-sensation seekers. Consequently, high-sensation seekers give little attention to these ads (Lorch et al., 1994) and begin searching for more stimulating media content (or other activity). People who are relatively less prone to sensation-seeking behavior are far less attracted to arousing message features, often preferring content that seems familiar or calming (Donohew et al., 1991). Like their counterparts, though, when the level of stimulation is too high or too low, low-sensation seekers also experience negative affect, pay less attention, and seek out an alternative message or medium (Lorch et al., 1994). Although MSV refers to the objective features of a message, viewer reactions to these message features indicate perceived message sensation value (see Morgan et al., 2003, or Palmgreen et al., 2002, for an extended discussion of these twin concepts). Comprehensive reviews of efforts to use sensation seeking and MSV research to prevent substance use are available (Donohew et al., 1998; Palmgreen & Donohew, 2002), but it is useful to provide brief examples of such work. For instance, laboratory research by Everett and Palmgreen (1995) on anticocaine public service announcements, for example, concluded that high–sensation value anticocaine ads were more effective than low–sensation value ads in enhancing cued and uncued recall, promoting anticocaine attitudes, and reducing intentions to try cocaine among high-sensation seekers. A different study by Lorch et al. (1994) used an eyes-on-screen measure of attention to investigate the role of sensation value in both program context and ads. They found that high–sensation seeking young adults not only paid more attention to high–sensation value programs (vs. low–sensation value programs), they also paid greater attention to high–sensation value antidrug public service announcements embedded in these shows. Most recently, in a mass media antidrug campaign focused on teenage marijuana use, researchers produced high–sensation value television ads and targeted them to sensation seekers. The estimated drop in the past-month marijuana use among high-sensation seekers was 27% in Knoxville, Tennessee, and 36% in Lexington, Kentucky (Palmgreen, Donohew, Lorch, Hoyle, & Stephenson, 2001, 2002). Cancer communication and the Activation Model Despite the link between illicit drug use and sensation seeking, there is little theoretical reason to think that drug use is the only area of behavior to which we can prospectively apply the approach outlined here. Cancer communication efforts should benefit from this approach in a variety of ways. In terms of reduction of high-risk behaviors, at least three behaviors—tobacco use, alcohol use, and unprotected sex—are both directly relevant to cancer prevention and control and also bear a relationship to sensation seeking. Moreover, on a different plane, the activation approach also in some cases points to different audience targeting possibilities than those conventionally associated with past activation research, especially for behaviors related to diet, exercise, and screening adherence. Alcohol and tobacco use We can first focus on alcohol use, which has been linked to cancer of the esophagus and of the breast (Razvodovsky, 2003) as well as to oral cavity cancers in general (Corrao, Bagnardi, Zambon, & LaVecchia, 2004) and tobacco use, which has been linked to lung cancer (Das, 2003). In addition to the association between each of these behaviors and various cancers, it appears that a person’s sensation-seeking tendency is predictive of whether a person engages in these behaviors as well as of positive perceptions of these behaviors. This relationship is not surprising in light of the promise of sensory experience often associated with such smoking and drinking, but nonetheless warrants our attention as we build a case for the relevance of the Activation Model for cancer communication. Given the relationship that sensation seeking holds with illicit drug use, it is logical that sensation seeking also would positively predict tobacco and alcohol use, and evidence supports this idea (Stephenson & Helme, 2005). Frankenberger (2004), for example, found a positive relationship between sensation seeking and smoking experience among 14- to 18-year-old survey participants. Outcomes by Stephenson et al. (2003) and Hoyle et al. (2002) are similar to Frankenberger’s (2004), with zero-order correlations between tobacco use and sensation seeking almost mirroring the relationships between marijuana use and sensation seeking. With regard to alcohol use, Ratsma, van der Stelt, Schoffelmeer, Westerveld, and Gunning (2001) have reported that sensation seeking predicts risk of alcoholism, whereas others have found a similar relationship of sensation seeking to alcohol consumption (Stein et al., 1994; Stephenson et al. (2003); Zuckerman & Kuhlman, 2000). Stephenson et al. (2003), as an example, report that past-month and lifetime alcohol use was correlated with four different measures of sensation seeking displaying zero-order correlations between.23 and.35. What is less clear at this stage is whether the relationship between sensation seeking and substance use is a direct function of the chemical reward of substance use itself, such as the mood-elevation characteristics of nicotine, or instead is somehow a function of the novel experience (i.e., forbidden, illegal, exciting activity either socially or personally) offered by tobacco or alcohol experimentation. Dinn, Ayciceqi, and Harris (2004), for example, found no support for a self-medication explanation for tobacco use, which suggests that other types of perceptions about the tobacco use experience might hold the key to enticement of high-sensation seekers. Moreover, Zuckerman (1994) maintains that substance use is initially motivated by curiosity and sustained either by the drug-induced pleasure of its use or by addiction. In addition to reporting a positive correlation between sensation seeking and alcohol consumption among college students, Forsyth and Hundleby (1987) reported that higher sensation seekers were more likely to drink when they were bored. Hence, the causal mechanism linking alcohol and tobacco use to sensation seeking is not entirely clear, but it nonetheless appears an association exists between sensation seeking and substance use with distinct implications for cancer communication efforts. Some researchers have already begun to seize upon these empirical relationships to pursue relevant cancer communication research. Niederdeppe (2005) examined the message features of ads used in various waves of the Florida “truth” Anti-tobacco media campaign (e.g., Sly, Heald, & Ray, 2001). In that work, for example, Niederdeppe found that some specific message features often associated with MSV apparently invited increased processing of the “truth” ads among 12- to 18-year-old audience members. Meanwhile, funded by the State of Colorado Tobacco Research Program, Helme and his associates are studying the persuasive effects of high–sensation value tobacco prevention ads (Helme, Donohew, Baier, & Zittleman, in press). Additionally, Lerman (personal communication, November 2003) has focused on smoking cessation (rather than prevention of trial) and currently is evaluating the impact of ads that vary in MSV on current adult smokers. Stages of smoking, including experimentation, initiation, and cessation, each appear to be associated with different levels of nicotine dependency, which may suggest a new opportunity for use of the Activation Model as a stage-specific strategy. Unprotected sex Beyond substance use, sensation seeking also holds a particularly interesting relationship to sexual behavior without protection, which is a risk factor for cervical cancer (American Cancer Society, 2003; Bosch et al., 1995). It appears that those high in sensation seeking often are tempted not simply to engage in sexual activity, but to do so in high-risk ways, for example, with relatively unknown or multiple partners or without condoms. Sheer and Cline (1994), for example, have reported a positive relationship between sensation-seeking predisposition and both the number of partners and incidence of unprotected sex reported by college students. McCoul and Haslam (2001) also found that a measure of sensation seeking was associated with the number of sex partners reported by heterosexual men. It also appears that tendencies toward substance use and high-risk sexual behavior covary. In particular, alcohol use and risky sex appear to be related. Kalichman and Cain (2004), in fact, see sensation seeking as a link holding together a constellation of risky behaviors that include alcohol use and unprotected sex. Henderson et al. (2005) offer some insight as to a mechanism for this relationship as they recently found a main effect of sensation seeking on judgments of potential sexual partner attractiveness and likelihood of desire to date a hypothetical partner. Importantly, compared with low-sensation seekers, high-sensation seekers also tended to rate potential partners as less risky. Not only are hypothetical sexual partners valued by high-sensation seekers because of their simple novelty, then, but it also appears that high-sensation seekers may underestimate the health risk involved in such encounters. Researchers also have begun to apply Activation Model concepts to the realm of safe sex promotion, though most such applications have focused more explicitly on preventing HIV infection or unintended pregnancy than on the ultimate goal of reducing cervical cancer risk to date. Nonetheless, as Donohew et al. (2000) note, the clear tendency of high-sensation seeking individuals to pursue high-risk sexual encounters, in combination with knowledge gained about the potential mechanisms underlying this relationship, makes sensation seeking a useful targeting variable. Those likely to take such risks in the bedroom or elsewhere are likely to share specific medium and message preferences. Physical activity In addition to helping to reduce substance use and unprotected sex, other opportunities to apply the Activation Model to cancer communication exist. Specifically, exercise behaviors offer a relevant site for intervention with implications for cancer prevention. Physical activity plays an important role in maintaining a healthy weight and we know increased physical activity has long-term implications for reducing risk for cancer (Knoops et al., 2004; Slattery, 2004; Zhang, Xie, Lee, & Binns, 2004). There are at least two ways we might consider sensation seeking in the arena of physical activity interventions. Some researchers have found an association between sensation seeking and physical activity (see Sallis, Prochaska, & Taylor, 2000). In light of that idea, an opportunity exists for interventions focused on increasing physical activity by attempting to reach those who are lower in sensation seeking tendency. At the same time, we also might actively encourage physical activity among high-sensation seekers as well, building on earlier work by Palmgreen et al. (1995), which promoted thrill-seeking activities as alternatives to substance use. Issues and challenges facing future Activation Model research By heeding the Activation Model perspective, we can usefully acknowledge the idea that the formal features of a message, such as editing patterns or sound effects, and the media placement of a message matter in gaining and maintaining the attention of people at risk for certain cancers. With that noted, however, we ought to move beyond the simple suggestion that the Activation Model be included in a list of theoretical perspectives available to campaign designers and evaluators. Ultimately, we should be able to point to linkages between the Activation Model and theories of behavioral decision making. We also should strive to improve our methods for studying these linkages. Although it appears that sensation-seeking tendency generally predicts evaluative preference for high-sensation value messages, we know far less about the main effects of MSV components on outcomes such as memory encoding. On the one hand, we noted above some instances in which MSV apparently encouraged processing and unaided recall, especially among high-sensation seekers (Everett & Palmgreen, 1995). Some recent work in this area, however, also suggests that there might be important general limits on human ability for retention of such messages. Southwell (2005) and Southwell and Lee (2004), for example, illustrate instances in which advertisements that employ complex editing appear to hinder recognition memory. Some of the content editing in those studies is of direct relevance for our discussion because it both contributed to the high MSV of the content in question as Morgan et al. (2003) would define it and, at the same time, also introduced time and space jumps that are more mentally taxing than a continuous sequence of scenes. For example, a direct cut from one frame depicting a person in a room at one point in time to a frame showing the same person in an entirely different setting 2 years later likely introduces new information to process. In both cases, Southwell hypothesized (and found support for) a negative effect of exposure to such ads on memory. The fact that ads with many such cuts were also the least likely to be remembered, in turn, suggests that MSV might not always equal memory value. These findings resonate with recent work by Lang and colleagues on the impact of tabloid production features and postproduction techniques on processing and memory for television news (Grabe, Lang, & Zhao, 2003; Lang, Potter, & Grabe, 2003). Specifically, there exists the distinct possibility that some of the so-called bells and whistles popular with television news editors to attract attention, such as sudden noises or complex frame transitions, may undermine audience memory for news content that is already arousing in its own right (e.g., simply because of the topic of the story). Consequently, a question worth further investigation is whether MSV invites unintended consequences for memory for materials presented in a high MSV message. If messages with high levels of stimulation draw the attention of targeted audiences but do so at the cost of long-term memory for the message’s content, then high-MSV messages might undermine there own success—initial attention that distracts processing from important aspects of message content. If this is the case, then questions remain about the mechanisms that mediate any relationship between MSV and long-term behavior change. We also do not yet know enough about the mechanisms that explain the impact of sensation seeking on short-term message evaluation, though Stephenson (Stephenson, 2002, 2003; Stephenson & Palmgreen, 2001) has examined the processing of high–sensation value antidrug ads and found a variety of patterns that are suggestive. In these studies, increased MSV actually encouraged the processing of antimarijuana ads by all viewers, not just those labeled as high-sensation seekers. More importantly, however, reported emotional responses to the antimarijuana ads varied considerably between high- and low-sensation seekers, and the mechanisms mediating the link between ad exposure and marijuana opposition was considerably different for high- and low-sensation seekers (Stephenson, 2002, 2003). For example, in one of those studies (Stephenson, 2003), those higher in sensation seeking experienced sympathetic distress in response to high-MSV ads, a finding we can potentially attribute to high-sensation seekers identifying with advertisement characters that were depicted as experiencing negative drug use consequences. High-sensation seekers also may have related their own experiences to those presented in the ads and judged the experiences to be consistent with their own. In light of these ideas, it is perhaps logical then that reported sympathetic distress, in turn, was positively related to postexposure opposition to marijuana for those high in sensation seeking. At the same time, low-sensation seeking participants did not demonstrate a positive link between reported emotional response and marijuana opposition. Perhaps these viewers identified with the characters in the ads less than their high–sensation seeking peers and instead focused more on the simple arguments presented against use of the featured illicit substance. An additional and unexplained issue from existing research is the possibility that MSV can lead to distraction that, in turn, invites peripheral message acceptance among high-sensation seekers. Stephenson (Stephenson, 2003; Stephenson & Palmgreen, 2001) has argued, for example, that high-sensation value message characteristics may distract from counterarguing among high-sensation seekers. This assertion awaits empirical verification. Other scholars also are working to connect the Activation Model to broader thinking about persuasion. Harrington, Lane, Donohew, and Zimmerman (2006), for example, offer perhaps the first major attempt to integrate core Activation Model ideas with the major dual-process models of persuasion (e.g., Petty & Cacioppo, 1986). Harrington et al. (2006) identify the gap in the persuasion process left by relying solely on the Activation Model suggesting that, beyond attention, cognitive forces related to message evaluation must become operative for persuasion to occur. Harrington et al. introduce message cognition value as a counterpart to MSV, whereas optimal level of cognitive engagement is introduced as a complement for optimal level of arousal. These concepts hinge on the more well-known need for cognition variable from the Elaboration Likelihood Model. The implication for communication interventions is that, in some cases, messages will need to attract attention and encourage central or systematic processing in order to be maximally persuasive. Additional testing of this idea is needed (Harrington et al., 2003). Conclusions The Activation Model theorizes how stimulating and novel media messages affect viewers who vary in their need for arousal. In this article, our goal was to introduce this perspective to the realm of cancer communication. At the very least, the concept of sensation seeking is relevant for researchers and practitioners working on lung, breast, oral, and cervical cancers and we encourage efforts to apply these ideas to communication campaigns in those areas. The construct might also prove useful, in a different way, in reaching sedentary audiences. At the same time, we also note that questions about the Activation Model and sensation seeking remain unanswered and that there is need for methodological refinement. Regardless of challenges, though, what is clear is that we have at least one tool for categorizing people who both engage in behavior that puts them at risk for some cancers and enjoy certain types of media content. By incorporating measures of sensation seeking into our formative and summative evaluation work, we should be able to make progress in addressing potential and actual cancer. Acknowledgments We thank Joseph Cappella for the opportunity to write this piece and for his editorial assistance. We also thank Lewis Donohew, Philip Palmgreen, and Nancy Harrington for their assistance. Notes 1 " In fact, the Activation Model was introduced as an alternative to dominant cognitive consistency theories such as selective exposure approaches, which suggested that individuals sought exposure to, and retained information from, media sources that simply were consistent with their beliefs and attitudes (Festinger, 1957; Heider, 1958; Klapper, 1960). Its earliest applications assessed how different people process news stories that were seen by some as “too threatening or otherwise exciting” (Donohew et al., 1988, p. 206). 