Show Summary Details

Page of

PRINTED FROM OXFORD REFERENCE ( (c) Copyright Oxford University Press, 2013. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single entry from a reference work in OR for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 21 June 2018


Encyclopedia of Rhetoric

Daniel J. O'Keefe


has been one of the great continuing mysteries of rhetoric and related disciplines. Somehow, discourse has the capacity to move hearts and minds, to transform people and situations, in remarkably powerful ways. From the beginnings of rhetorical study, identifying underlying principles of persuasion has been a focus of attention. The flowering of the social sciences in the twentieth century quite naturally included attention to persuasion, one of two defining ends of rhetoric. [See Eloquence.] What follows is a road map to social–scientific work concerning persuasive communication. Such work represents the application of social–scientific methods to enduring questions of rhetoric—how people direct and shape belief, achieve consensus, move others to action. Sometimes this research confirms intuitions of long standing, and sometimes it yields remarkably counterintuitive findings.

However, persuasion research is not unified within any single disciplinary or conceptual framework. Research has been conducted in a number of academic fields, with few efforts after integration or connection. Nearly all the social sciences (including psychology, communication, sociology, political science, and anthropology) and related applied endeavors in which social–scientific questions and methods appear (such as advertising, marketing, and public health) contain relevant research. An appropriate overview thus will acknowledge the different guises under which persuasion-relevant research has appeared.


In one way or another, persuasion involves influencing the audience's mental state, commonly as a precursor to action. Although a number of mental states may be the focus of a persuader's attention, social–scientific persuasion research has given pride of place to attitude, understood as the general evaluation of an object, such as a policy, proposal, product, or person. Hence, much of the relevant social–scientific work concerns attitude change, because such change represents an exemplary case of rhetorical success.

This research is predominantly experimental work, in which persuasive effectiveness is assessed under systematically controlled conditions. In the simplest research design, experimental participants are randomly assigned to hear one of two versions of a given message, where the versions differ only with respect to the property of research interest; for example, the two messages might vary in the source to which the message is attributed or in the order in which arguments are presented. If the two versions differ dependably in the attitude change evoked, that difference is presumably attributable only to the property that was experimentally varied. However, a given variable may have different effects in different messages; a particular message variation might substantially enhance persuasion in one case but have little effect in another. Hence dependable generalizations about persuasive effects require evidence derived from multiple messages, either within a single study or across many studies.


Three general kinds of theories have informed social–scientific work on persuasion: theories of attitude, theories of voluntary action, and theories of persuasion proper. The first two, though not directly concerned with persuasion, have nevertheless proved influential in shaping understandings of persuasion processes.

Attitude theories.

Because persuasion research has focused especially on attitude change as a paradigm of persuasion, theories of the nature and structure of attitude have been important sources of insight into persuasion. A straightforward example is given by “expectancy-value” models of attitude. Broadly expressed, these models describe the underlying bases of attitude as consisting of beliefs about the attitude object, such as beliefs about properties of the object. Each belief has some associated evaluation, representing the perceived desirability of the attribute, and each belief is held with some degree of certainty or strength, indicating the perceived likelihood that the object has the attribute. Across beliefs, these two facets of belief (the “value” of each attribute and the “expectancy” of its association with the object) are seen to combine to yield the person's overall assessment of, or attitude toward, the object.

This image of underlying attitude structure immediately suggests a number of alternative possible strategies for attitude change. One possibility is to attempt to add some new belief (of appropriate valence) about the object; a second is to try to change the evaluation of some existing belief; a third is to try to change the strength with which some existing belief is held. Naturally, different persuasion situations will require different approaches. In one circumstance, a persuader might conclude that the audience evaluates the outcomes of the persuader's advocated policy just as the advocate wishes, but needs to be convinced that the proposed policy will actually produce those outcomes; in another circumstance, the audience may already agree about the policy's attributes, but disagree about the evaluation of those properties. It will be noticed that this way of thinking can be seen as a particular realization of the familiar rhetorical idea that successful persuasion requires adapting one's discourse to the audience's state of mind.

