Advances in Consumer Research Volume 10, 1983 Pages 540-542
THE VIVIDNESS EFFECT: MAKING A MOUNTAIN OUT OF A MOLEHILL?
Shelley E. Taylor, University of California, Los Angeles
Joanne V. Wood, University of California, Los Angeles
Although it is widely accepted that vividly presented information is inherently more persuasive than information presented in a more pallid and dull form, the empirical evidence reveals little support for this belief. Five hypotheses that purport to explain the lack of evidence for the vividness effect are outlined and evaluated.
For the past several years, we have been exploring how vividly presented information affects attitudes. Information may be described as vivid, that is, as likely to attract and hold our attention and to excite the imagination to the extent that it is (a) emotionally interesting, (b) concrete and imagery-provoking, and (c) proximate in a sensory, temporal, or spatial way" (Nisbett and Ross 1980 p.45). We will refer to the "vividness effect" as the differentially persuasive impact that such information is thought to have on attitudes, relative to information that is presented in a more pallid and dull form. Our investigations have examined vividness effects as a function of age (Gottlieb Taylor and Ruderman 1977), type of message (Winkler Taylor and Falcone 1979), and type of appeal (Winkler Taylor Tebbetts Jemmott and Johnson 1979). In all our efforts to pin down the parameters of the vividness effect, we have been hampered by one striking problem: He nave found very little evidence for the existence of a vividness effect.
Surprised by our own failures to demonstrate a so seemingly obvious effect, we combed the available published and unpublished literature purporting to examine the impact of vividly presented information on attitudes. What we found uas more surprising still. There is very little evidence generally for a vividness effect. Information that is presented in concrete, colorful terms does not have a consistently greater persuasive impact on judgements compared to information presented in a more pallid and dull form. Information accompanied by pictures or presented via videotape is not inherently more persuasive than information presented without these visual aids. Direct, rather than vicarious experience, does not in and of itself appear to be more inherently persuasive. Although information presented in case history form does appear to be more persuasive than comparable information presented in base-rate or other statistical form, it does not appear that the increased vividness of the case history information is the mechanism by which this effect occurs rather, it appears that people under-utilize base-rate and other statistical information, rather than over-utilize case history information (see Taylor and Thompson 1982). In short, then, our review of more than fifty individual experimental studies indicates little support for a vividness effect. The results of this analysis were presented by Taylor and Thompson (1982) in a psychological review paper this year.
It would be premature, however, to assume that no vividness effect exists. There could be a number of reasons why the experimental literature fails to demonstrate a consistent or robust effect. At least five possibilities are credible; (1) researchers have obscured persuasive messages by placing them in vivid but distracting contexts thus the context, but not the message itself, is vivid (2) Vivid information is only persuasive when it produces affective responses, and few studies assessing vividness have engaged affect; (3) People remember vivid presentations better, and erroneously infer that they were also persuaded (4) People find vivid presentations to be colorful and graphic, and erroneously assume that they were persuaded by them (5) Vivid information only affects attitudes when it must compete with other, less vivid, information, for selective attention. As can be seen, three of these hypotheses assume that the absence of evidence for vividness effects occurs because experimental tests have failed to mirror the conditions under which the vividness effects occur in everyday life (Hypotheses l, 2 and 5). The other two hypotheses (3, 4) maintain that the vividness effect is itself illusory, and that people mistakenly assume they have been persuaded by vivid information because it is striking and/or memorable. In the next few pages, we will examine recent experimental evidence assessing each of these alternatives.
With respect to the first explanation, an examination of the experimental literature reveals that researchers have indeed often obscured the content of messages with irrelevant but vivid presentations. Advertisers may be guilty of this as well. 'or example, a voluptuous product endorser may lead viewers to ignore the product itself or a catchy jingle that fails to include the product name may leave the listener humming the tune all day without knowing its origin. Under such circumstances, vivid presentations are not persuasive, since the persuasive intent of the message is itself obscured. However, despite the fact that messages may be obscured by the vivid format through which they are presented, this explanation alone does not account for the lack of research evidence for vividness effects. Even when investigators have been careful to make the message itself vivid, rather than its trappings, there is no consistent effect of vividly presented information on attitudes (e.g. Borgida 1979 Winkler Taylor and Falcone 1979; Taylor Wood and Thompson in progress). Thus, this hypothesis is judged to be inadequate to explain the absence of evidence for a vividness effect.