2 " Impulsive sensation seeking can be measured with a 17-item scale, but is divorced from its conceptual predecessor and contains no subscales. See Joireman and Kuhlman (2004) and Schmitz (2004) for an extensive comparison of an alternate five-factor personality model, in which impulsive sensation seeking resides, to other similar theories of personality. References American Cancer Society . ( 2003 ). What are risk factors for cervical cancer? Atlanta, GA : Author . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Arnett , J . ( 1994 ). Sensation seeking: A new conceptualization and a new scale . Personality and Individual Differences , 16 , 289 – 296 . Google Scholar Crossref Search ADS WorldCat Ball , S. A. , & Schottenfeld , R. S. ( 1997 ). A five-factor model of personality and addiction, psychiatric, and AIDS risk severity in pregnant and postpartum cocaine misusers . Substance Use and Misuse , 32 , 25 – 41 . Google Scholar Crossref Search ADS PubMed WorldCat Bardo , M. T. , Donohew , R. L., & Harrington , N. G. ( 1996 ). Psychobiology of novelty seeking and drug seeking behavior . Behavioural Brain Research , 77 , 23 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat Bardo , M. T. , & Mueller , C. W. ( 1991 ). Sensation seeking and drug abuse prevention from a biological perspective. In L. Donohew, H. E. Sypher, & W. J. Bukoski (Eds.), Persuasive communication and drug abuse prevention (pp. 195 – 207 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Benjamin , J. , Li , L., Patterson , C., Greenberg , B. D., Murphy , D. L., & Hamer , D. H. ( 1996 ). Population and familial association between the D4 dopamine receptor gene and measures of Novelty Seeking . Nature Genetics , 12 , 81 – 84 . Google Scholar Crossref Search ADS PubMed WorldCat Berlyne , D. E . ( 1960 ). Conflict, arousal, and curiosity . New York : McGraw-Hill . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Bosch , F. X. , Manos , M. M., Munoz , N., Sherman , M., Jansen , A. M., Peto , J., et al. . ( 1995 ). Prevalence of human papillomavirus in cervical cancer: A worldwide perspective . Journal of the National Cancer Institute , 87 , 796 – 802 . Google Scholar Crossref Search ADS PubMed WorldCat Cacioppo , J. T. , & Bernston , G. G. ( 1994 ). Relationship between attitudes and evaluate space: A critical review, with emphasis on the separability of positive and negative substrates . Psychological Bulletin , 115 , 401 – 423 . Google Scholar Crossref Search ADS WorldCat Corrao , G. , Bagnardi , V., Zambon , A., & LaVecchia , C. ( 2004 ). A meta-analysis of alcohol consumption and the risk of 15 diseases . Preventive Medicine , 38 , 613 – 619 . Google Scholar Crossref Search ADS PubMed WorldCat Das , S. K . ( 2003 ). Harmful health effects of cigarette smoking . Molecular & Cellular Biochemistry , 253 ( 1–2 ), 159 – 165 . Google Scholar Crossref Search ADS WorldCat Dinn , W. M. , Ayciceqi , A., & Harris , C. L. ( 2004 ). Cigarette smoking in a student sample: Neurocognitive and clinical correlates . Addictive Behaviors , 29 , 107 – 126 . Google Scholar Crossref Search ADS PubMed WorldCat Donohew , L . ( 1990 ). Public health campaigns: Individual message strategies and a model. In E. B. Ray & L. Donohew (Eds.), Communication and health: Systems and applications (pp. 136 – 152 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Donohew , L. , Bardo , M. T., & Zimmerman , R. S. ( 2004 ). Personality and risky behavior: Communication and prevention. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 223 – 245 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Donohew , L. , Finn , S., & Christ , W. ( 1988 ). The nature of news revisited: The roles of affect, schemas, and cognition. In L. Donohew, H. Sypher, & E. T. Higgins (Eds.), Communication, social cognition, and affect (pp. 195 – 218 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Donohew , L. , Lorch , E., & Palmgreen , P. ( 1991 ). Sensation seeking and targeting of televised anti-drug PSAs. In L. Donohew, H. E. Sypher, & W. J. Bukoski (Eds.), Persuasive communication and drug abuse prevention (pp. 209 – 226 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Donohew , L. , Lorch , E., & Palmgreen , P. ( 1998 ). Applications of a theoretic model of information exposure to health interventions . Human Communication Research , 24 , 454 – 468 . Google Scholar Crossref Search ADS PubMed WorldCat Donohew , L. , & Palmgreen , P. ( 2003 ). Constructing theory. In G. Stemple, D. Weaver, & G. C. Wilhoit (Eds.), Mass communication theory and research (pp. 111 – 128 ). Boston : Allyn & Bacon . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Donohew , L. , Palmgreen , P., & Duncan , J. ( 1980 ). An activation model of information exposure . Communication Monographs , 47 , 295 – 303 . Google Scholar Crossref Search ADS WorldCat Donohew , L. , Palmgreen , P., Zimmerman , R., Harrington , N., & Lane , D. ( 2003 ). Health risk takers and prevention. In D. Romer (Ed.), Reducing adolescent risk: Toward an integrated approach (pp. 165 – 170 ). Thousand Oaks, CA : Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Donohew , L. , Zimmerman , R. S., Cupp , P. S., Novak , S., Colon , S., & Abell , R. ( 2000 ). Sensation seeking, impulsive decision-making, and risky sex: Implications for risk-taking and design of interventions . Personality and Individual Differences , 28 , 1079 – 1091 . Google Scholar Crossref Search ADS WorldCat Ebstein , R. P. , Novick , O., Umansky , R., Priel , B., Osher , Y., Blaine , D., et al. . ( 1996 ). Dopamine D4 receptor (D4DR) exon III polymorphism associated with the human personality trait of Novelty Seeking . Nature Genetics , 12 , 78 – 80 . Google Scholar Crossref Search ADS PubMed WorldCat Everett , M. W. , & Palmgreen , P. ( 1995 ). Influences of sensation seeking, message sensation value, and program context on effectiveness of anticocaine public service announcements . Health Communication , 7 , 225 – 248 . Google Scholar Crossref Search ADS WorldCat Festinger , L . ( 1957 ). A theory of cognitive dissonance . Evanston, IL : Row, Peterson . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fishbein , M. , & Cappella , J. N. ( 2006 ). The role of theory in developing effective health communications . Journal of Communication , 56 , S1 – S17 . Google Scholar Crossref Search ADS WorldCat Forsyth , G. , & Hundleby , J. D. ( 1987 ). Personality and situation as determinants of desire to drink in young adults . International Journal of the Addictions , 22 , 653 – 669 . Google Scholar Crossref Search ADS PubMed WorldCat Frankenberger , K. D . ( 2004 ). Adolescent egocentrism, risk perceptions, and sensation seeking among smoking and nonsmoking youth . Journal of Adolescent Research , 19 , 576 – 590 . Google Scholar Crossref Search ADS WorldCat Freimuth , V . ( 1992 ). Theoretical foundations of AIDS media campaigns. In T. Edgar, M. A. Fitzpatrick, & V. Freimuth (Eds.), AIDS: A communication perspective (pp. 92 – 110 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fulker , D. W. , Eysenck , S. B. G., & Zuckerman , M. ( 1980 ). The genetics of sensation seeking . Journal of Personality Research , 14 , 261 – 281 . Google Scholar Crossref Search ADS WorldCat Furnham , A . ( 2004 ). Personality and leisure activity: Sensation seeking and spare-time activities. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 167 – 183 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gomà-i-Friexanet , M . ( 2004 ). Sensation seeking and participation in physical risk sports. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 185 – 201 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Grabe , M. E. , Lang , A., & Zhao , X. ( 2003 ). News content and form: Implications for memory and audience evaluations . Communication Research , 30 , 387 – 413 . Google Scholar Crossref Search ADS WorldCat Harrington , N. G. , Lane , D. R., Donohew , L., & Zimmerman , R. S. ( 2006 ). An extension of the activation model of information exposure: The addition of a cognitive variable to a model of attention . Media Psychology , 8 , 139 – 164 . Google Scholar Crossref Search ADS WorldCat Harrington , N. G. , Lane , D. R., Donohew , L., Zimmerman , R. S., Norling , G. R., An , J. H., et al. . ( 2003 ). Persuasive strategies for effective anti-drug messages . Communication Monographs , 70 , 16 – 38 . Google Scholar Crossref Search ADS WorldCat Hebb , D . ( 1955 ). Drives and the CNS (conceptual nervous system) . Psychological Review , 62 , 243 – 254 . Google Scholar Crossref Search ADS PubMed WorldCat Heider , F . ( 1958 ). The psychology of interpersonal relations . New York : Wiley . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Helme , D. W. , Donohew , R. L., Baier , M., & Zittleman , L. ( in press ). A classroom-administered simulation of a television campaign on adolescent smoking: Testing an activation model of information exposure . Journal of Health Communication . OpenURL Placeholder Text WorldCat Henderson , V. R. , Hennessy , M., Barrett , D. W., Curtis , B., McCoy-Roth , M., Trentacoste , N., et al. . ( 2005 ). When risky is attractive: Sensation seeking and romantic partner selection . Personality and Individual Differences , 38 , 311 – 325 . Google Scholar Crossref Search ADS WorldCat Hoyle , R. H. , Stephenson , M. T., Palmgreen , P., Lorch , E. P., & Donohew , L. ( 2002 ). Reliability and validity of scores on a brief measure of sensation seeking . Personality and Individual Differences , 32 , 401 – 414 . Google Scholar Crossref Search ADS WorldCat Hur , Y. , & Bouchard , T. J. Jr. ( 1997 ). Genetic influences on impulsivity and sensation-seeking . Behavior Genetics , 27 , 455 – 463 . Google Scholar Crossref Search ADS PubMed WorldCat Ito , I. A. , Cacioppo , J. T., & Lang , P. J. ( 1998 ). Eliciting affect using the International Affective Picture System: Trajectories through evaluative space . Personality and Social Psychology Bulletin , 24 , 855 – 879 . Google Scholar Crossref Search ADS WorldCat Joireman , J. , & Kuhlman , D. M. ( 2004 ). The Zuckerman-Kuhlman personality questionnaire: Origin, development, and validity of a measure to assess an alternative five-factor model of personality. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 49 – 64 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kalichman , S. C. , & Cain , D. ( 2004 ). A prospective study of sensation seeking and alcohol use as predictors of sexual risk behaviors among men and women receiving sexually transmitted infection clinic services . Psychology of Addictive Behaviors , 18 , 367 – 373 . Google Scholar Crossref Search ADS PubMed WorldCat Klapper , J. T . ( 1960 ). The effects of mass communications . New York : Free Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kluger , A. N. , Siegfried , Z., & Ebstein , R. P. ( 2002 ). A meta-analysis of the association between DRD4 polymorphism and novelty seeking . Molecular Psychiatry , 9 , 197 – 202 . OpenURL Placeholder Text WorldCat Koopmans , J. R. , Boomsma , D. I., Heath , A. C., & Van Doornen , L. J. P. ( 1995 ). A multivariate genetic analysis of sensation seeking . Behavior Genetics , 25 , 349 – 356 . Google Scholar Crossref Search ADS PubMed WorldCat Koopmans , J. R. , Van Doornen , L. J., & Boomsma , D. I. ( 1997 ). Association between alcohol use and smoking in adolescent and young adult twins: A bivariate genetic analysis . Alcoholism: Clinical and Experimental Research , 21 , 537 – 546 . OpenURL Placeholder Text WorldCat Knoops , K. T. , De Groot , L. C., Kromhout , D., Perrin , A. E., Moreiras-Varela , O., Menotti , A., et al. . ( 2004 ). Mediterranean diet, lifestyle factors, and 10-year mortality in elderly European men and women: The HALE project . Journal of the American Medical Association , 292 , 1433 – 1439 . Google Scholar Crossref Search ADS PubMed WorldCat Lang , A . ( 2006 ). Using the limited capacity model of motivated mediated message processing (LC4MP) to design effective cancer communication messages . Journal of Communication , 56 , S57 – S80 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Potter , R., & Grabe , M. E. ( 2003 ). Making news memorable: Applying theory to the production of local television news . Journal of Broadcasting & Electronic Media , 47 , 113 – 123 . Google Scholar Crossref Search ADS WorldCat Lang , A. , Shin , M., & Lee , S. ( 2005 ). Sensation seeking, motivation, and substance use: A dual system approach . Media Psychology , 7 , 1 – 29 . Google Scholar Crossref Search ADS WorldCat Le Moal , M. , & Simon , H. ( 1991 ). Mesocorticolimbic dopaminergic network: Functional and regulatory roles . Physiological Review , 71 , 155 – 234 . OpenURL Placeholder Text WorldCat Lorch , E. P. , Palmgreen , P., Donohew , L., Helm , D., Baer , S. A., & D’Silva , M. U. ( 1994 ). Program context, sensation seeking, and attention to televised anti-drug public service announcements . Human Communication Research , 20 , 390 – 412 . Google Scholar Crossref Search ADS WorldCat Martin , C. A. , Kelly , T. H., Rayens , M. K., Brogli , B., Himelreich , K., Brenzel , Al., et al. . ( 2004 ). Sensation seeking and symptoms of disruptive disorder: Association with nicotine, alcohol, and marijuana use in early and mid-adolescence . Psychological Reports , 94 , 1075 – 1082 . Google Scholar Crossref Search ADS PubMed WorldCat McCoul , M. D. , & Haslam , N. ( 2001 ). Predicting high risk sexual behaviour in heterosexual and homosexual men: The roles of impulsivity and sensation seeking . Personality and Individual Differences , 31 , 1303 – 1310 . Google Scholar Crossref Search ADS WorldCat Morgan , S. E. , Palmgreen , P., Stephenson , M. T., Lorch , E. P., & Hoyle , R. H. ( 2003 ). The relationship between message sensation value and perceived message sensation value: The effect of formal message features on subjective evaluations of anti-drug public service announcements . Journal of Communication , 53 , 512 – 526 . Google Scholar Crossref Search ADS WorldCat Netter , P. , Hennig , J., & Roed , I. S. ( 1996 ). Serotonin and dopamine as mediators of sensation seeking behavior . Neuropsychobiology , 34 , 155 – 165 . Google Scholar Crossref Search ADS PubMed WorldCat Netter , P. , & Rammsayer , T. H. ( 1991 ). Reactivity to dopaminergic drugs and aggression related personality traits . Personality and Individual Differences , 12 , 1009 – 1017 . Google Scholar Crossref Search ADS WorldCat Newcomb , M. D. , & Felix-Ortiz , M. ( 1992 ). Multiple protective and risk factors for drug use and abuse: Cross sectional and prospective findings . Journal of Personality and Social Psychology , 63 , 280 – 296 . Google Scholar Crossref Search ADS PubMed WorldCat Niederdeppe , J . ( 2005 ). Syntactic indeterminancy, perceived message sensation value-enhancing features, and message processing in the context of anti-tobacco advertisements . Communication Monographs , 72 , 284 – 300 . Google Scholar Crossref Search ADS WorldCat Palmgreen , P. , & Donohew , L. ( 2002 ). Effective mass media strategies for drug abuse prevention campaigns. In Z. Sloboda & W. Bukoski (Eds.), Effective strategies for drug abuse prevention (pp. 27 – 43 ). New York : Plenum Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Palmgreen , P. , Donohew , L., Lorch , E. P., Hoyle , R. H., & Stephenson , M. T. ( 2001 ). Television campaigns and adolescent marijuana use: Tests of a sensation seeking targeting . American Journal of Public Health , 91 , 292 – 296 . Google Scholar Crossref Search ADS PubMed WorldCat Palmgreen , P. , Donohew , L., Lorch , E. P., Hoyle , R. H., & Stephenson , M. T. ( 2002 ). Television campaigns and sensation seeking targeting of adolescent marijuana use: A controlled time-series approach. In R. C. Hornik (Ed.), Public health communication: Evidence for behavior change (pp. 35 – 56 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Palmgreen , P. , Lorch , E. P., Donohew , L., Harrington , N. G., D’Silva , M., & Helm , D. ( 1995 ). Reaching at-risk populations in a mass media drug abuse prevention campaign: Sensation seeking as a targeting variable . Drugs and Society , 8 , 29 – 45 . Google Scholar Crossref Search ADS WorldCat Palmgreen , P. , Stephenson , M. T., Everett , M. W., Basehart , J. R., & Francies , R. ( 2002 ). Perceived message sensation value (PMSV) and the dimensions and validation of a PMSV scale . Health Communication , 14 , 403 – 428 . Google Scholar Crossref Search ADS PubMed WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1986 ). Communication and persuasion: Central and peripheral routes to attitude change . New York : Springer-Verlag . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rammsayer , T. H . ( 1997 ). Are there dissociable roles of the mesostriatal and mesolimbocortical dopamine systems on temporal information processing in humans? Neuropsychobiology , 35 , 36 – 45 . Google Scholar Crossref Search ADS PubMed WorldCat Rammsayer , T. H . ( 2004 ). Extraversion and the dopamine hypothesis. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 409 – 427 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ratsma , J. E. , Van Der Stelt , O., Schoffelmeer , A. N. M., Westerveld , A., & Gunning , W. B. ( 2001 ). P3 event-related potential, dopamine D2 receptor A1 allele, and sensation-seeking in adult children of alcoholics . Alcoholism , 25 , 960 – 967 . Google Scholar Crossref Search ADS PubMed WorldCat Razvodovsky , Y. E . ( 2003 ). Aggregate level time series association between alcohol consumption and cancer mortality rate . Alcoholism , 39 , 11 – 20 . OpenURL Placeholder Text WorldCat Roberti , J. W . ( 2004 ). A review of behavioral and biological correlates of sensation seeking . Journal of Research in Personality , 38 , 256 – 279 . Google Scholar Crossref Search ADS WorldCat Russo , M. F. , & Beidel , D. C. ( 1993 ). Co-morbidity of childhood anxiety and externalizing disorders: Prevalence, associated characteristics, and validation issues . Clinical Psychology Review , 14 , 199 – 221 . Google Scholar Crossref Search ADS WorldCat Russo , M. F. , Lahey , B. B., Christ , M. A. G., Frick , P. J., McBurnett , K., Walker , J. L., et al. . ( 1991 ). Preliminary development of a sensation seeking scale for children . Personality and Individual Differences , 12 , 399 – 405 . Google Scholar Crossref Search ADS WorldCat Sallis , J. F. , Prochaska , J. J., & Taylor , W. C. ( 2000 ). A review of correlates of physical activity of children and adolescents . Medicine & Science in Sports & Exercise , 32 , 963 – 975 . Google Scholar Crossref Search ADS WorldCat Schinka , J. A. , Letsch , E. A., & Crawford , F. C. ( 2002 ). DRD4 and novelty seeking: Results of meta-analyses . American Journal of Medical Genetics , 114 , 643 – 648 . Google Scholar Crossref Search ADS PubMed WorldCat Schmitz , P. G . ( 2004 ). On the alternative five-factor model: Structure and correlates. In R. M. Stelmack (Ed.), On the psychobiology of personality: Essays in honor of Marvin Zuckerman (pp. 65 – 87 ). Ottawa, CA : Elsevier . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sheer , V. C. , & Cline , R. J. ( 1994 ). The development and validation of a model explaining sexual behavior among college students: Implications for AIDS communication campaigns . Human Communication Research , 21 , 280 – 304 . Google Scholar Crossref Search ADS WorldCat Slattery , M. L . ( 2004 ). Physical activity and colorectal cancer . Sports Medicine , 34 , 239 – 252 . Google Scholar Crossref Search ADS PubMed WorldCat Sly , D. F. , Heald , G. R., & Ray , S. ( 2001 ). The Florida “truth” anti-tobacco media evaluation: Design, first year results, and implications for planning future state media evaluations . Tobacco Control , 10 , 9 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat Southwell , B. G . ( 2005 ). Between messages and people: A multilevel model of memory for television content . Communication Research , 32 , 112 – 140 . Google Scholar Crossref Search ADS WorldCat Southwell , B. G. , & Lee , M. ( 2004 ). A pitfall of new media? User controls exacerbate editing effects on memory . Journalism & Mass Communication Quarterly , 81 , 643 – 656 . Google Scholar Crossref Search ADS WorldCat Stein , J. A. , Newcomb , M. D., & Bentler , P. M. ( 1994 ). Psychosocial correlates and predictors of AIDS risk behaviors, abortion, and drug use among a community sample of young adult women . Health Psychology , 13 , 308 – 318 . Google Scholar Crossref Search ADS PubMed WorldCat Stephenson , M. T . ( 2002 ). Sensation seeking as a moderator of the processing of anti-heroin PSAs . Communication Studies , 53 , 358 – 380 . Google Scholar Crossref Search ADS WorldCat Stephenson , M. T . ( 2003 ). Examining adolescents’ responses to antimarijuana PSAs . Human Communication Research , 29 , 343 – 369 . Google Scholar Crossref Search ADS WorldCat Stephenson , M. T. , & Helme , D. W. ( 2005 ). The effects of authoritative parenting on adolescent cigarette and marijuana use by high sensation seekers . Unpublished manuscript. Texas A & M , College Station, TX . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Stephenson , M. T. , Hoyle , R. H., Slater , M. D., & Palmgreen , P. ( 2003 ). Brief measures of sensation seeking for screening and large-scale surveys . Drug and Alcohol Dependence , 72 , 279 – 286 . Google Scholar Crossref Search ADS PubMed WorldCat Stephenson , M. T. , & Palmgreen , P. ( 2001 ). Sensation seeking, message sensation value, personal involvement, and processing of anti-drug PSAs . Communication Monographs , 68 , 49 – 71 . Google Scholar Crossref Search ADS WorldCat Xiaoming , L. , Stanton , B., Cottrell , L., Burns , J., Pack , R., & Kaljee , L. ( 2000 ). Patterns of initiation of sex and drug-related activities among urban low-income African American adolescents . Journal of Adolescent Health , 28 , 46 – 54 . OpenURL Placeholder Text WorldCat Zhang , M. , Xie , X., Lee , A. H., & Binns , C. W. ( 2004 ). Sedentary behaviours and epithelial ovarian cancer risk . Cancer Causes & Control , 15 , 83 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat Zillmann , D . ( 1991 ). Television viewing and physiological arousal. In J. Bryant & D. Zillmann (Eds.), Responding to the screen: Reception and reaction processes (pp. 281 – 303 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zillmann , D . ( 2000 ). Mood management in the context of selective exposure theory. In M. E. Roloff (Ed.), Communication yearbook 23 (pp. 103 – 123 ). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zillmann , D. , & Bryant , J. ( 1985 ). Affect, mood, and emotion as determinants of selective exposure. In D. Zillmann & J. Bryant (Eds.), Selective exposure to communication (pp. 157 – 190 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zuckerman , M . ( 1969 ). Variables affecting deprivation results. In J. P. Zubek (Ed.), Sensory deprivation: Fifteen years of research (pp. 47 – 84 ). New York : Appleton-Century-Crofts . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zuckerman , M . ( 1979 ). Sensation seeking: Beyond the optimal level of arousal . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zuckerman , M . ( 1984 ). Sensation seeking: A comparative approach to a human trait . Behavioral and Brain Sciences , 7 , 413 – 471 . Google Scholar Crossref Search ADS WorldCat Zuckerman , M . ( 1994 ). Behavioral expressions and biosocial bases of sensation seeking . Cambridge, UK : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zuckerman , M. , Eysenck , S., & Eysenck , H. J. ( 1978 ). Sensation seeking in England and America: Cross-cultural age and sex comparisons . Journal of Consulting and Clinical Psychology , 46 , 139 – 149 . Google Scholar Crossref Search ADS PubMed WorldCat Zuckerman , M. , & Kuhlman , S. ( 2000 ). Personality and risk-taking: Common biosocial factors . Journal of Personality , 68 , 999 – 1029 . Google Scholar Crossref Search ADS PubMed WorldCat Zuckerman , M. , Kuhlman , D. M., Joireman , J., Teta , P., & Kraft , M. ( 1993 ). A comparison of three structural models for personality: The big three, the big five, and the alternative five . Journal of Personality and Social Psychology , 65 , 757 – 768 . Google Scholar Crossref Search ADS WorldCat Author notes 3 " Both authors contributed equally to the preparation of the manuscript. © 2006 International Communication Association

Cappella, Joseph, N.

doi: 10.1111/j.1460-2466.2006.00293.xpmid: N/A

Abstract Three broad classes of theories about message effects in cancer control are presented in this special supplement to the Journal of Communication. These are behavior change, information processing, and message effects theories. All three types have implications for the design of messages for cancer control. The theories are not just different approaches to a complex problem but offer complementary perspectives on the effects of messages on audiences. This summary article explores why theory is so important to efficient research in message effects and speculates about the interrelation among behavior change, information processing, and message effects models. This special issue of the Journal of Communication resulted from papers submitted to a workshop entitled “Integrating Theories of Message Effects and Health Behavior Change to Improve Cancer Control” held at the Annenberg School for Communication at the University of Pennsylvania in November 2003. The conference was sponsored by the National Cancer Institute and cosponsored by the International Communication Association and the Annenberg Public Policy Center. The workshop grew from the recognition that advances in communication science for cancer control required not just reflection on theories about messages and their effects but integration across theories pertinent to message effects and across the many different kinds of scientists who conduct such research. The workshop’s goals were to investigate the integration of three classes of theories in their application to problems of the prevention of risky behavior and adoption of healthy behaviors. The three classes include (a) human information processing theories pertinent to behavior change; (b) behavior change theories describing rational, emotional, social, and personal predictors of healthy and risky behavior; and (c) message effect theories predicting the format and content of messages that produce effects on cognitive, attitudinal, and emotional outcomes. The workshop sought to create a dialogue among researchers in each of these areas to propel greater theoretical coherence and improve applications in health communication. The results of the workshop are the 13 papers that comprise this special issue of the Journal of Communication. In this overview article, my goal is to elaborate the connections among the theories and research presented and to raise some questions that are as yet unanswered about the design of messages for effective cancer control. To accomplish the first goal, it is useful to think of the various articles as representatives of certain categories such as behavior change, information processing, and message effects theories. Although the distinctions are sometimes blurred, the candidate articles within each group are described in Table 1. The articles from the workshop do not represent all theories and models of behavior change or information processing that are pertinent to message effects. But they are central theories. Similarly, a number of other approaches to message effects could have been included, for example, focusing on priming effects (Ybarra & Trafimow, 1998) or normative appeals (Cialdini, 2003). However, the articles in this issue represent some of the core approaches across the domain of message effects. Table 1 Groups of Theoretical Approaches Pertinent to Message Effects . Topic . Authors . Behavior change theories Integrated model of behavior change Fishbein and Cappella Behavioral beliefs in smoking initiation Krosnick et al. Information processing Activation model of information exposure Stephenson and Southwell Limited capacity model of motivated mediated message processing Lang Elaboration likelihood model Briñol and Petty Unimodel Kruglanski et al. Message effects Emotional appeals Dillard and Nabi Emotion functions Peters et al. Tailoring Rimer and Kreuter Narratives Green Frames (gain and loss) Rothman et al. Exemplars Zillman Systemic factors Cultural and social Viswanath and Emmons . Topic . Authors . Behavior change theories Integrated model of behavior change Fishbein and Cappella Behavioral beliefs in smoking initiation Krosnick et al. Information processing Activation model of information exposure Stephenson and Southwell Limited capacity model of motivated mediated message processing Lang Elaboration likelihood model Briñol and Petty Unimodel Kruglanski et al. Message effects Emotional appeals Dillard and Nabi Emotion functions Peters et al. Tailoring Rimer and Kreuter Narratives Green Frames (gain and loss) Rothman et al. Exemplars Zillman Systemic factors Cultural and social Viswanath and Emmons Open in new tab Table 1 Groups of Theoretical Approaches Pertinent to Message Effects . Topic . Authors . Behavior change theories Integrated model of behavior change Fishbein and Cappella Behavioral beliefs in smoking initiation Krosnick et al. Information processing Activation model of information exposure Stephenson and Southwell Limited capacity model of motivated mediated message processing Lang Elaboration likelihood model Briñol and Petty Unimodel Kruglanski et al. Message effects Emotional appeals Dillard and Nabi Emotion functions Peters et al. Tailoring Rimer and Kreuter Narratives Green Frames (gain and loss) Rothman et al. Exemplars Zillman Systemic factors Cultural and social Viswanath and Emmons . Topic . Authors . Behavior change theories Integrated model of behavior change Fishbein and Cappella Behavioral beliefs in smoking initiation Krosnick et al. Information processing Activation model of information exposure Stephenson and Southwell Limited capacity model of motivated mediated message processing Lang Elaboration likelihood model Briñol and Petty Unimodel Kruglanski et al. Message effects Emotional appeals Dillard and Nabi Emotion functions Peters et al. Tailoring Rimer and Kreuter Narratives Green Frames (gain and loss) Rothman et al. Exemplars Zillman Systemic factors Cultural and social Viswanath and Emmons Open in new tab These categories are offered not to suggest that behavior change and information processing theories have nothing to say about message design nor that message effects theories are driven more by practicalities of message design than by theoretical considerations. To the contrary, all the articles address message design in some (more and less specific) ways, and all the articles are grounded in established theory in some way. But the categories provide at least one way to think about a hierarchical organization for an enlarged approach—if not yet a theory—about the design of effective messages. In what follows, I will offer a series of summary statements about the 13 articles of this special issue. These statements may help us understand how the approaches described here contribute to the design of effective messages in cancer control. Theories of message effects, not just careful testing of messages, are necessary for effective cancer control Messages designed to achieve persuasive goals can be conceptualized in an infinite number of different ways. Too, the number of different message components and their combinations that could possibly affect a message’s overall effectiveness is virtually unfathomable. For example, consider designing a simple brochure to be placed in a doctor’s offices encouraging male patients over 40 years to obtain a regular prostate-specific antigen (PSA) test along with a digital rectal exam. In addition to questions about color, font, the use of pictures, reading levels, technical terms, length, and so on, the brochure’s creator must decide whether to use gain or less frames for presenting positive and negative consequences, personal stories, statistics, intense language, strong arguments and evidence, and a thousand other possible considerations. In the face of such complexity, why not simply test each message that might be used in a public health campaign, message by message, and population by population? Given an infinity of message features, a plausible case can be made for testing the effectiveness of each unique message in each context rather than developing theories that might provide insight across contexts. The obvious counterargument is inefficiency. Individualized message testing in the absence of theory to direct the testing and in the absence of controlled testing procedures produces little knowledge that can be transferred to other contexts. The goal of theory-driven message testing is to generate knowledge that can be transferred to parallel cancer control contexts. To find that posttest intentions to get a PSA test annually are higher after reading a brochure than before provides no information about what particular feature of the brochure produced elevated intention or whether it was some particular combination of features. Even if the testing procedure allows for isolation of the key message feature responsible for the favorable effects on intention, not knowing why the feature had its effects creates other inefficiencies that stymie the development of useful knowledge. For example, suppose two brochures are compared that differ only in the use of gain versus loss frames, with the loss-framed version producing greater elevation in the intention to screen. If we also do not know why the loss frame was more effective, we are at a loss for connecting findings with this message feature to other findings on different features that employ the same mechanism. So if the effects are mediated by emotional processes linked to the arousal of threat and efficacious means of alleviating that threat, then the effects of gain and loss frames might be understood as operating in a manner that is parallel to other message features that activate withdrawal emotions such as fear, anxiety, and apprehension. The results would be a much more parsimonious way to think about the very complex—indeed infinitely complex—world of message features and their effectiveness. In short, message effects theories must be the choice over simple message testing approaches for very practical reasons. We want our research to be both efficient and effective. Theory-driven research about message effects allows us to isolate which message features are most responsible for the consequences of a given message and, when the mechanisms are understood as well, to connect message features that would otherwise be seen as completely disconnected as operating through the same theoretical mechanisms. In the articles in this issue, three classes of theories are pertinent to message effects—behavior change theories, information processing theories, and message effects theories. We next explore some of the linkages among these groups of theories. Behavior change and information processing theories are complementary Behavior change and information processing theories are not simply two different traditions of theorizing about attitude processes but are complementary, offering information to message designers, which in combination answers questions the other approach ignores. They do not simply answer different questions, they answer complementary questions. Behavior change theories answer questions about routes to behavioral intention and behavior change. The theories can be used to target beliefs that are plausible and viable options for change. Fishbein (this issue) points out that behavior change theories can provide useful guidance about routes and beliefs that could produce behavior changes when based on thorough and careful formative research. The research presented by Krosnick, Chang, Sherman, Chassin, and Presson on smoking initiation provides strong evidence of the role of belief in the negative consequences of smoking by adolescents as a clear barrier to initiation. This work underscores the importance of antismoking campaigns that intensify and reinforce negative consequence beliefs about smoking in young people. Behavior change theories guide researchers to routes to persuasion and to beliefs to target in persuasive efforts but, as Fishbein points out, they do not tell us how to design messages to achieve these changes. By contrast, information processing theories tell us little about which routes to choose or which beliefs to target but do provide guidance about which combinations of audiences and messages are likely to be more amenable or more resistant to specific interventions. For example, even if behavior change research about initiation of or quitting smoking identifies beliefs about negative consequences as predictive of intention, these theories can give no guidance—and are not intended to give guidance—about how to make a targeted belief salient, memorable, or to alter that belief and have it accepted by the target audience. Information processing theories offer answers to this question precisely addressing psychological processes and contextual characteristics associated with attention (Lang; Stephenson & Southwell), encoding, storage, and retrieval (Lang), and acceptance (Briñol & Petty; Kruglanski, Chen, Pierro, Mannetti, Erb, and Spiegel). The link between information processing theories of messages and behavior change theories is in many ways obvious. The behavioral change (BC) theories are not about belief change mechanisms, whereas the information processing models are. The unimodel and the elaboration likelihood model (ELM) are mute about which beliefs should be the targets of persuasion. The two classes of theories complement one another with the BC theories, directing attention to routes and topics of persuasion and the other theories to audience and context characteristics that need to be considered once the topics of persuasion are selected. Together, the two approaches can guide researchers toward (a) selecting routes and beliefs to target and (b) considerations necessary to achieve attention, sustain attention, encode and store for later retrieval, and achieve acceptance and possibly yielding (à la McGuire, 1999). Even though they provide guidance about complex problems of message design, they do not solve these problems. Although some of the discussion about messages within information processing theories is quite specific, other discussion directs message design in only general terms. For example, activation model of information exposure (AMIE) (Stephenson & Southwell) treats message sensation value (MSV) and limited capacity model of motivated mediated message processing (LC4MP) (Lang) treats “information introduced,” both of which identify specific message characteristics that are linked theoretically to mechanisms of attention. On the other hand, the ELM and LC4MP identify more general message characteristics (argument strength and arousing message content, respectively) that are less properties of messages than they are of audience reactions to messages. Message characteristics can be manipulated by designers during message creation although audience reactions require careful formative pretesting. Information processing theories focus on different stages of the persuasion process In his pioneering work on communication and persuasion, McGuire (1999) did a great service for both theoreticians and practitioners of persuasion by identifying a set of stages in the persuasion process that still comports well with more contemporary views of information processing. He suggested that behavior change resulting from persuasive messages required several steps including exposure, attention, encoding, storage, acceptance, accessibility, and yielding.1 The four information processing articles included in this special issue treat different aspects of McGuire’s (1999) communication–persuasion sequence. Although each of the information processing articles deals with the acceptance phase in some