Functional approaches to attitude exemplify a second set of attitude theories useful to students of persuasion. These approaches suggest that attitudes can serve various psychological functions, such as defending the person's self-image, organizing information about the attitude object, and expressing the person's values. A number of different schemes have been put forward that identify and elaborate these various functions, but there is no consensus yet on any one detailed analysis. However, one broad distinction embodied in nearly all functional attitude classifications is that between symbolic and instrumental (utilitarian) functions. Attitudes focused on symbolic associations of an object—the values it expresses, the moral beliefs it symbolizes—serve symbolic functions; attitudes focused on the intrinsic properties of the object—appraising the object in terms of intrinsic attributes or consequences—serve instrumental functions. For instance, a person's favorable attitude toward a given automobile might serve mainly instrumental functions and so be based on beliefs about gas mileage, luggage capacity, and so on, or mainly symbolic ones, and so be based on beliefs about what sort of personal identity is projected by driving this car or how driving the car makes one feel.

Approached from this perspective, the key to successful persuasion is the matching of the persuasive appeal to the attitude's functional basis, and hence this approach offers another realization of the general idea that rhetorical effectiveness requires audience adaptation. If a negative attitude toward a neighborhood facility for persons with AIDS is based on a symbolic association of AIDS and homosexuality, then changing that attitude might involve providing information that heterosexuals are also susceptible to AIDS. But if that negative attitude is based on instrumental concerns about contagiousness, then presumably a different persuasive approach will be required: indeed, in such a circumstance, emphasizing heterosexual susceptibility to AIDS would presumably backfire. In a number of studies of consumer advertising, instrumentally oriented appeals (emphasizing intrinsic product qualities) have been found more persuasive than symbolically oriented appeals (emphasizing image-based considerations) when the audience's attitudes have an instrumental basis; by contrast, with attitudes that have a symbolic basis, symbolically oriented appeals have been found more persuasive than instrumentally oriented appeals.

Individual personality differences play a role in shaping the function a given attitude serves. Some persons (“high self-monitors”) are generally more concerned than others (“low self-monitors”) about the image they project, and hence their attitudes are more likely to have symbolic bases. The nature of the attitude object also constrains the kind of function served. Some objects, such as air conditioners, easily accommodate only an instrumental function, others (such as class rings) only a symbolic function. But some objects, such as automobiles, easily permit multiple attitude functions, and hence persuasion concerning such objects needs to be especially attentive to the underlying functional basis of the audience's attitudes. Different appeals will be wanted for persons whose automobile attitudes are based on beliefs about gas mileage and frequency-of-repair records than for persons whose attitudes are based on beliefs about the image projected by driving a given car.

Voluntary action theories.

A second relevant group of theories is also not directly concerned with persuasion but rather aims at identifying factors that appear to influence voluntary action. This kind of theory offers insight into persuasion indirectly, because the factors influencing behavior provide natural foci for persuasive efforts. A leading example is Fishbein and Ajzen's (1975) “theory of reasoned action,” which suggests that a person's behavioral intentions are influenced jointly by attitudinal considerations (the person's attitude toward the action in question) and by normative ones (the person's “subjective norm,” that is, the person's assessment of whether significant others desire performance of the behavior). These two factors can vary in their impact on intention; attitudinal considerations may weigh more heavily than normative ones in some circumstances, but less heavily in others.

For a persuader, the theory of reasoned action can be used to identify useful foci for persuasive efforts. For example, if adolescent tobacco use is influenced more heavily by normative than by attitudinal factors, then interventions designed to discourage such behavior should presumably give special attention to addressing those normative factors. Moreover, because the theory also provides an account of the determinants of these attitudinal and normative factors, that is, an account of what underlies each of these, it can supply even further direction to persuaders.

A good deal of research evidence has indicated the usefulness of the theory of reasoned action, to the point where it has become the standard against which potential competitors are compared. Broadly put, the question is whether some additional general factor beyond attitude and subjective norm can be identified that dependably improves the prediction of behavioral intention. Among a number of suggestions, the one with the most research support is that represented by Ajzen's “theory of planned behavior,” which recommends additionally considering the person's perceived control over the behavior, that is, whether the person thinks it is easy or difficult to perform the action. The potential merit of this addition can be seen by considering behaviors such as exercise: persons might think that exercising is desirable (positive attitude) and that significant others think they should exercise (positive subjective norm), but believe themselves incapable of performing the behavior because expensive specialized equipment is needed but not owned, the gym is far away, exercise can't be fit into one's schedule, and so forth. Plainly, in such a circumstance, reiterating the advantages of exercise is unlikely to be a successful avenue to persuasion; instead, the perceived obstacles to performance of the behavior will need to be addressed.