way, there are broad differences in emphasis among the approaches, making them complementary to a large extent. For example, both Stephenson and Southwell’s AMIE and Lang’s LC4MP are concerned primarily with components of the process prior to acceptance and yielding, whereas Kruglanski et al.’s unimodel and Brinol and Petty’s paper on the ELM assume exposure and attention focusing instead on the factors associated with acceptance. In the real world of cancer control, intervention messages cannot be effective, unless they break through the clutter of information and are encoded for later retrieval. At the same time, having information is not the same as agreeing with the information’s implications—that is, accepting it as the basis for attitude and intention formation. Without understanding the necessary conditions that make change possible, even if these conditions are not sufficient ones, then effective message interventions are impossible. The AMIE and LC4MP models both deal with attention to messages but more narrowly in the former case and more broadly in the latter. AMIE was design in response to a particular group of adolescents at risk for drug use (Donohew, Palmgreen, & Duncan, 1980), namely high sensation seekers (Zuckerman, 1994), and to a particular problem, namely insuring attention to antidrug messages. The model has clear implications beyond this context, and the authors draw out these implications for cancer control. Attention is used in two different senses in AMIE. The first is attention as in selective exposure to some messages rather than others. This sense helps researchers deal with the message factors that might operate to “break through the clutter,” that is, the modern media experience. The second sense of attention is attention given exposure; that is, once the message is in the immediate consciousness of the audience, are cognitive resources given over to the message as it unfolds? For example, studies using “eyes on screen” as a measure of attention given exposure can indicate visual attention to a message if not the allocation of cognitive resources (Lorch, Palmgreen, Helm, Baer, & Dsilva, 1994). Lang’s LC4MP model focuses on attention, information encoding, storage, and retrieval. The theory operates from assumptions about limited capacity processing and, as a result, invites message designers to consider attentional limits and distractions which messages and persuasive situations can introduce limiting—for good or for ill—the audience’s ability to process the message’s content. The ELM and the unimodel appear to take up the process of influence where the AMIE and LC4MP approaches end. They both assume exposure and attention. The appearance is deceptive. For their apparent differences in approach to what constitutes central and peripheral information in a persuasive message, both the ELM and the unimodel accept the general assumption that the audience’s ability and motivation are crucial to determining whether the information in a message will be elaborated—that is, processed deeply—or not. Lang’s LC4MP is centrally concerned with ability and motivation in at least two different senses, namely as characteristics of the audience and as features of the message that can motivate (e.g., through arousal) or can undermine ability (e.g., through distraction). The ELM and unimodel offer theoretical guidance for successful persuasion through the audience’s ability and motivation to process the information in messages and assume exposure and attention to this information. At the same time, features of messages can affect the more transient aspects of ability and motivation and in the process undermine or intensify elaborative processing so central to theories of acceptance. So although the information processing theories are complementary and, on a superficial reading, seem to focus primarily on different stages of McGuire’s (1999) conceptualization of the persuasion process, they are more deeply interconnected at the level of the audience’s motivation and ability. Motivation and ability operate as central to all four of these theories. Motivation and ability are determined by relatively enduring audience characteristics such as knowledge and personality as well as relatively transient characteristics such as arousing music and evocative visuals. Although such enduring characteristics of target audiences are exogenous determinants of how much elaboration of the message occurs, characteristics of messages themselves can affect transient motivation and ability, altering the resources given over to elaboration of message content. Behavior change, information processing, and message effect theories provide guidance for message design but at different levels of abstraction and application Each of the articles in this special issue offers suggestions about the structure of messages for cancer control. The behavior change theories direct routes to persuasion and topics for messages. Information processing theories direct general strategies for message choice–about the importance of message content and the use of attention-getting message features to orient audiences to a message or to a component of the message. Message effect theories direct the selection of specific appeals (fear and efficacy, gain and loss frames) and message features (vivid case studies rather than statistical information) and toward broader approaches through emotion or tailoring. Consider some of the message features that are discussed by each approach. Behavior change theories do not direct message design but direct what messages should be about Formative research that describes the more important and less important routes to behavioral intention and behavior change can assist in determining which approaches to intervention are likely to be most effective. This formative work can also identify which behavioral beliefs are most likely to differentiate between intenders and nonintenders. These beliefs can be targeted in message interventions. Information processing theories do not dictate specific message choices but instead use the audience’s information processing predilections in combination with general message features to guide message choices Information processing theories are not just about the audience’s psychological processes, but they are also about the interaction between audience characteristics and message features that might interact with those characteristics to produce effective messages. The AMIE focuses on the interaction between an audience’s desire for stimulation measured by sensation seeking and a message’s ability to satisfy that need measured by a series of message characteristics captured in the construct MSV. Audience members are assumed to be motivated by control of arousal (Berlyne, 1960). Those in need of higher levels of stimulation will be drawn to messages that have the capacity to provide that stimulation. Lang’s LC4MP argues that a message’s impact is determined in part by individual differences among audience members as indicated, for example, by behavioral approach and withdrawal tendencies measured through testing for positivity offset and negativity bias in combination with arousal messages can create through their design (e.g., information introduced) and content. The theory assumes limited processing capacity on the part of the audience. The limited capacity is responsible for the interaction of individual differences in processing capacity and message characteristics in determining variation in attention, encoding, storage, and retrieval. The ELM’s predictions about message effects derive from the interaction between audience characteristics of ability and motivation and “message variables” (i.e., characteristics such as argument strength or rhetorical questions). Two competing processes are assumed to drive message processing: the drive to be correct in the face of ambiguous data and the drive for efficiency in information processing. The first pushes toward full information methods of decision making and deeper (or more elaborated) processing, whereas the latter invites shortcuts in decision making in favor of less elaboration. Although research from the ELM perspective has investigated a variety of message characteristics, the one that has received the greatest attention and had the most profound effect on how researchers think about persuasion has been argument strength. Elaborative processing of the message’s content, particulary the arguments and evidence, depends in the ELM on the audience’s ability and motivation. Elaborative processing is determined by ability and motivation such that audiences who are able and motivated are more likely to carry out elaborated processing of the message, counterarguing weak messages and supporting strong messages (ignoring for the moment the effects of biased reactions). Ability and motivation are potentially endogenous factors even though they are often manipulated in experimental studies as exogenous (or fixed) factors. Exploring this possibility opens the links between ELM and AMIE and LC4MP models as more general models of message effects. Kruglanski et al.’s unimodel makes the important point that all message content is information—at least potentially—and precategorizing it as central or peripheral may inappropriately prejudge what is central to the audience and what is not. The unimodel does not question the role of ability and motivation in affecting message elaboration, and, as a result, the linkages between the ELM and the unimodel on the one hand and AMIE and LC4MP on the other hand remain open avenues for integration. Message effect theories focus primarily, but not exclusively, on the “acceptance” phase of the persuasion process The claim that message effect theories such as gain and loss framing focus, primarily on what McGuire (1999) has identified as the acceptance phase of the persuasion process is a reasonable, but overly simple characterization of the role of certain classes of persuasive appeals. For example, gain and loss frames generally assume that audiences are exposed and attentive to message content, otherwise gain and loss framing could not work. At the same time, emotionally evocative images, as found in some antismoking ads to increase a sense of personal threat, can affect an audience’s ongoing visual attention to a message functioning to undermine the ability to process other message content through distraction or enhancing attention by introducing new information. So although some message effect theories offer narrow attention to specific message characteristics and assume exposure, attention, and encoding of the information, others, which are often broader in scope, examine effects on audiences that are necessary conditions to the persuasion process. Green’s (this issue) work on narratives is a case in point. Narrative forms of communication may invite inadvertent persuasive effect in part through the ability of effective narratives to sustain attention in low-motivation audiences. Although a story’s “lesson” to monitor one’s PSA levels may affect the intention to get a PSA, it may do so in part by sustaining attention to a narrative context that would be lost in an alternative format such as expository. The effect of narrative on intention may occur in part because narratives have a greater capacity to sustain attention throughout the message than would expository discourse. By affecting a necessary condition for persuasive effect (i.e., attention), narrative may allow an impact on the acceptance phase. Message effect theories offer recommendations about message design that differ in their level of specificity: strategic approaches, emotional appeals, and specific, narrow appeals. More strategic approaches include tailoring and narrative. By strategic approaches, I mean that the approach tells us how to craft messages so that the targeted behavior change will be accepted by the audience but do not tell message designers what to tailor or what the narrative should be about. Neither of these approaches represent a persuasive appeal. Instead, they represent a strategy for presenting a persuasive appeal. For example, suppose that formative research showed that women older than 40 would be more likely to get a mammogram screening on a regular basis if they were told about the negative consequences of avoiding early detection rather than the positive consequences of early detection (i.e., a loss frame). This loss appeal could be employed as a part of tailoring (to women who find such statements especially risky) or within a narrative framework. The delivery vehicles for appeals are tailored messages and narratives. Strategic approaches to message design are not persuasive appeals. They are content free in the sense that they do not offer a reason for an action in themselves. The beliefs that are targeted by a tailored or narrative approach must come from outside the approach—for example, from behavior change theories. Emotional appeals offer a unique class of persuasive appeals. The articles by Dillard and Nabi and by Peters, Lipkus, and Diefenbach suggest in the breadth of their treatment the importance and complexity of emotional appeals. I want only to make three brief points about emotional appeals to identify their unique role in persuasion. First, emotional appeals are really a set of potential persuasive appeals. They identify specific content and hence, are appeals, as the reason for doing something or avoiding doing it. For example, an antismoking message might appeal to adolescents not take up smoking because the behavior is disgusting. The same behavior with pregnant women might employ an appeal to fear over consequences for the fetus. Both approaches are emotional, but each involves a different appeal—that is, a reason for not smoking. The reasons are it is disgusting or it is a threat to my baby. Second, emotionally evocative messages are not only appeals that can affect acceptance but also potential distractions that can affect ability to process the message or motivators enhancing attention to the message and its contents. We ignore these effects at our peril in message design. For example, if an emotionally evocative image in a persuasive message was to draw the audience’s attention to the image but away from the message’s strong arguments, the image could undermine rather than enhance elaboration and undermine acceptance. Third, the creation of emotionally evocative messages—itself a major issue in message design—remains more art than science although the science is starting to get some attention. Specifically, research on appraisal processes has identified detailed components of situations that, once construed by the audience, form the determinants of emotional reaction to the situation (Omdahl, 1995; Smith & Ellsworth, 1985). The translation of these “stimulus evaluation checks,” as Scherer (1984) calls them, to the context of message appraisal holds promise for a more scientifically based assessment of message factors that can produce reliable emotional reactions without making unnatural claims about reducing artful message design to simplistic principles. Some message effects theories have targeted very specific message features exploring in a variety of contexts, including cancer control, the impact of those features on attitudes, intentions, and behaviors. Two of the more comprehensive and effective examples of message effects theories in this arena are included in this special issue—Rothman, Bartels, Wlaschin, and Salovey’s gain and loss frames and Zillman’s exemplification theory. Each offers specific direction in message design in a cancer control context, and each derives from a broader theoretical base. In the case of gain–loss frames, the theoretical base is prospect theory (Slovic, Finucane, Peters, & MacGregor, 2002), whereas in the case of Zillman’s approach, exemplification theory has its roots in cognitive heuristics (Fiske & Taylor, 1991) that has been so central to understanding human decision making. Both approaches have produced bodies of important and useful research. Both have also struggled to produce data supporting a clear explanation of the mechanism through which gain–loss and exemplification effects operate. The most commonly cited explanation for gain–loss effects is emotional mediation in gain–loss (Salovey, Schneider, & Apanovich, 2002) and cognitive availability in exemplification studies (Zillman & Brosius, 2000). If these expected mechanisms are found valid, then the connections between the research in gain–loss frames or vivid exemplars and the information processing approaches to message effects become ripe for examination. If a vivid exemplar about skin cancer draws attention and enhanced encoding, it may also have the capacity, if poorly designed, to distract the audience from other message content or lead to cognitive overload and defensive processing as some have suggested (Keller & Block, 1997). The point is that even very specific message effects theories such as gain–loss and exemplification offer explanatory mechanisms that engage with the broader explanatory mechanisms of certain of the information processing theories. Exploring these connections will enrich information processing theories by making them more specific regarding message design and enrich the narrow message effect theories by linking them to a broader array of potential message features. Some directions for message effect theories Choosing one appeal framework rather than another Message effects theories have yet to address questions about what kind of appeal should be selected from the appeals or strategies available to the message designer. To appreciate this issue, assume for the moment that formative research has been conducted on treatment seeking for smoking so that it is already known that the attitudinal route is the most salient route to the intention to seek treatment and that a particular behavioral belief has been identified as a potential target of persuasion—for example, avoiding manipulation by tobacco companies (Fishbein). Assume, further, that exposure to the message and attention to its contents are not issues because the messages will be delivered in a context where both can be assured. The question that still remains is what approach to belief change should be undertaken. Should the message use a rational or emotional appeal? If the appeal is a rational one, should the message be manipulated by framing it clearly as gain or loss or manipulated by making the consequences carried by vivid exemplars or both? In short, researchers and message designers in public health have had little guidance from our theories about message effects for choosing among appeals.2 More is known about choosing between appeals within a message framework (e.g., gain vs. loss) than between frameworks (e.g., vivid exemplars vs. strong arguments in an expository frame). This issue raises a second problem. When multiple features are incorporated in the same message, how can we study the effectiveness of combinations of features? In the example of treatment seeking, will a loss-framed vivid exemplar that activates fear work better or worse than a gain-framed vivid exemplar that activates disgust? The complexity of consequences when three or more message features are involved calls for new methods and integration across theories. Recent work reported in the Rimer and Kreuter and manuscript suggests the potential value of fractional factorial designs in addressing this question empirically (see also Collins, Murphy, Nair, & Strecher, 2005). Moving from the art of message design to the science of message design The creation of effective persuasive messages is a matter of both science and art. Much of the research reported in this special issue emphasizes the science of message effects and little on the science of message design. Part of the reason for this imbalance is that certain aspects of message design are much less under the control of a priori principles than others. For example, ascribing gain and loss frames to an appeal is relatively straightforward, making this approach to message design particularly easy to implement. On the other hand, potentially powerful techniques such as strong arguments, transporting narratives, and emotionally evocative texts or videos are not susceptible to implementation via simple design principles. Tools for the evaluation of argument strength (Cacioppo, Harkins, & Petty, 1981) transportation through narrative worlds (Green) and emotional involvement are available. The tools for evaluation of each are really just a way to replace message design with audience evaluation. Researchers are clearly aware of this problem. Areni (2002), Morley and Walker (1987), and Johnson and Smith-McLallen (2006) are working on the logical and psychological features of strong and weak arguments. Others have carried out studies translating appraisal theories of emotional response into the message, especially video message, domain (Dillard, Kinney, & Cruz, 1996). The structure of effective narrative has received attention in cognitive psychology (Bruner, 1986; Schank & Abelson, 1995) and film theory (Bordwell, 1985; Grodal, 1994). Although helpful, there is much distance to be covered. Multilevel approaches The theories in this special issue are oriented primarily toward the design and effects of messages aimed at individuals for cancer control. Viswanath and Emmons, on the other hand, draw our attention to institutional, social, and cultural factors that can operate to maintain differences in healthy behavioral choices and in disparities in risk. The next generation of thinking about message effects should give serious consideration to both the methods for multilevel data gathering and analysis (Raudenbush & Bryk, 2002) and the multilevel theory development (e.g., agent-based models, Wooldridge, 2004). Summary In conclusion, all the theories presented in this issue are pertinent to the effects that messages can have. The theories are complex not only because each offers a different lens through which to see but because the light they can provide on the impact of persuasive messages on cancer control is refracted through the contiguous lenses that the other theories offer. Understanding the relationships among the theories is a major challenge, but one for which the payoff could be significant improvement in health. Acknowledgment This research was supported by a grant from the National Cancer Institute P50 CA 095856-01. Notes 1 " McGuire’s (1999) steps are more complex than those listed here, but this subset is sufficient for our purposes. 2 " Some research has been done on matching the basis for the appeal with the audience’s thinking style (Rosenthal & Epstein, 2000) and also with the emotional or rational basis for the attitude (Fabrigar & Petty, 2003). References Areni , C. S . ( 2002 ). The proposition-probability model of argument structure and message acceptance . Journal of Consumer Research , 29 , 168 – 187 . Google Scholar Crossref Search ADS WorldCat Berlyne , D. E . ( 1960 ). Conflict, arousal, and curiosity . New York : McGraw-Hill . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Bordwell , D . ( 1985 ). Narration in the fiction film . Madison : University of Wisconsin Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Briñol , P. , & Petty , R. E. ( 2006 ). Fundamental processes leading to attitude change: Implications for cancer prevention communications . Journal of Communication , 56 , S81 – S104 . Google Scholar Crossref Search ADS WorldCat Bruner , J . ( 1986 ). Actual minds, possible worlds . Cambridge, MA : Harvard University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cacioppo , J. T. , Harkins , S. G., & Petty , R. E. ( 1981 ). The nature of attitudes and cognitive responses and their relationships to behavior. In R. E. Petty, T. M. Ostrom, & T. C. Brock (Eds.), Cognitive responses in persuasion (pp. 32 – 54 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Cialdini , R. B . ( 2003 ). Crafting normative messages to protect the environment . Current Directions in Psychological Science , 12 , 105 – 109 . Google Scholar Crossref Search ADS WorldCat Collins , L. M. , Murphy , S. A., Nair , V. N., & Strecher , V. J. ( 2005 ). A strategy for optimizing and evaluating behavioral interventions . Annals of Behavioral Medicine , 30 , 65 – 73 . Google Scholar Crossref Search ADS PubMed WorldCat Dillard , J. P. , Kinney , T. A., & Cruz , M. G. ( 1996 ). Influence, appraisals, and emotions in close relationships . Communication Monographs , 63 , 105 – 130 . Google Scholar Crossref Search ADS WorldCat Dillard , J. P. , & Nabi , R.L. ( 2006 ). The persuasive influence of emotion in cancer prevention and detection messages . Journal of Communication , 56 , S123 – S139 . Google Scholar Crossref Search ADS WorldCat Donohew , L. , Palmgreen , P., & Duncan , J. ( 1980 ). An activation model of information exposure . Communication Monographs , 47 , 295 – 303 . Google Scholar Crossref Search ADS WorldCat Fabrigar , L. R. , & Petty , R. E. ( 1999 ). The role of the affective and cognitive bases of attitudes in susceptibility to affectively and cognitively based persuasion . Personality and Social Psychology Bulletin , 25 , 363 – 381 . Google Scholar Crossref Search ADS WorldCat Fishbein , M. , & Cappella , J. N. ( 2006 ). The role of theory in developing effective health communications . Journal of Communication , 56 , S1 – S17 . Google Scholar Crossref Search ADS WorldCat Fiske , S. T. , & Taylor , S. E. ( 1991 ). Social cognition (2nd ed.). New York : McGraw Hill . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green , M. C . ( 2006 ). Narratives and cancer communication . Journal of Communication , 56 , S163 – S183 . Google Scholar Crossref Search ADS WorldCat Grodal , T. K . ( 1994 ). Cognition, emotion, and visual fiction . Copenhagen, Denmark : University of Copenhagen . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Johnson , B. T. , & Smith-McLallen , A. ( 2006 ). The good, the bad and the likely: Arguments gain strength more via desirability than via truthfulness . Unpublished manuscript, Department of Psychology, University of Connecticut , Storrs . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Keller , P. A. , & Block , L. G. ( 1997 ). Vividness effects: A resource-matching perspective . Journal of Consumer Research , 24 , 295 – 305 . Google Scholar Crossref Search ADS WorldCat Krosnick , J. , Chang , L., Sherman , S. J., Chassin , L., & Presson , C. ( 2006 ). The effects of beliefs about the health consequences of cigarette smoking on smoking onset and quitting . Journal of Communication , 56 , S18 – S37 . Google Scholar Crossref Search ADS WorldCat Kruglanski , A. W. , Chen , X., Pierro , A., Mannetti , L., Erb , H. P., & Spiegel , S. ( 2006 ). Persuasion according to the unimodel: Implications for cancer communication . Journal of Communication , 56 , S105 – S122 . Google Scholar Crossref Search ADS WorldCat Lang , A . ( 2006 ). Using the limited capacity model of motivated mediated message processing (LC4MP) to design effective cancer communication messages . Journal of Communication , 56 , S57 – S80 . Google Scholar Crossref Search ADS WorldCat Lorch , E. P. , Palmgreen , P., Helm , D., Baer , S. A., & Dsilva , M. U. ( 1994 ). Program context, sensation-seeking, and attention to televised anti-drug public service announcements . Human communication Research , 20 , 390 – 412 . Google Scholar Crossref Search ADS WorldCat McGuire , W. J . ( 1999 ). Constructing social psychology: Critical and creative processes . New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Morley , D. D. , & Walker , K. B. ( 1987 ). The role of importance, novelty, and plausibility in producing belief change . Communication Monographs , 54 , 436 – 442 . Google Scholar Crossref Search ADS WorldCat Omdahl , B. L . ( 1995 ). Cognitive appraisal, emotion, and empathy . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Peters , E. , Lipkus , I., & Diefenbach , M. A. ( 2006 ). The functions of affect in health communications and in the construction of health preferences . Journal of Communication , 56 , S140 – S162 . Google Scholar Crossref Search ADS WorldCat Raudenbush , S. W. , & Bryk , A. S. ( 2002 ). Hierarchical linear models: Application and data analysis methods (2nd ed.). Thousand Oaks, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rimer , B. K. , & Kreuter , M. W. ( 2006 ). Advancing tailored health communication: A persuasion & message effects perspective . Journal of Communication , 56 , S184 – S201 . Google Scholar Crossref Search ADS WorldCat Rosenthal , L. , & Epstein , S. ( 2000 ). Rational and experiential thinking styles as related to receptivity to messages syntonic and dystonic with thinking style . Unpublished manuscript, University of Massachusetts at Amherst . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rothman , A. J. , Bartels , R. D., Wlaschin , J., & Salovey , P. ( 2006 ). The strategic use of gain and loss framed messages to promote healthy behavior: How theory can inform practice . Journal of Communication , 56 , S202 – S220 . Google Scholar Crossref Search ADS WorldCat Salovey , P. , Schneider , T. R., & Apanovich , A. M. ( 2002 ). Message framing in the prevention and early detection of illness. In J. P. Dillard, & M. Pfau (Eds.), The persuasion handbook: Developments in theory and practice (pp. 391 – 406 ). Thousand Oaks, CA , Sage . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Schank , R. C. , & Abelson , R. P. ( 1995 ). Knowledge and memory: The real story. In R. S. Wyer (Ed.), Knowledge and memory: The real story. Advances in social cognition (Vol. 8 , pp. 1 – 85 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Scherer , K. R . ( 1984 ). Emotion as a multicomponent process: A model and some crosscultural data. In P. Shaver (Ed.), Review of personality and social psychology. Volume 5. Emotions, relationships, and health (pp. 37 – 63 ). Beverly Hills, CA : Sage . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Slovic , P. , Finucane , M. L., Peters , E., & MacGregor , D. G. ( 2002 ). The affect heuristic. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment (pp. 397 – 420 ). New York : Cambridge University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Smith , C. A. , & Ellsworth , P. C. ( 1985 ). Patterns of cognitive appraisal in emotion . Journal of Personality and Social Psychology , 48 , 813 – 838 . Google Scholar Crossref Search ADS PubMed WorldCat Stephenson , M. T. , & Southwell , B. G. ( 2006 ). Sensation seeking, the activation model, and mass media health campaigns: Current findings and future directions for cancer communication . Journal of Communication , 56 , S38 – S56 . Google Scholar Crossref Search ADS WorldCat Viswanath , K. , & Emmons , K. M. ( 2006 ). Message effects and social determinants of health: Its application to cancer disparities . Journal of Communication , 56 , S238 – S264 . Google Scholar Crossref Search ADS WorldCat Wooldridge , M . ( 2002 ). An introduction to multi-agent systems . Chichester, UK : Wiley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Ybarra , O. , & Trafimow , D. ( 1998 ). How priming the private self or collective self affects the relative weights of attitudes and subjective norms . Personality and Social Psychology Bulletin , 24 , 362 – 370 . Google Scholar Crossref Search ADS WorldCat Zillman , D . ( 2006 ). Exemplification effects in the promotion of safety and health . Journal of Communication , 56 , S221 – S237 . Google Scholar Crossref Search ADS WorldCat Zillman , D. , & Brosius , H. ( 2000 ). Exemplification in communication: The influence of case reports on the perception of issues . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zuckerman , M . ( 1994 ). Behavioral expressions and biosocial bases of sensation seeking . Cambridge, UK : Cambridge University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association

Green, Melanie, C.

doi: 10.1111/j.1460-2466.2006.00288.xpmid: N/A

Abstract Narratives can be an effective means of communicating cancer-related information. Transportation into narrative worlds, or immersion into a story, is a primary mechanism of narrative persuasion (Green & Brock, 2000, 2002). Transportation theory extends the domain of traditional message effects theories, as well as providing mechanisms for behavior change. Transporting narratives can both change beliefs and motivate action, and may be particularly useful for conveying cancer information because they reduce counterarguments (and thus help individuals overcome barriers to treatment seeking); facilitate the mental simulation of unknown, difficult, or frightening procedures (e.g., screening); provide role models for behavior change; and create strong attitudes that are based on both cognition and emotion. Throughout human history, narratives have been used to share information, change beliefs, and inspire action. Indeed, narrative has been described as a fundamental mode of thinking (Schank & Abelson, 1995). In the health domain, narratives may be as simple as a brief public service announcement (commercial) or as complex as a multiepisode telenovela (soap opera). Engaging, transporting stories may be especially well suited to convey cancer-related information for a variety of reasons—they reduce counterarguments (and thus help individuals overcome barriers to treatment seeking); facilitate the mental simulation of unknown, difficult, or frightening procedures (e.g., screening, seeking treatment for smoking cessation); provide role models for behavior change; and create strong attitudes that are based on both cognition and emotion. Traditional theories of message effects, particularly those in the persuasion literature, have tended to emphasize nonnarrative communications (with a few exceptions, such as cultivation theory; Gerbner, Gross, Morgan, & Signorielli, 1994). Health behavior theories have often highlighted processes within an individual, such as stages of change or health beliefs, without implying a particular type of intervention. Communicators may or may not use narratives in their interventions (e.g., a tailored communication could be a story of a person going through a similar treatment or an expository presentation of relevant health information), but the effect of stories per se has not typically been the focus of study. In contrast, transportation theory (Green & Brock, 2000, 2002) focuses specifically on the experience of becoming immersed in a story and the ways in which this “transportation into a narrative world” can lead to real-world belief (and behavior) change. This perspective complements existing attitude change theories (e.g., the Elaboration Likelihood Model; Petty & Cacioppo, 1986), which are most relevant to argument-based messages, such as editorials and advertisements. Persuasion theories that center on reactions to message arguments do not adequately address stories, which illustrate events rather than presenting explicit arguments and lists of facts. Narratives rely on characters rather than external sources and often are presented as entertainment rather than education. A narrative structure—with a beginning, middle, and end—ties actions and implications together in a causal chain, rather than relying on a set of propositions that may be more or less well integrated. This paper describes the transportation theory of narrative persuasion and the ways in which transportation into narrative worlds (even fictional ones) can lead to cancer-relevant belief and behavior change, as well as linking this theory with other perspectives on message effects and behavior change. The theoretical framework described below (see also Green & Brock, 2000, 2002) has received initial empirical support; however, a goal of this paper is to encourage further testing of the transportation theory propositions, particularly in the area of cancer communication. Transportation into a narrative world Transportation into a narrative world is defined as an integrative melding of attention, imagery, and feelings, focused on story events (Green & Brock, 2000, 2002; see also Gerrig, 1993; Nell, 1988). Transportation, psychologically similar to flow (Csikszentmihalyi, 1990) or absorption, is a form of experiential response to narratives (Prentice & Gerrig, 1999). Transportation is a pleasurable state that contributes to media enjoyment (Green, Brock, & Kaufman, 2004).1 Indeed, the fact that film, television, and book publishing are multibillion dollar industries illustrates the extent to which individuals value and seek out transportation experiences. Individuals can be transported into factual or fictional communications, and into written, spoken, or visual narratives. Measurement and manipulation Transportation can be measured with a 15-item self-report scale (Green & Brock, 2000). The scale has shown good internal consistency, as well as discriminant and convergent validity. Participants answer each item on a scale of 1 (not at all) to 7 (very much). Example items include “I was emotionally involved in the narrative while reading it” and “I could picture myself in the scene of the events described in the narrative.” Transportation can also be manipulated by varying story quality or by varying the instructions given to readers. For example, instructions to focus on the surface aspects of the story, such as difficulty and grammar, produce lower transportation (Green & Brock, 2000). The transportation scale taps cognitive, emotional, and imagery processes. These processes work together to create the transportation experience; therefore, using the entire scale often provides the strongest predictive power. However, it is possible that particular dimensions of transportation may be more predictive of some outcomes than others; emotion may be a key factor in some circumstances and imagery in others. Individual differences Across studies, there is no gender difference in transportation, although men may be more transported into some kinds of stories and women into others. Transportation does show a moderate correlation with dispositional empathy, so it is possible that individuals who are more empathic may be better candidates for narrative-based persuasion. Similarly, individuals with greater tendencies toward becoming absorbed in activities in general (e.g., Tellegen & Atkinson, 1974) and those with higher propensities toward forming mental images (Sheehan, 1967) are likely to experience stronger effects from narratives. “Transportability,” the extent to which individuals readily become deeply transported into stories, can be measured as an individual difference (Dal Cin, Zanna, & Fong, 2004). Transportation effects Research has shown that individuals who are transported into a narrative world are likely to change their real-world beliefs in response to information, claims, or events in a story. For example, transported readers of a story about an attack on a young girl at a shopping mall were more likely than their less-transported counterparts to believe that malls were dangerous places and that the world was unjust (Green & Brock, 2000). Transportation was also associated with increased positivity toward sympathetic characters and a reduction in negative cognitive responses to the story. There are at least three possible means by which transportation can affect readers: creating connections with characters, reducing counterarguing, and making narrative events seem more like real experience (including providing concrete examples of events and vivid mental images of story events or characters). Transportation into a narrative world also helps individuals to engage in mental simulations of events or behaviors. Relationships with characters Transportation is associated with increased positivity toward sympathetic characters (Green & Brock, 2000). Transported