These voluntary action theories thus can be seen as identifying general possible points of resistance to a persuader's views—and hence as identifying general possible targets for persuasive efforts (attitude, subjective norm, perceived behavioral control). In a sense, these targets parallel the “stock issues” familiar to students of stasis theory, as these identify possible points at issue in a dispute. [See Stasis.]

Persuasion theories.

The third relevant kind of theory aims at providing an account of persuasion itself. Of these, the most prominent and successful have been “dual-process” models of persuasion, exemplified by Petty and Cacioppo's elaboration likelihood model (ELM). The ELM suggests that there are two broad “routes to persuasion”; which one is activated depends on the degree of elaboration, or issue-relevant thinking, in which the receiver engages. One is the central route, in which the outcomes of persuasive efforts are the result of the receiver's thoughtful consideration of issue-relevant material, such as the message's arguments. [See Logos.] The other is the peripheral route, in which persuasive outcomes arise from less thoughtful processes such as the receiver's invocation of some heuristic (a simplifying decision rule); for example, instead of carefully considering the arguments and evidence, a receiver might reach a conclusion based on the communicator's credibility, the communicator's likeability, or the reactions of other audience members to the message. [See Ēthos.] These two prototypical forms of persuasion actually represent the ends of an elaboration continuum; at intermediate levels of elaboration, both centralroute and peripheral-route processes may be at work.

The degree of elaboration in which a receiver engages is a function of a variety of factors. Some of these factors influence elaboration motivation, such as the receiver's degree of involvement with the message topic, that is, the personal relevance of the topic to the receiver. Others influence elaboration ability, such as the receiver's preexisting knowledge of the topic or the degree to which the persuasion setting permits undivided attention to the message.

As elaboration and, correspondingly, the kinds of persuasion processes involved varies, different factors will play a role in determining persuasive outcomes. Under conditions of high elaboration, for example, one key factor influencing the success of persuasive messages appears to be the evaluative direction of the receiver's thoughts, whether the receiver's issue-relevant thoughts are generally favorable or unfavorable to the position advocated. This in turn is influenced by, among other things, the quality of the message's arguments (their cogency, strength, importance). When elaboration is high, receivers carefully scrutinize the message's arguments, and hence the quality of those arguments becomes an important determinant of persuasive success. But under conditions of low elaboration, variations in argument quality make less difference to persuasive outcomes, and peripheral considerations—such as the receiver's liking of the communicator, or the reactions of other audience members to the advocacy—play larger roles.

One especially attractive aspect of such dual-process models is that they offer the prospect of reconciling apparently inconsistent research findings. For example, experiments examining the effects of accompanying persuasive messages with a distracting stimulus or task have found that distraction can either enhance or reduce persuasive effectiveness. From the perspective of the ELM, just such variation is to be expected, however. Because distraction interferes with elaboration, it will impair persuasive success in circumstances in which the receiver would otherwise have had predominantly favorable thoughts about the advocated view. Distraction will enhance success, however, if the circumstance is one in which predominantly unfavorable thoughts would have occurred. In a number of areas, dual-process models such as the ELM have proved quite helpful in illuminating complex research results; such models represent an important step forward in the social–scientific understanding of persuasion processes.

But dual-process-model research has been strangely inattentive to the particulars of argument quality. Experimental argument quality variations have been created by varying an amalgam of not carefully conceptualized message features, including the importance of the outcomes of the advocated action or view (e.g., strong arguments discussing important rather than trivial consequences) and the quality of the evidence provided (e.g., strong arguments invoking opinions of disinterested rather than self-interested parties). It will plainly take some time to identify the various components of argument quality, tease out their separate and joint contributions, and clarify the mechanisms by which such properties produce the observed effects.

Variable-Oriented Research.

Although some persuasion research is guided by the sorts of theories just discussed, a good deal of persuasion inquiry might more appropriately be described as not motivated by any specific theoretical framework, instead being aimed at illuminating the roles that various particular variables play in persuasion. This work can usefully be organized by whether the variable under investigation is a characteristic of the source, the message, or the receiver.