individuals may identify with those characters or come to view them as friends. If an individual likes or identifies with a particular character, the implications of events experienced by the character or assertions made by the character may carry special weight in shifting a reader’s beliefs. Increasing identification Identification has been conceptualized in different ways by different researchers (see Cohen, 2001, for a discussion). Broadly speaking, identification refers to relating to characters, caring about them, and putting oneself in the character’s place. Separating out the different aspects of identification—liking, similarity, and empathy, for example—may be useful in providing a fuller understanding of identification effects. Factors that make it easier for readers to identify with characters may facilitate the experience of transportation and thus should increase belief change. Research suggests that a preexisting similarity between a narrative character and the recipient (reader, viewer) of a narrative can increase transportation (Green, 2004), for example, readers who had a gay friend or family member were more transported into a story that dealt with prejudice against homosexuals in fraternities. This effect appears to be stronger when the preexisting similarity is on a story-relevant dimension rather than on a simple demographic characteristic (e.g., gender, school attended), although there is not strong evidence on this point. These findings are consistent with research that suggests that tailoring health messages to specific groups or individuals can be an effective strategy (e.g., Kreuter, Strecher, & Glassman, 1999; Rimer & Kreuter, 2006). Therefore, during the message design or story selection phase of a cancer communication campaign, attention should be given to finding stories that match the audience on key characteristics (values, experiences, and so on). Modeling Characters may serve as role models for appropriate behavior, as described by social cognitive theory (e.g., Bandura, 1977). Characters illustrate the costs and benefits of different courses of action. People may be inspired to emulate the actions of admired characters, avoid the problems of negative characters, or follow in the footsteps of transitional characters, who experience a transformation (typically a move from negative to positive attitudes or behaviors) over the course of the story (Rogers et al., 1999). The effectiveness of narrative modeling has been demonstrated by studies of entertainment education in a variety of health domains such as family planning (see Slater, 2002, for a review). Modeling has been shown to increase perceived self-efficacy, or individuals’ beliefs that they are able to perform a specific behavior, such as breast self-examinations (Anderson, 2000). Seeing similar others succeed at a task provides confidence to viewers. Consistent with the identification research discussed above, there is some evidence that models who are similar to the target audience are more effective at motivating the desired behavior (e.g., that African American models are more influential for African American women; Anderson & McMillion, 1995). Modeling may also contribute to response efficacy, or a belief that the action will have the intended effect (e.g., that having a mammogram will decrease the chances of dying of breast cancer). When individuals identify with characters, these characters may provide templates for “possible selves” (Markus & Nurius, 1986). An individual may wish to avoid a future possible self as a cancer sufferer, or a person already diagnosed with cancer may strive for a “cancer survivor” possible self through adherence to a treatment regimen. Being able to identify with or relate to a character may be particularly useful in helping individuals overcome barriers to screening or treatment seeking, including embarrassment about needing help or the belief that cancer “couldn’t happen to me.” Furthermore, seeing characters struggle with cancer fears may provide the emotional benefit of making individuals feel that they are not alone in confronting these difficult issues. Individuals in narratives may also provide inspiration postdiagnosis. Reading Lance Armstrong’s story of survival and triumph over testicular cancer can give hope of living a full life after cancer treatment; hearing from survivors of breast cancer provides living proof that a breast cancer diagnosis is not a death sentence (Erwin, Spatz, & Turturro, 1992). Norms Although perceived norms are likely to be most influenced by real others, characters in a narrative may also create shifts in normative beliefs (perceptions that important others think that the person should or should not perform a behavior, a component of the theory of reasoned action). These shifts in norms can occur simply from exposure to popular culture. If beloved characters on a favorite television show go for cancer screening or mention a preference for healthy foods, viewers of the show may come to believe that those activities are generally viewed as beneficial.2 On the negative side, however, smoking scenes in movies glamorize smoking and can increase the young people’s intent to smoke (Pechmann & Shih, 1999). A large-scale study of adolescents found that viewing smoking in movies was linked to smoking initiation (Dalton et al., 2003) and positive attitudes toward smoking even among nonsmokers (Sargent et al., 2002). Viewing on-screen smoking can lead to actual smoking through multiple pathways. Not only can movie smoking lead to positive images of smokers but can also lead to perceptions that smoking is more frequent (McCool, Cameron, & Petrie, 2005). However, antismoking advertisements can counteract these effects, and specifically, recent studies of adolescents’ responses to antismoking advertisements indicated that ads focusing on the social disapproval that could result from smoking were more effective in creating nonsmoking intentions than other types of advertisements, such as those focused on health risks (Pechmann, Zhao, Goldberg, & Reibling, 2003). A related approach involves evoking or creating stereotypes of individuals who engage in healthy or unhealthy behavior (Pechmann, 2001). Reinforcing negative stereotypes about smokers (that they are unsuccessful, “stinky,” and dumb), for example, can reduce intentions to smoke (Pechmann & Knight, 2002). Emotional response Characters help create emotional responses to narratives. Emotion is a core component of narrative impact (Oatley, 2002). Research on dual-process models of attitude change suggests that attitudes can be based in affect, cognition, or both (e.g., Fabrigar & Petty, 1999). Narratives may be a particularly effective means of forming and changing affectively based attitudes, as well as attitudes based in both affect and cognition. Emotional responses to narratives can also motivate behavior change (e.g., by evoking sadness for a character’s poor outcome and a desire to avoid a similar outcome oneself, or fear about the consequences of risky behaviors). Narratives may also help people manage their emotions postdiagnosis. Listening to others’ stories (or even telling one’s own story) may allow individuals to gain some distance from the fear, uncertainty, sadness, or anger that can accompany the discovery that one has cancer (Carlick & Biley, 2004). Reduction of counterarguing/negative cognitive responses A second way that transportation can lead to attitude change is by lowering resistance and reducing counterarguments. Resistance can be broadly defined as a reaction against change or a motivation to oppose persuasive appeals (Knowles & Linn, 2004). Resistance to persuasion is a key obstacle to belief and behavior change, and resistance may be a particular concern for cancer communication because individuals are likely to be highly motivated to maintain an illusion of invulnerability. Narratives may be especially effective at overcoming resistance; for example, Slater and Rouner (1996) found that in processing alcohol education messages, college students rated statistical evidence as more persuasive when the message was congruent with their values but narrative evidence as more persuasive when the message was incongruent with their values (counterattitudinal). Stories may be a relatively subtle form of persuasion that is less likely to inspire reactance, particularly if they are not presented as overt persuasive attempts (e.g., Dal Cin et al., 2004). A uses-and-gratifications perspective suggests that individuals seek out entertainment products to meet particular needs (Rubin, 2002). Presenting a narrative as entertainment rather than education may be helpful in reducing selective exposure (i.e., individuals avoiding health information, perhaps because they expect it to be either frightening or boring). Individuals may be drawn to the narrative because of its plot, interest, or entertainment value but can then be affected by cancer-related information contained in the story. Concern about selective exposure may be most relevant to the earlier stages of the cancer continuum (prevention and screening); postdiagnosis, individuals may be quite motivated to seek out cancer-relevant information. Message effects theories based on the cognitive response perspective (Petty, Ostrom, & Brock, 1981) suggest that the thoughts individuals have in response to a message are primary determinants of persuasion. Positive thoughts lead to greater persuasion, whereas negative thoughts lead to a lack of persuasion (or even a boomerang effect, in which individuals adopt a position opposite the one being advocated). Transportation into a narrative world leads to a reduction in negative cognitive responding or counterarguing of story assertions (Deighton, Romer, & McQueen, 1989; Green & Brock, 2000).3 Studies of the way in which individuals correct (or fail to correct) beliefs that are false or inaccurate suggest that the default is for individuals to accept all propositions that they hear as true, unless they have the motivation and ability to refute them (Gilbert, 1991). Transportation may reduce individuals’ ability to counterargue a story’s assertions because the reader’s mental capacity is devoted to imagining story events. It may also be more difficult to counterargue conclusions that are implied by the story rather than stated directly as arguments. Transportation may also reduce individuals’ motivation to counterargue because interrupting the narrative flow to dispute the author’s claims or descriptions would likely destroy the pleasure of the experience. Even when individuals have finished a narrative, they may not be motivated to go back and critique the implications of the story, especially if they do not believe the story has influenced them. If individuals are not refuting claims made in a narrative or disputing the realism of the narrative situations, the story events are likely to become integrated into real-world belief structures. The integration of story information into real-world beliefs has been demonstrated with explicit attitudes (e.g., Green, 2004; Green & Brock, 2000), and Dal Cin et al. (2004) suggest that it is likely that narratives would also affect implicit attitudes. Implicit attitudes are unconscious associations. For example, tobacco use may be embedded in a network of either positive or negative associations (Swanson, Rudman, & Greenwald, 2001). Even though an individual may not be aware of these links, implicit attitudes can influence behavior, especially spontaneous behavior (e.g., Marsh, Johnson, & Scott-Sheldon, 2001). Thus, they may be particularly important for influencing cancer risk behaviors such as smoking or going out in the sun without sunscreen. Future research should extend the investigation of narratives to include implicit attitudes, as well as testing the impact of implicit attitudes on cancer-relevant behaviors specifically. Increasing realism Part of the power of narratives is their ability to give concrete form to abstract ideas. Providing a list of the benefits of cancer screening may not capture recipients’ attention or inspire action in the same way that hearing a woman talk about how getting a mammogram allowed her to catch her cancer in time to save her life. Transportation enhances this aspect of narratives by making narrative experience seem more like real experience (Green, 2004). Direct experience with attitude objects is a powerful predictor of attitudes (see Fazio & Zanna, 1981, for a review). Perceived realism matters; if the story seems more like an actual event, the plausibility and impact of the story are increased (see Potter, 1986). Concrete examples Zillmann’s exemplification theory suggests that concrete incidents, particularly ones related to danger or risk, attract attention and are stored in long-term emotional memory (Zillmann, 2002). Such specific examples are also highly accessible and thus tend to influence decision making in a heuristic fashion. Although exemplification theory states that relevant instances may be conveyed through pictures or language, it is likely that narratives are an especially frequent and powerful source of exemplars. A picture of a person suffering from advanced cancer may be striking, but the story of how that person’s actions contributed to the cancer (e.g., years of smoking) or other narrative elements (e.g., reactions of family members, opportunities lost due to the disease) are likely to be even more touching and memorable (Green & Brock, 2002). People tend to generalize from stories even when the cases presented in the story are not typical (see Strange & Leung, 1999). People often disregard the base rates of an event. It is difficult to argue against a character’s concrete experience. These generalizations may be useful when the cases imply good health practices (e.g., a colonoscopy that helped a person avoid dying from colon cancer), but a problem may arise when these narratives or exemplars are counter to good health practices (e.g., a person who smoked his entire life but never suffered any ill effects). It is possible that other stories may be the most effective way to combat inaccurate perceptions. Direct comparisons between narrative/report and statistical information have shown mixed results, although many of these studies show that reports are more persuasive (see Baesler & Burgoon, 1994, for a brief review). However, these experiments typically have not measured the degree of transportation evoked by the narrative accounts; pallid narratives may be less effective than statistical evidence, whereas absorbing stories are more effective. Mental imagery Story-based mental imagery may be a particularly powerful means by which narratives can influence beliefs. Visual images, or mental pictures, can be evoked by a transporting narrative or provided by a visual narrative (television, movies). These images are likely to be enduring and are difficult to change with other types of arguments. Images have a long history in health communication; antismoking campaigns have used striking pictures of diseased lungs to convey the dangers of smoking (see Hammond, Fong, McDonald, Brown, & Cameron, 2004, for a recent example). Transportation-imagery theory (Green & Brock, 2002) suggests that these images are most powerful when they are evoked by a story, rather than provided in isolation. Theories of mental imagery are related to research on the vividness effect, but these two effects differ in some key respects. Vividness refers to the extent to which a message is “emotionally interesting, concrete and imagery provoking, proximate in a sensory, temporal, or spatial way” (Nisbett & Ross, 1980, p. 45). As this definition shows, vividness encompasses a range of elements. Furthermore, vividness studies, which have shown mixed effects, have typically focused on features of the message itself rather than the imagery experienced by readers (e.g., Taylor & Thompson, 1982). An important point from the vividness studies is that vivid imagery that is not central to the message themes may be a distraction rather than an enhancement and thus may actually reduce persuasive impact (Smith & Shaffer, 2000). Imagery should be relevant to the central message being conveyed; for example, a story should encourage imagery related to a family’s joy after a mother’s successful treatment for breast cancer, rather than creating vivid mental images of (say) the family’s hometown. Imagery or vividness may also interact with other factors. For instance, studies by Block and Keller (1997) of messages about human papilloma virus (HPV) (a cause of cervical cancer) and skin cancer suggested that vivid messages (operationalized as stories in one experiment and pictures in the other) were more persuasive, but only among high self-efficacy participants, whereas Cox and Cox (2001) found that a loss-framed anecdote about breast cancer was more effective than statistical evidence, but that a gain-framed anecdote was less effective. Further research and theory is needed to understand the moderators of imagery and vividness effects. Mental simulation The effects of transportation—creating connections with characters, providing realistic, concrete situations, and evoking mental imagery—combine to facilitate the mental simulation of new situations (cf. Oatley, 2002). This ease of imagining can help change beliefs and behavior (Gregory, Cialdini, & Carpenter, 1982; Sherman, Cialdini, Schwartzman, & Reynolds, 1985). For example, if individuals find it easier to imagine themselves suffering from cancer, they may form a stronger behavioral intention to engage in actions to prevent cancer (changing diet or exercise habits). To the extent that the story can provide specific pathways to goals (e.g., showing how a character quit smoking or integrated sunscreen use into their daily routine), it may be especially effective in motivating individuals to reach a desired future self, and may increase individuals’ optimism about their ability to achieve their goal and their feelings of self-efficacy (e.g., Taylor & Schneider, 1989). Mental simulation can serve as a form of behavioral rehearsal, which can be a means of increasing self-efficacy. If individuals can imagine themselves going through the steps to get a mammogram, they may feel more confident in their ability to do so in real life. Mental simulations could also be useful postdiagnosis as individuals try to decide among various treatment options. Patients may be faced with an overload of information, and a narrative can serve as a simplifying or organizing structure. Narratives that help patients imagine what life would be like after surgery versus chemotherapy might aid decision making. For example, the Comprehensive Health Enhancement Support System (CHESS), an interactive health support system for cancer patients, provides some narrative tools (stories from others, opportunities for journaling) among other informational and social support resources. CHESS has been effective at improving quality of life and satisfaction with treatment decision making; however, the contribution of the specifically narrative components is unknown (Gustafson et al., 2001). Attitude strength Because the end goal of health communication is behavior change, it is essential that cancer communications lead to strong attitudes. Strong attitudes are persistent over time and resistant to counterpersuasion; these attitudes are held with certainty and are highly accessible (see Petty & Krosnick, 1995). Transportation is likely to lead to strong attitudes for several reasons. First, transportation focuses an individual completely on the narrative; the transported state is one of high cognitive engagement. The more thought or attention an individual has given to the attitude issue, the stronger the attitude should be. Second, transportation into narratives can meld affective and cognitive reactions, going beyond simple learning. Attitudes that have both cognitive