Source characteristics.

Two attributes of communicators have been especially prominent in persuasion research: credibility and liking. Credibility, the perceived believability of a communicator, is based on the conjunction of perceived competence (expertise, knowledgeability) and perceived trustworthiness (honesty, sincerity). These perceptions are influenced by knowledge of the communicator's background and circumstance: for example, training and experience, and whether the communicator is self-interested in a way that might induce bias. They can also be affected by aspects of the message or its delivery: for example, nonfluencies in delivery can diminish perceptions of competence. Credibility, thus conceived, is not quite identical to Aristotle's conception of ēthos, but plainly there is an underlying common recognition of the role that aspects of character can play in persuasion. [See Credibility.]

As one might expect, higher-credibility communicators, those perceived as more competent and trustworthy, are commonly more persuasive than lower-credibility sources, but this generalization needs to be tempered in two ways. First, the impact that credibility has on persuasive outcomes varies depending on such matters as the audience's involvement with the issue (as suggested by the ELM); credibility variations make less difference as the personal relevance of the issue to the audience increases. Second, lower-credibility communicators have been observed to be more persuasive than high-credibility sources in circumstances in which the advocated view is one toward which the audience is initially at least somewhat favorable; it appears that hearing a lower-credibility source advocate one's own viewpoint encourages (covert) compensatory arguing in the audience (something not encouraged when an apparently expert source is defending one's view), which leads to greater persuasion.

Unsurprisingly, better-liked communicators are commonly more persuasive than their less well-liked counterparts. But, as with credibility, this effect weakens as audience involvement increases. Moreover, several studies have reported effects in which disliked communicators proved more persuasive than liked communicators. This counterintuitive effect is not yet well understood, but appears to arise only when the receiver has chosen to listen to the message; it may be that when receivers have chosen to listen to a communicator who turns out to be unlikable, they search for some reason for having done so—and finding merit in the advocated view might provide such justification.

Other communicator characteristics appear to play roles in persuasion primarily through their influence on credibility and liking. For example, similarities between source and audience, or more precisely, the audience's perception of similarities between source and audience, seem to affect persuasive outcomes only indirectly, by influencing perceived credibility and liking, which then have more direct, if complicated effects on persuasion. The same appears to be true for other communicator characteristics, such as ethnicity or physical attractiveness.

Message characteristics.

A large number of different message variations have been studied for their possible contributions to persuasive effects. Three examples can serve to illustrate the nature of this research: studies of message sidedness variations, fear appeals, and conclusion explicitness.

First, considerable attention has been given to the persuasive effects of different means of handling opposing arguments. Broadly, a persuader might either ignore such arguments (what is termed a “one-sided” message) or discuss them (a “two-sided” message); if opposing arguments are discussed, a persuader either might try to undermine those opposing considerations (a refutational two-sided message) or might simply mention some opposing arguments (a nonrefutational two-sided message). A two-sided message, as understood here, thus differs from the “two-sided argumentation” associated with dissoi logoi or in ultramque partem. There has often been speculation that the choice among these alternatives should depend on such factors as whether the audience is already inclined to favor the advocated view, or whether the audience is familiar with possible counterarguments; in fact, none of these factors plays a significant role in influencing the relative effectiveness of message sidedness variations. Broadly speaking, one-sided messages are less persuasive than refutational two-sided messages, and do not differ significantly from nonrefutational two-sided messages; thus persuaders would generally be best advised to attempt straightforward refutation of possible objections. However, the advantage that refutational two-sided messages have over nonrefutational two-sided messages, when each is compared against a one-sided message, is diminished when the messages are consumer advertisements; it may be that the initial skepticism with which consumer advertising is commonly met creates a circumstance in which nonrefutational acknowledgement of counterarguments can enhance the advertisement's credibility and hence its persuasiveness.

Fear appeals have been another long-standing focus of research interest in persuasion. A fear appeal is a message designed to arouse a sense of threat in the audience, in the hope of motivating acceptance of the communication's recommended course of action, which is aimed at alleviating or avoiding fear. For example, a message might describe the terrible consequences of skin cancer, and then recommend various sun-protection actions such as using sunscreen or wearing a hat as means of avoiding these outcomes. In this research area, the message variation of interest is the intensity or explicitness of the fear-arousing material; a message might have relatively mild or relatively strong fear-arousing material, and the question is what effects such variations have on fear arousal and persuasive effectiveness. The research evidence makes it clear that it is not easy for communicators to manipulate fear levels through messages; for instance, more intense message material does not always arouse more fear. However, messages that do arouse greater fear are likely to be more persuasive than messages that arouse less fear. This effect is contrary to long-standing beliefs about the impact on persuasion of fear arousal. A common expectation has been that the impact of aroused fear on persuasion would take the form of an inverted U-shaped curve, with the greatest persuasion occurring at intermediate levels of aroused fear. But the evidence to date is inconsistent with such suppositions. Students of rhetoric will recognize fear appeals as one species of emotional appeal. [See Pathos.] Other emotional appeals, such as appeals to pity or guilt, have received less research attention than have fear appeals.

A third example of message-variable research is provided by studies of variations in the explicitness with which the message's overall conclusion is put forward. The specific experimental contrast of interest compares messages in which the conclusion is stated explicitly and ones in which it is left implicit, that is, where the audience is left to infer the message's conclusion. It has widely been thought that the relative persuasiveness of these two message forms will depend on the audience's ability and willingness to reason to the desired conclusion. Specifically, the expectation has been that with receivers who cannot draw the desired conclusion themselves, because of intellectual ability, or will not do so because of holding opposing views, explicit-conclusion messages will be more persuasive, but otherwise implicit-conclusion messages are to be preferred since such messages will invite the audience's active participation, in an enthymematic fashion. But the research evidence in hand indicates that in fact messages with explicit conclusions are generally more persuasive than those that leave the conclusion implicit, regardless of the audience's initial opinion or intellectual ability. [See Enthymeme; and Tacit dimension, the.]

This provides only a sampling of the message features that have been studied in persuasion research. Many other aspects of message organization (e.g., alternative ways of ordering argumentative materials) and content (e.g., the use of altruistic appeals, rhetorical questions, or figurative language) have also been investigated. Visual aspects of persuasive messages, however, have received relatively little systematic research attention. In some ways this is unsurprising; our vocabularies for describing verbal variation are rather better articulated than corresponding vocabularies for imagery in persuasive messages (not to mention vocabularies for describing the interplay of visual and verbal materials). Still, the prominence of visual images in persuasive messages makes it likely that this will be an increasingly significant focus for research.

Receiver characteristics.

Receiver personality traits appear to play rather complicated roles in persuasion. Many personality characteristics can either enhance or inhibit persuasion, depending on the circumstance. For example, as mentioned above in the discussion of functional attitude theories, receivers may differ in self-monitoring (the degree to which they are sensitive to the image they project), with this then predisposing receivers to vary in the degree to which symbolic or instrumental functions are served by their attitudes. Thus a high self-monitoring receiver might be either easier or more difficult to persuade than a low self-monitoring receiver, depending on whether the message's persuasive appeal matched the receiver's underlying symbolic attitude function. Studies of such phenomena (the effects of variation in receiver attributes such as self-monitoring, intelligence, self-esteem, age, and so forth) can be seen as reflecting a continuation of the rhetorical tradition's preoccupation with analyzing the roles that particular audiences' characteristics play in persuasion.

One especially significant area of research concerns how receivers can be made resistant to persuasion, for example, how voters inclined to vote for a given candidate can be made resistant to counterpersuasion by opponents, or how adolescents can be made resistant to offers of drugs. What persuades a person may be different from what makes a person resistant to counterpersuasion, and hence several lines of research have been undertaken concerning distinctive aspects of resistance-to-persuasion phenomena.

The most useful general conception of persuasion resistance relies on an analogy with disease inoculation: by exposing receivers to refutations of weak versions of opposing arguments, receivers can be made resistant to subsequent attacks. For example, in political campaign contexts, such inoculation treatments have been found to reduce the effectiveness of subsequent negative advertising. Additionally, there have been a number of studies of how best to teach children to resist social pressures to use tobacco, alcohol, or illegal drugs. One approach focuses on direct social pressures in the form of offers of such substances, and teaches children skills for refusing such offers, commonly through a combination of modeling (seeing others perform refusals) and practice (role-playing exercises in which the child refuses an offer). A second approach attempts to defuse the indirect social pressure arising from children's normative misperceptions, for example, overestimation of how many of their peers use such substances. The evidence to date gives little reason to think that refusal-skill training prevents subsequent substance abuse, but normative interventions appear more promising.