and affective bases are likely to be stronger. The narrative structure itself may also contribute to enduring attitudes. Specifically, if individuals recall one part of a story (perhaps a particularly striking mental image or compelling character), they are likely to also recall the rest of the story due to the cause-and-effect structure of the narrative message. In contrast, if a rhetorical persuasive message presents a set of arguments that are relatively independent of one another, recalling one argument may not aid in bringing others to mind. Of course, argument recall is not a necessary condition for attitude change, but repetition or rehearsal of the primary message is likely helpful in maintaining the persuasive impact. Fiction Unlike other kinds of persuasive messages, narratives are often fictional or “made up.” Because information about cancer screening, prevention, and treatment can literally be a matter of life and death, communicators have a clear ethical responsibility to ensure that the information conveyed is accurate. However, this responsibility does not preclude including cancer-related information in fictional communications. Indeed, fiction may be an especially effective way of reaching audiences that otherwise might avoid such information. Meaning and impact of fiction Individuals do not typically interpret the term “fiction” to mean “false” (at least when they are referring to narratives). Rather, fiction refers to products of imagination, which may be based on reality to a greater or lesser degree; there is no explicit claim of accuracy. Even though most fiction is a mix of truth and creative invention, both anecdotal and empirical evidence indicate that individuals readily adopt information from fictional programs as truth. For example, writers on the popular television program ER regularly include health information as part of the drama of the show, and indeed, viewers learn from the program. For example, after an episode focusing on a teenage girl with HPV, more viewers were aware of the connection between HPV and cervical cancer (Brodie et al., 2001). A breast cancer story line and public service announcement on The Young & the Restless, a popular soap opera, led to viewer calls to the Cancer Information Service hotline (Beck, 2004). Laboratory studies also provide support for the persuasive power of fiction (Green & Brock, 2000; Green, Garst, Brock, & Chung, in press; Prentice, Gerrig, & Bailis, 1997; Strange & Leung, 1999; Wheeler, Green, & Brock, 1999). Extent of belief change does not differ when individuals read materials labeled fiction versus those labeled as fact. Entertainment education is a prime example of how fictional vehicles can be used to convey real health information (see Singhal, Cody, Rogers, & Sabido, 2004; Slater, 2002). This approach may involve creating entirely new television or radio programs or altering the story lines of existing programs to integrate health themes. Entertainment education has been used around the world to address a variety of health issues (although one of the most frequent themes has been family planning, particularly in developing nations) and has resulted in both attitude and behavior change. Like other forms of narrative persuasion, entertainment education efforts must strike a delicate balance; if they do not include enough health information, the message may not be effective, but if they include too much didactic content, viewers may tune out. Stories in general, but perhaps fictional stories in particular, may seem less complicated or less threatening to recipients. Oatley (1999), following Scheff (1979), suggested that narratives provide a middle ground where emotions are experienced enough for their meaning to be understood, but, at the same time, these emotions do not overwhelm the reader. Narratives provide a safe space for individuals to explore the implications of their experiences. These opportunities may be particularly valuable in the cancer domain. Individuals may not be prepared to confront fears about their own illness or mortality but can explore these issues in the context of a fictional other person. One additional function that fictional stories may serve is by sparking real-world discussions of issues raised in the narrative. For example, in-depth interviews of viewers indicated that an ovarian cancer event on the popular ABC series thirtysomething stimulated peer discussions of cancer (Sharf, Freimuth, Greenspon, & Plotnick, 1996). Postdiagnosis, viewing a fictional narrative together may help individuals raise sensitive issues with partners or family members. Separating fact and fiction A failure of source monitoring (e.g., Johnson, Hastroudi, & Lindsay, 1993; Mares, 1996) may be one explanation for fictional influence. Source monitoring refers to keeping track of where a piece of information originated. If individuals forget whether a particular claim came from a sitcom versus a documentary, fictional information may carry more weight than it should. That is, the recalled fictional material may be treated as factual because individuals mistakenly believe that the information came from a factual source. However, the studies cited above show that fiction’s power persists even when individuals are fully aware that the material they are reading or watching comes from the mind of an author, with no responsibility to journalistic standards of truth. Determining how readers or viewers separate real information from a fictional context is an important question. Readers appear to be using a plausibility rather than an accuracy criterion; what is important is whether the characters act like real people would act, rather than whether a particular event actually took place (see Strange, 2002). Recipients also may compare fictional information to their prior knowledge on a topic. Slater (1990) suggests that fiction has a greater effect on beliefs when the topic is unfamiliar rather than familiar. Readers or viewers may be more willing or able to discount fictional sources when they have other knowledge about an issue. Thus, fictional communications may be more effective for aspects of cancer communication that are relatively less well known, such as the details of screening procedures or treatment options, rather than for issues that individuals believe they already know about (benefits of exercise or a healthy diet). Fiction as a cue to processing style My collaborators and I have proposed that fiction or narrative may serve as a cue to a reader to engage in a less critical, more immersive form of mental engagement (Green, Garst, & Brock, 2004). The idea that stories are treated differently from scientific or logical argument, and may be held to different truth standards than rhetorical messages, is not new (see, e.g., Bruner, 1986). Prentice and Gerrig (1999) suggested that individuals typically do not take a critical, elaborative approach to fiction, and that fiction has its greatest influence when readers respond experientially (i.e., with immersion and emotion) rather than rationally. We further propose that this less critical processing may take a strong or a weak form. In the weak form of fictional processing, an individual may simply disengage critical or evaluative processing. The person may be focused on relaxation, may feel that the material is not particularly important, or may simply wish to be entertained. Thus, they may be passively influenced by the communication. The strong form of fictional processing is transportation into a narrative world. The Elaboration Likelihood Model suggests that attitudes that are formed through extensive thinking (the central route to persuasion) are stronger than those formed with only shallow processing (the peripheral route); similarly, attitudes formed or changed via transportation will likely be stronger than those formed through a more passive exposure to fictional material. Although the weak form of fictional processing may be useful when individuals are particularly resistant to cancer-relevant information, it will generally be most helpful to encourage the higher levels of processing created by transportation. Transportation and behavior change models The discussion of transportation theory thus far has touched upon elements related to behavior change theories; this section will briefly highlight some of these links more explicitly. The integrated theoretical model states that a behavior is likely to occur if a person has a strong intention and the skills and abilities to perform the behavior, and there are no environmental constraints preventing the performance. Fishbein notes an implication of this model is that different approaches may be needed for individuals who have formed relevant behavioral intentions (e.g., to eat more healthy foods or to quit smoking) and those who have not (Fishbein & Yzer, 2003). For people who have not yet formed behavioral intentions, stories may provide the motivation to do so. Stories can also change attitudes, by presenting emotionally powerful information along with vivid mental images, while reducing counterarguing (Green & Brock, 2000). Schank suggests that stories may be most effective for learning (which, by extension, may include forming behavioral intentions) when individuals have encountered a surprise or an expectation failure (e.g., Schank & Berman, 2002). For example, individuals who have just found out about a risk factor for cancer or those who have gotten a test result suggesting the presence of disease may be particularly open to stories related to cancer treatment. (Of course, they may be particularly open to other forms of cancer communication as well.) This perspective suggests that one of the challenges of narrative persuasion is to create expectation failures and convince people that their previous models of the world need to be revised, which is precisely what good narratives are capable of doing. If people have already formed appropriate intentions, stories may be an aid to mental simulation that helps them translate the intentions to actual behavior, as described above. People may have increased feelings of efficacy after seeing a character like them perform the appropriate behavior (e.g., going for a mammogram, giving up fatty foods). Models of behavior change, such as the theory of reasoned action, emphasize the importance of identifying the beliefs that contribute to behavioral intentions. This step is important in narrative persuasion as well; once these beliefs have been determined, narratives can be constructed to address the appropriate beliefs. Beyond public narratives Thus far, this paper has focused primarily on public narratives—health messages conveyed through narrative advertisements, entertainment education products, or video stories. However, personal narratives can also be important in cancer prevention, treatment, and survivorship, and may be especially powerful postdiagnosis. A full review of this literature is beyond the scope of the current paper, but two areas are especially relevant. Writing and health A large body of evidence now supports the idea that writing about traumatic or emotional events in one’s life can lead to improved health. The expressive writing paradigm, pioneered by Pennebaker (e.g., Pennebaker, 2000; see Niederhoffer & Pennebaker, 2002, for a review), typically involves writing about the event or experience for 20 minutes on 3 consecutive days. (However, it is possible that other forms of writing, such as keeping a journal or participating in online support groups, may have similar benefits.) A study of prostate cancer patients suggested that individuals who engaged in expressive writing showed improvements in physical symptoms and health care utilization compared to a control group, but did not show greater immunocompetence (Rosenberg et al., 2002). Further research with cancer survivors might help identify when and for what outcomes writing will be most useful. Narratives for making meaning A cancer diagnosis is a life-changing event. Cancer patients must deal with fear, an altered sense of the future, and often a reevaluation of relationships and priorities. Constructing personal illness and/or recovery narratives can be a key step in coming to terms with and making sense of what has happened. These narratives may help give individuals some distance from their illness and allow them to better cope with the emotional experiences that accompany it (Carlick & Biley, 2004). Narratives may help provide meaning and deepen one’s self-knowledge. The form of these illness narratives can also matter; narratives that include hope for the future can aid coping (Folkman, 1997). These narratives need not come from completely within an individual. Cognitive behavioral interventions (which include a narrative component) with breast cancer patients have been shown to increase social support and immune function (Cruess et al., 2000; McGregor et al., 2004). These effects are attributed to benefit finding (e.g., cancer leading to a greater sense of meaning, reorganized life priorities, or stronger relationships). Because studies of cognitive behavioral interventions and illness narratives frequently include both narrative and other elements (optimism, positive imagery, skill building, etc.), more research is needed to identify the role of narrative per se in creating these benefits. Challenges Although narrative has the potential to be a powerful tool in the fight against cancer, communicators also face important challenges, including differing levels of literacy (with lower literacy among traditionally underserved populations), health disparities, and a crowded media environment. Literacy/health disparities One issue that cancer communicators face is that the individuals who may be most in need of information about screening, prevention, or treatment may be the ones who are most difficult to reach due to socioeconomic factors, limited literacy, or distrust of authorities. To overcome some of these barriers, health communicators using narrative approaches have taken advantage of oral traditions or personal narratives. For example, the Witness Project brings African American breast and cervical cancer survivors into church groups and local organizations to share their stories (Erwin et al., 1992). This approach not only targets otherwise hard-to-reach audiences but also adds the power of personal contact and modeling. Relying on these in person narratives does entail some loss of control; however, a pamphlet or video can be carefully designed to convey particular messages, whereas an individual person’s story may be more likely to also include idiosyncratic or erroneous beliefs about cancer. Another option is to use technology such as videotaped stories (e.g., Borrayo, 2004, who created a brief video about breast cancer screening aimed at low-literacy Latina women). The use of video presentations requires that recipients have access to a television (and possibly a VCR/DVD player); however, the large majority of American homes do have televisions. Stories can also be presented on computers or over the Internet, which has the further benefit of allowing interactivity. A study of an interactive breast cancer education CD-ROM showed that even low-income, minority women with little computer experience had positive reactions to the hypermedia health education program (Slater et al., 1994), so such interventions are not limited to high–socioeconomic status populations (see also Rogers, 2004, for a description of ongoing web-based cancer interventions among Hispanics in the southwest United States). When targeting specific populations, especially those that have traditionally been underserved, it is important to be culturally appropriate. For narratives as with other types of communications, formative research (surveys, focus groups) can be essential in identifying beliefs and barriers in the target populations. Competing messages Like other persuasive messages, narratives face a crowded media environment. This may be a particularly important challenge for cancer prevention messages because of the millions of dollars spent advertising fast food, cigarettes, and other consumer products that increase rather than decrease cancer risk. Not surprisingly, then, the effects of any single message may be quite small. For example, a survey study of reactions to a special program on testicular cancer showed that viewers had stronger behavioral intentions to perform testicular self-examination than nonviewers, but this effect was quite weak (Trumbo, 2004). However, one advantage that narratives have is that viewers may be motivated to seek them out, especially when they are presented as entertainment (e.g., an episode of ER or a favorite soap opera). Coordinating approaches Entertainment education efforts often combine narratives with other kinds of interventions—discussion groups, nonnarrative information services (such as cancer hotlines), improved access to health services, one-on-one communications from physicians, and so on. Although such efforts may be logistically challenging, the effects of narrative cancer interventions may be especially powerful when they are combined with other efforts to make cancer information and services available in communities. Conclusions In sum, narratives have the potential to both change cancer-related beliefs and motivate health behaviors. Effective narratives should be well written, have engaging characters (perhaps matched to recipients on relevant dimensions), and evoke mental imagery; they should be able to transport their readers. Even when the narratives are fictional, the health information provided should be as accurate as possible. Formative research on the existing beliefs of the target audience can also be used to help shape narrative content and ensure that the narrative will not be rejected by the audience. The transportation experience can lead to attitude change through connections with characters, reduction of counterarguing, and increases in perceptions of realism, as well as providing role models for health-promoting behavior. Transportation theory extends current message effects theories with its specific focus on narratives and fiction. It complements theories such as the Elaboration Likelihood Model, which focus on argument-based messages, and incorporates aspects of other theories, such as exemplification theory. Narratives can be useful both in leading people to form appropriate behavioral intentions and in motivating them to act on those intentions. Acknowledgments This paper benefited from helpful feedback from Joe Cappella, Barbara Rimer, and the anonymous reviewers. I am also grateful for the insights from the National Cancer Institute Working Group on Narratives in Cancer Communication. Notes 1 " Although transportation is described as a pleasurable state, the narrative events that evoke transportation do not have to be positive. Indeed, many of the most powerful stories across the ages have been about human suffering, and it is likely that stories with negative elements or outcomes can be influential in changing attitudes toward cancer-related behaviors. The enjoyment of a transportation experience does not necessarily lie in the valence of the emotions evoked by a narrative but in the process of temporarily leaving one’s one reality behind. For a more detailed discussion of this issue, see Green, Brock, et al. (2004). 2 " Although these social norm changes may occur through narrative means, as a result of identifying with characters or becoming immersed in a story, it is also possible that some portion of these effects may also be due to simple exposure to certain patterns of behavior. 