Interest in persuasion, and hence in conducting persuasion-related research, naturally arises in a number of arenas of practical activity, such as consumer advertising; political communication (e.g., election campaign messages); legal communication, such as witness testimony or attorney argumentation; and health communication, including mass media campaigns aimed at disease prevention, communications concerning biological or environmental risks, and messages communicating product warning information.

Some of the persuasion-related research in these domains represents applications of more general ideas such as those previously discussed. For example, a large number of studies have examined the usefulness of the theory of reasoned action or the theory of planned behavior for understanding various health-related behaviors: exercising, participating in health screening programs such as mammography, taking protective action against skin cancers arising from sun exposure, engaging in breast or testicle self-examination, and so on. Similarly, researchers have investigated the differences between symbolic and instrumental functions of attitudes toward persons with AIDS, the credibility that various sources of drug information have for adolescents, the effectiveness of fear appeals in encouraging seatbelt use, and so on.

Additionally, however, one can find domainspecific research, that is, research focused on questions or variables of distinctive interest in a particular area of application. For example, studies of consumer advertising have examined how one's attitude toward an advertisement (one's evaluation of the ad, as distinct from one's evaluation of the object being advertised) influences advertising effectiveness; better-liked ads are, unsurprisingly, generally more persuasive, but this effect weakens as the audience becomes more familiar with the product being advertised. Similarly, research on political campaign persuasion has given special attention to the effects of negative political advertising; such advertising has often been presumed to be especially effective, but the available empirical evidence suggests that negative political advertising is typically unsuccessful and may even be damaging to the sponsoring candidate.

Within such domain-specific research, one development of special interest for students of persuasion has been the articulation of various “stage” models of health-related behavior, exemplified by the “transtheoretical” model of health behavior, so named because putatively it integrates a number of different theoretical perspectives. The transtheoretical model (sometimes called the “stages-of-change” model) identifies a number of distinct stages in a person's adoption of a given health-related behavior such as engaging in an exercise program. In the precontemplation stage, a person is not even thinking about undertaking an exercise program anytime soon; in the contemplation stage, the person is at least seriously thinking about doing so; a person in the preparation stage is ready to change and may have undertaken planning or other preparatory action (such as signing up for a health club); in the action stage, the person has undertaken the exercise program; finally, a person who has continued to engage in exercise for some time is said to be in the maintenance stage.

From a persuader's perspective, stage models are appealing because of their potential usefulness in suggesting how best to tailor persuasive efforts to a particular audience. For example, for persons in the precontemplation stage, the persuader's challenge will be to get the audience thinking about the target behavior (i.e., moving the audience from precontemplation to contemplation). By contrast, for persons in the preparation stage, the persuader will presumably want to help the audience translate their plans and intentions into actions. These stage models thus offer yet another way of thinking about audience analysis and adaptation.

One especially intriguing finding from transtheoretical-based research concerns “decisional balance,” the perceived importance of the advantages and the disadvantages of a given action. Studies of diverse health-related actions (including using sunscreen, undergoing mammography screening, reducing dietary fat, and exercising) have found that as persons move from precontemplation to action, the perceived importance of the action's advantages increases, and the perceived importance of the disadvantages decreases. This much is unsurprising, but the research evidence indicates that these two changes are not symmetrical: the increase in the perceived importance of the advantages is substantially larger than the decrease in the perceived importance of the disadvantages. At face value, this suggests that the adoption of such behaviors may be less a matter of the person's deciding that the disadvantages are insignificant than it is a matter of deciding that the advantages make the action worthwhile; correspondingly, in encouraging movement from precontemplation to action, persuaders may wish to give less attention to undermining potential disadvantages than to increasing the perceived importance of the action's advantages.

Only recently has much research evidence begun to accumulate concerning stage models of health behavior, and a number of thorny conceptual and methodological questions remain unsettled. It is not yet clear, for example, whether the transtheoretical model's typology of stages will prove to be the most generally useful one; more generally, researchers appear not yet to have considered carefully exactly what sorts of evidence will be needed to assess the various claims implicit in stage models. Still, stage models plainly offer the prospect of continuing contributions to our understanding of persuasion processes.


One question that naturally arises is why there has not been more aggressive integration of the variable-analytic and applied-research findings within the various theoretical frameworks. One reason is simply the span of academic fields in which persuasion-relevant research is to be found. Traditional discipline-centered training models do not always encourage researchers to look abroad for relevant work, and only recently has there been much concerted effort at retrieving and organizing the scattered research literature. Moreover, extant theoretical frameworks have not attempted to speak to the broad range of relevant issues. For example, although emotional and visual aspects of persuasion are plainly important, theoretical models to date have not been designed to accommodate easily, much less focus on, such facets. One may hope that future frameworks are more expansive, both in the sense of being open to taking up a broader range of concerns and in the sense of being ready to engage relevant work across disciplinary boundaries.

At the same time, the development of larger frameworks will encounter a natural and inevitable tension in social–scientific persuasion work—a tension familiar in rhetorical studies—between general frameworks and case- or context-specific treatments. This can be clearly illustrated by considering the question of whether to add this or that particular factor to the theory of reasoned action or the theory of planned behavior, so as to enhance the prediction of intention. It is possible that adding some given factor improves the prediction of intention for one specific behavior, but does not prove generally useful across a variety of behavioral domains, and so would not be an appropriate addition to the general model. That is to say, there is some trade-off between having a parsimonious and widely applicable general account and having a maximally satisfactory account of some particular circumstance. As helpful as general images of persuasion can be, then, it will almost certainly remain the case that individual circumstances require correspondingly individualized treatment. But this will be no surprise to the student of rhetoric.

[See also overview article on Audience; Conviction; Identification; and Judgment.]


Ajzen, Icek. “The Theory of Planned Behavior.Organizational Behavior and Human Decision Processes 50 (1991), pp. 179–211.Find this resource:

    Conner, Mark, and Paul Norman, eds. Predicting Health Behaviour. Buckingham, U.K., 1996. Chapters provide critical analysis of the application of various models (e.g., the theory of planned behavior) to the explanation of health-related behavior.Find this resource:

      Eagly, Alice H., and Shelly Chaiken. The Psychology of Attitudes. Fort Worth, Tex., 1993. An excellent comprehensive treatment of the research literature on attitudes.Find this resource:

        Fishbein, Martin, and Icek Ajzen. Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, Mass., 1975. Organized by the theory of reasoned action.Find this resource:

          Jackson, Sally. Message Effects Research. New York, 1992. A careful discussion of methodological issues arising in persuasion research.Find this resource:

            Maibach, Edward, and Roxanne Louiselle Parrott, eds. Designing Health Messages: Approaches from Communication Theory and Public Health Practice. Thousand Oaks, Calif., 1995. Chapters illustrate a variety of theory-based approaches to the design of persuasive messages on health topics.Find this resource:

              Messaris, Paul. Visual Persuasion: The Role of Images in Advertising. Thousand Oaks, Calif., 1997. Thoughtful treatment of a long-neglected aspect of persuasion.Find this resource:

                O'Keefe, Daniel J. Persuasion: Theory and Research. Newbury Park, Calif., 1990. A broad survey of theory and research; a new edition is planned.Find this resource:

                  Petty, Richard E., and John T. Cacioppo. Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York, 1986. Detailed presentation of the elaboration likelihood model.Find this resource:

                    Pfau, Michael, and Henry C. Kenski. Attack Politics: Strategy and Defense. New York, 1990. Reports two field studies of inoculation treatments against negative political advertising.Find this resource:

                      Prochaska, James O., and Carlo C. DiClemente. The Transtheoretical Approach: Crossing the Traditional Boundaries of Therapy. Homewood, Ill., 1984.Find this resource:

                        Weinstein, Neil D., Alexander J. Rothman, and Stephen R. Sutton. “Stage Theories of Health Behavior: Conceptual and Methodological Issues.Health Psychology 17 (1998), pp. 290–299.Find this resource:

                          —Daniel J. O'Keefe