3 " Counterarguments to nonnarrative persuasive messages are usually assessed through a thought-listing procedure. Thoughts listed by participants are coded as positive, negative, or neutral toward the message. However, because the points conveyed by narratives are often implied rather than directly stated, thought listing is not always an appropriate measure. Green and Brock (2000) presented a “Pinocchio circling” technique for assessing negative responses to parts of a narrative. In this procedure, participants circle parts of the story that did not ring true to them (“false notes”). References Anderson , R. B . ( 2000 ). Vicarious and persuasive influences on efficacy expectations and intentions to perform breast self-examination . Public Relations Review , 26 , 97 – 114 . Google Scholar Crossref Search ADS WorldCat Anderson , R. B. , & McMillion , P. Y. ( 1995 ). Effects of similar and diversified modeling on African American women’s efficacy expectations and intentions to perform breast self-examination . Health Communication , 7 , 327 – 343 . Google Scholar Crossref Search ADS WorldCat Baesler , E. J. , & Burgoon , J. K. ( 1994 ). The temporal effects of story and statistical evidence on belief change . Communication Research , 21 , 582 – 602 . Google Scholar Crossref Search ADS WorldCat Bandura , A . ( 1977 ). Social learning theory . Englewood Cliffs, NJ : Prentice-Hall . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Beck , V . ( 2004 ). Working with daytime and prime-time television shows in the United States to promote health. In A. Singhal, M. J. Cody, E. M. Rogers, & M. Sabido (Eds.), Entertainment-education and social change: History, research, and practice . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Block , L. G. , & Keller , P. A. ( 1997 ). Effects of self-efficacy and vividness on the persuasiveness of health communications . Journal of Consumer Psychology , 6 ( 1 ), 31 – 54 . Google Scholar Crossref Search ADS WorldCat Borrayo , E. A . ( 2004 ). Where’s Maria? A video to increase awareness about breast cancer and mammography screening among low-literacy Latinas . Preventive Medicine , 39 , 99 – 110 . Google Scholar Crossref Search ADS PubMed WorldCat Brodie , M. , Foehr , W., Rideout , V., Baer , N., Miller , C., Flournoy , R., et al. . ( 2001 ). Communicating health information through the entertainment media . Health Affairs , 20 , 192 – 199 . Google Scholar Crossref Search ADS PubMed WorldCat Bruner , J. S . ( 1986 ). Actual minds, possible worlds . Cambridge, MA : Harvard University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Carlick , A. , & Biley , F. C. ( 2004 ). Thoughts on the therapeutic use of narrative in the promotion of coping in cancer care . European Journal of Cancer Care , 13 , 308 – 317 . Google Scholar Crossref Search ADS PubMed WorldCat Cohen , J . ( 2001 ). Defining identification: A theoretical look at the identification of audiences with media characters . Mass Communication and Society , 4 , 245 – 264 . Google Scholar Crossref Search ADS WorldCat Cox , D. , & Cox , A. D. ( 2001 ). Communicating the consequences of early detection: The role of evidence and framing . Journal of Marketing , 65 ( 3 ), 91 – 103 . Google Scholar Crossref Search ADS WorldCat Cruess , D. G. , Antoni , M. H., McGregor , B. A., Kilbourn , K. M., Boyers , A. E., Alferi , S. M., et al. . ( 2000 ). Cognitive-behavioral stress management reduces serum cortisol by enhancing benefit finding among women being treated for early stage breast cancer . Psychosomatic Medicine , 62 , 304 – 308 . Google Scholar Crossref Search ADS PubMed WorldCat Csikszentmihalyi , M . ( 1990 ). Flow: The psychology of optimal experience . New York : Harper and Row . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dal Cin , S. , Zanna , M. P., & Fong , G. T. ( 2004 ). Narrative persuasion and overcoming resistance. In E. S. Knowles & J. Linn (Eds.), Resistance and persuasion (pp. 175 – 191 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Dalton , M. A. , Sargent , J. D., Beach , M. L., Titus-Ernstoff , L., Gibson , J. J., Ahrens , M. B., et al. . ( 2003 ). Effect of viewing smoking in movies on adolescent smoking initiation: A cohort study . Lancet , 362 , 281 – 285 . Google Scholar Crossref Search ADS PubMed WorldCat Deighton , J. , Romer , D., & McQueen , J. ( 1989 ). Using drama to persuade . Journal of Consumer Research , 16 , 335 – 343 . Google Scholar Crossref Search ADS WorldCat Erwin , D. O. , Spatz , T. S., & Turturro , C. L. ( 1992 ). Development of an African American role model intervention to increase breast self-examination and mammography . Journal of Cancer Education , 7 , 311 – 319 . Google Scholar Crossref Search ADS PubMed WorldCat Fabrigar , L. R. , & Petty , R. E. ( 1999 ). The role of the affective and cognitive bases of attitudes in susceptibility to affectively and cognitively based persuasion . Personality and Social Psychology Bulletin , 25 , 363 – 381 . Google Scholar Crossref Search ADS WorldCat Fazio , R. H. , & Zanna , M. P. ( 1981 ). Direct experience and attitude-behavior consistency. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 14 , pp. 162 – 202 ). New York : Academic Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Fishbein , M. , & Yzer , M. C. ( 2003 ). Using theory to design effective health behavior interventions . Communication Theory , 13 ( 2 ), 164 – 183 . Google Scholar Crossref Search ADS WorldCat Folkman , S . ( 1997 ). Positive psychological states and coping with severe stress . Social Science and Medicine , 45 , 1207 – 1221 . Google Scholar Crossref Search ADS PubMed WorldCat Gerbner , G. , Gross , L., Morgan , M., & Signorielli , N. ( 1994 ). Growing up with television: The cultivation perspective. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 17 – 41 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gerrig , R. J . ( 1993 ). Experiencing narrative worlds: On the psychological activities of reading . New Haven, CT : Yale University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Gilbert , D. T . ( 1991 ). How mental systems believe . American Psychologist , 46 ( 2 ), 107 – 199 . Google Scholar Crossref Search ADS WorldCat Green , M. C . ( 2004 ). Transportation into narrative worlds: The role of prior knowledge and perceived realism . Discourse Processes , 38 , 247 – 266 . Google Scholar Crossref Search ADS WorldCat Green , M. C. , & Brock , T. C. ( 2000 ). The role of transportation in the persuasiveness of public narratives . Journal of Personality and Social Psychology , 79 , 701 – 721 . Google Scholar Crossref Search ADS PubMed WorldCat Green , M. C. , & Brock , T. C. ( 2002 ). In the mind’s eye: Transportation-imagery model of narrative persuasion. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 315 – 341 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green , M. C. , Brock , T. C., & Kaufman , G. F. ( 2004 ). Understanding media enjoyment: The role of transportation into narrative worlds . Communication Theory , 14 , 311 – 327 . Google Scholar Crossref Search ADS WorldCat Green , M. C. , Garst , J., & Brock , T. C. ( 2004 ). The power of fiction: Persuasion via imagination and narrative. In L. J. Shrum (Ed.), The psychology of entertainment media: Blurring the lines between entertainment and persuasion (pp. 161 – 176 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Green , M. C. , Garst , J., Brock , T. C., & Chung , S. ( in press ). Fact versus fiction labeling: Persuasion parity despite heightened scrutiny of fact . Media Psychology . OpenURL Placeholder Text WorldCat Gregory , W. L. , Cialdini , R. B., & Carpenter , K. M. ( 1982 ). Self-relevant scenarios as mediators of likelihood estimates and compliance: Does imagining make it so? Journal of Personality and Social Psychology , 43 ( 1 ), 89 – 99 . Google Scholar Crossref Search ADS WorldCat Gustafson , D. H. , Hawkins , R., Boberg , E. W., McTavish , F., Owens , B., Wise , M., et al. . ( 2002 ). CHESS: Ten years of research and development in consumer health informatics for broad populations, including underserved . International Journal of Medical Informatics , 65 ( 3 ), 169 – 177 . Google Scholar Crossref Search ADS PubMed WorldCat Hammond , D. , Fong , G. T., McDonald , P. W., Brown , S., & Cameron , R. ( 2004 ). Graphic Canadian cigarette warning labels and adverse outcomes: Evidence from Canadian smokers . American Journal of Public Health , 94 , 1442 – 1445 . Google Scholar Crossref Search ADS PubMed WorldCat Johnson , M. K. , Hastroudi , S., & Lindsay , D. S. ( 1993 ). Source monitoring . Psychological Bulletin , 114 , 3 – 28 . Google Scholar Crossref Search ADS PubMed WorldCat Knowles , E. S. , & Linn , J. (Eds.). ( 2004 ). Resistance and persuasion . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kreuter , M. W. , Strecher , V. J., & Glassman , B. ( 1999 ). One size does not fit all: The case for tailoring print materials . Annals of Behavioral Medicine , 21 , 276 – 283 . Google Scholar Crossref Search ADS PubMed WorldCat Mares , M. L . ( 1996 ). The role of source confusions in television’s cultivation of social reality judgements . Human Communication Research , 23 , 278 – 297 . Google Scholar Crossref Search ADS WorldCat Markus , H. , & Nurius , P. ( 1986 ). Possible selves . American Psychologist , 41 , 954 – 969 . Google Scholar Crossref Search ADS WorldCat Marsh , K. L. , Johnson , B. T., & Scott-Sheldon , L. A. ( 2001 ). Heart versus reason in condom use: Implicit versus explicit attitudinal predictors of sexual behavior . Zeitschrift für experimentelle Psychologie , 48 ( 2 ), 161 – 175 . Google Scholar Crossref Search ADS PubMed WorldCat McCool , J. P. , Cameron , L. D., & Petrie , K. J. ( 2005 ). The influence of smoking imagery on the smoking intentions of young people: Testing a media interpretation model . Journal of Adolescent Health , 36 , 475 – 485 . Google Scholar Crossref Search ADS PubMed WorldCat McGregor , B. A. , Antoni , M. H., Boyers , A., Alferi , S. M., Blomberg , B. B., & Carver , C. S. ( 2004 ). Cognitive-behavioral stress management increases benefit finding and immune function among women with early-stage breast cancer . Journal of Psychosomatic Research , 56 ( 1 ), 9 – 11 . Google Scholar Crossref Search ADS PubMed WorldCat Nell , V . ( 1988 ). Lost in a book: The psychology of reading for pleasure . New Haven, CT : Yale University Press . Google Scholar Crossref Search ADS Google Scholar Google Preview WorldCat COPAC Niederhoffer , K. G. , & Pennebaker , J. W. ( 2002 ). Sharing one’s story: On the benefits of writing or talking about emotional experience. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 573 – 583 ). London : Oxford University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Nisbett , R. , & Ross , L. ( 1980 ). Human inference: Strategies and short-comings of social judgment . Englewood Cliffs, NJ : Prentice Hall . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Oatley , K . ( 1999 ). Why fiction may be twice as true as fact: Fiction as cognitive and emotional simulation . Review of General Psychology , 3 ( 2 ), 101 – 117 . Google Scholar Crossref Search ADS WorldCat Oatley , K . ( 2002 ). Emotions and the story worlds of fiction. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 39 – 69 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pechmann , C . ( 2001 ). A comparison of health communication models: Risk learning versus stereotype priming . Media Psychology , 3 ( 2 ), 189 – 210 . Google Scholar Crossref Search ADS WorldCat Pechmann , C. , & Knight , S. J. ( 2002 ). An experimental investigation of the joint effects of advertising and peers on adolescents’ beliefs and intentions about cigarette consumption . Journal of Consumer Research , 29 , 5 – 19 . Google Scholar Crossref Search ADS WorldCat Pechmann , C. , & Shih , C. F. ( 1999 ). Smoking scenes in movies and antismoking advertisements before movies: Effects on youth . Journal of Marketing , 63 ( 3 ), 1 – 13 . Google Scholar Crossref Search ADS WorldCat Pechmann , C. , Zhao , G., Goldberg , M. E., & Reibling , E. T. ( 2003 ). What to convey in antismoking advertisements for adolescents: The use of protection motivation theory to identify effective message themes . Journal of Marketing , 67 ( 2 ), 1 – 18 . Google Scholar Crossref Search ADS WorldCat Pennebaker , J. W . ( 2000 ). Telling stories: The health benefits of narrative . Literature & Medicine , 19 ( 1 ), 3 – 18 . Google Scholar Crossref Search ADS WorldCat Petty , R. E. , & Cacioppo , J. T. ( 1986 ). Communication and persuasion: Classic and contemporary approaches . Dubuque, IA : Brown . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , & Krosnick , J. A. (Eds.). ( 1995 ). Attitude strength: Antecedents and consequences . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Petty , R. E. , Ostrom , T., & Brock , T. C. ( 1981 ). Cognitive responses in persuasion . Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Potter , W. J . ( 1986 ). Perceived reality and the cultivation hypothesis . Journal of Broadcasting and Electronic Media , 30 ( 2 ), 159 – 174 . Google Scholar Crossref Search ADS WorldCat Prentice , D. A. , & Gerrig , R. J. ( 1999 ). Exploring the boundary between fiction and reality. In S. Chaiken & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 529 – 546 ). New York : Guilford . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Prentice , D. A. , Gerrig , R. J., & Bailis , D. S. ( 1997 ). What readers bring to the processing of fictional texts . Psychonomic Bulletin & Review , 5 , 416 – 420 . Google Scholar Crossref Search ADS WorldCat Rimer , B. K. , & Kreuter , M. W. ( 2006 ). Advancing tailored health communication: A persuasion & message effects perspective . Journal of Communication , 56 , S184 – S201 . Google Scholar Crossref Search ADS WorldCat Rogers , E. M . ( 2004 ). Delivering entertainment-education health messages through the Internet to hard-to-reach U.S. audiences in the Southwest. In A. Singhal, M. J. Cody, E. M. Rogers, & M. Sabido (Eds.), Entertainment-education and social change: History, research, and practice (pp. 281 – 298 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Rogers , E. M. , Vaughan , P. W., Swalehe , R. M. A., Rao , N., Svenkerud , P., & Sood , S. ( 1999 ). Effects of an entertainment-education radio soap opera on family planning behavior in Tanzania . Studies in Family Planning , 30 , 193 – 211 . Google Scholar Crossref Search ADS PubMed WorldCat Rosenberg , H. J. , Rosenberg , S. D., Ernstoff , M. S., Wolford , G. L., Amdur , R. J., Elshamy , M. R., et al. . ( 2002 ). Expressive disclosure and health outcomes in a prostate cancer population . International Journal of Psychiatry in Medicine , 32 ( 1 ), 37 – 53 . Google Scholar Crossref Search ADS PubMed WorldCat Rubin , A. M . ( 2002 ). The uses-and-gratifications perspective of media effects. In D. Zillmann & J. Bryant (Eds.), Media effects: Advances in theory and research (2nd ed., pp. 525 – 548 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sargent , J. D. , Dalton , M. A., Beach , M. L., Mott , L. A., Tickle , J. J., Ahrens , B., et al. . ( 2002 ). Viewing tobacco use in movies: Does it shape attitudes that mediate adolescent smoking? American Journal of Preventative Medicine , 22 ( 3 ), 137 – 145 . Google Scholar Crossref Search ADS WorldCat Schank , R. C. , & Abelson , R. P. ( 1995 ). Knowledge and memory: The real story. In R. S. Wyer , Jr. (Ed.), Advances in social cognition (Vol. VIII , pp. 1 – 85 ). Hillsdale, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Schank , R. C. , & Berman , T. ( 2002 ). The pervasive role of stories in knowledge and action. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 287 – 313 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Scheff , T. J . ( 1979 ). Catharsis in healing, ritual, and drama . Berkeley : University of California Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Sharf , B. F. , Freimuth , V. S., Greenspon , P., & Plotnick , C. ( 1996 ). Confronting cancer on thirtysomething: Audience response to health content on entertainment television . Journal of Health Communication , 1 , 157 – 176 . Google Scholar Crossref Search ADS PubMed WorldCat Sheehan , P. W . ( 1967 ). A shortened form of the Betts’ questionnaire upon mental imagery . Journal of Clinical Psychology , 23 , 386 – 389 . Google Scholar Crossref Search ADS PubMed WorldCat Sherman , S. J. , Cialdini , R. B., Schwartzman , D. F., & Reynolds , K. D. ( 1985 ). Imagining can heighten or lower the perceived likelihood of contracting a disease: The mediating effect of ease of imagery . Personality and Social Psychology Bulletin , 11 ( 1 ), 118 – 127 . Google Scholar Crossref Search ADS WorldCat Singhal , A. , Cody , M. J., Rogers , E. M., & Sabido , M. (Eds.). ( 2004 ). Entertainment-education and social change . Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Slater , M. D . ( 1990 ). Processing social information in messages: Social group familiarity, fiction versus nonfiction, and subsequent beliefs . Communication Research , 17 , 327 – 343 . Google Scholar Crossref Search ADS WorldCat Slater , M. D . ( 2002 ). Entertainment education and the persuasive impact of narratives. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 157 – 181 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Slater , M. D. , & Rouner , D. ( 1996 ). Value-affirmative and value-protective processing of alcohol education messages that include statistical evidence or anecdotes . Communication Research , 23 , 210 – 235 . Google Scholar Crossref Search ADS WorldCat Slater , M. D. , Zimmerman , D. E., Tipton , M. L., Halverson , H., Kean , T., & Rost , J. D. ( 1994 ). Delivering health information to the disadvantaged: Assessing a hypertext approach . Hypermedia , 6 ( 2 ), 67 – 86 . OpenURL Placeholder Text WorldCat Smith , S. M. , & Shaffer , D. R. ( 2000 ). Vividness can undermine or enhance message processing: The moderating role of vividness congruency . Personality and Social Psychology Bulletin , 26 , 769 – 779 . Google Scholar Crossref Search ADS WorldCat Strange , J. J . ( 2002 ). How fictional tales wag real world beliefs: Models and mechanisms of narrative influence. In M. C. Green, J. J. Strange, & T. C. Brock (Eds.), Narrative impact: Social and cognitive foundations (pp. 315 – 341 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Strange , J. J. , & Leung , C. C. ( 1999 ). How anecdotal accounts in news and in fiction can influence judgments of a social problem’s urgency, causes, and cures . Personality and Social Psychology Bulletin , 25 , 436 – 449 . Google Scholar Crossref Search ADS WorldCat Swanson , J. E. , Rudman , L. A., & Greenwald , A. G. ( 2001 ). Using the Implicit Association Test to investigate attitude-behaviour consistency for stigmatised behaviour . Cognition and Emotion , 15 , 207 – 230 . Google Scholar Crossref Search ADS WorldCat Taylor , S. E. , & Schneider , S. K. ( 1989 ). Coping and the simulation of events . Social Cognition , 7 , 174 – 194 . Google Scholar Crossref Search ADS WorldCat Taylor , S. E. , & Thompson , S. C. ( 1982 ). Stalking the elusive “vividness” effect . Psychological Review , 89 , 155 – 181 . Google Scholar Crossref Search ADS WorldCat Tellegen , A. , & Atkinson , G. ( 1974 ). Openness to absorbing and self-altering experiences (“absorption”): A trait related to hypnotic susceptibility . Journal of Abnormal Psychology , 83 , 268 – 277 . Google Scholar Crossref Search ADS PubMed WorldCat Trumbo , C. W . ( 2004 ). Mass-mediated information effects on testicular self-examination among college students . Journal of American College Health , 52 , 257 – 261 . Google Scholar Crossref Search ADS PubMed WorldCat Wheeler , S. C. , Green , M. C., & Brock , T. C. ( 1999 ). Fictional narratives change beliefs: Replications of Prentice, Gerrig, & Bailis (1997) with mixed corroboration . Psychonomic Bulletin & Review , 6 , 136 – 141 . Google Scholar Crossref Search ADS PubMed WorldCat Zillmann , D . ( 2002 ). Exemplification theory of media influence. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (2nd ed., pp. 19 – 41 ). Mahwah, NJ : Erlbaum . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC © 2006 International Communication Association