The hypothesis most central to the theme of this panel is that vivid information is only persuasive when it produces affective responses. In our review of the vividness literature (Taylor and Thompson 1987), we noted that few studies assessing vividness have engaged affect, and those that do have rarely measured it. Indeed, some investigators have treated affect and persuasion as synonymous, using liking for a product or other attitude object as a measure of persuasion. We think this is unwise conceptually and operationally. When affect and persuasion are measured separately (e.g. self-rated arousal or emotional response for affect and belief statements for persuasion) (Miller 1969 Knower 1935; 1936), the two sets of measures are unrelated (Taylor and Thompson 1982). Though these results are by no means definitive (they are only based on three studies), they do suggest that even when affect is engaged by vividly presented information, affect is not the mediator of persuasion effects. if and when they do exist (Taylor and Thompson 1982).
Vivid information does have a modest, though consistent, effect on memory, such that vividly presented information is better recalled than is information presented in a more pallid and dull form (see Taylor and Thompson 1982). This result suggests the third hypothesis, that people remember vivid information, and as a function of this memory, erroneously conclude that they were persuaded when they were not. In one of our recent studies (Taylor Wood and Thompson in progress), subjects were exposed to a series of messages that were presented in either a vivid or non-vivid fashion. In the vivid communications, information was presented in colorful, graphic terms with short, crisp sentences; the pallid-and-dull form of the information was presented in terms that were difficult to visualize and often pedantic in tone. The topics included the effects of color on mood, the need for a state lottery, and the dangers to children created by toxic substances in the home. Recall of the information and attitudes toward the topics were assessed. Perceived persuasiveness of the message was also measured: that is, subjects were asked how much they felt they had been persuaded by the message and how much they felt others would be persuaded by the message. Results were as follows: First, as has been true in the majority of prior studies, no effect of vividly presented information on attitudes was found. In contrast, strong effects of vivid presentations on recall were found, consistent with previous evidence. However, the hypothesis that people would mistake their recall of vivid messages for persuasion was not supported. There was no relationship between memory for vivid communications and perceived persuasiveness across a variety of messages. Accordingly, it appears that Hypothesis Three is incorrect People do not erroneously infer that they were persuaded by a vivid presentation solely because they recall it better.
The fourth hypothesis suggests that people perceive vivid communications as colorful and engaging and erroneously assume that they were persuaded. We have nicknamed this hypothesis the "Carl Sagan effect" for the following reason: In an article in Time magazine a couple of years ago (October 20, 1980) on Carl Sagan, entitled "A Gift for Vividness," one of Sagan's colleagues made the following observation: "Carl is very often right and always interesting. That is in contrast to most academicians, who are always right and not very interesting" (p.68). This amusing observation serves to remind us that merely because someone is entertaining may not make him right; very possibly, many people confuse the two factors. This hypothesized independence of perceived vividness and persuasiveness and the idea that people may con.use the two prompted the following study. Subjects were exposed to a series of persuasive messages on such topics as juvenile crime and outer space travel that were presented in either a vivid or a non-vivid fashion. The degree to which subjects were persuaded was then assessed by a series of attitude measures, and the perceived persuasive impact of the messages was also measured; that is, subjects were asked whether they felt they had been persuaded by the message. The results were strongly supportive of the Carl Sagan effect. First, as in other studies, there was no overall vividness effect, vividly presented information was no more persuasive than non-vividly presented information. However, when the rated colorfulness and graphicness or information was correlated with the perceived persuasive impact, virtually all of the correlations were highly significant across several messages. That is, when subjects perceived a communication to be colorful and graphic, they perceived that they had been persuaded by it, and they perceived that others would also be persuaded by it.
The subjects may, of course, be right It may indeed be the case that they are more persuaded by messages they perceive as vivid and graphic. Indeed, it may be the persuasiveness of the message itself that leads to perceptions of the message as vivid and graphic! Accordingly, we recalculated the correlations between rated vividness and colorfulness and perceived persuasive impact, controlling for actual persuasion. Although the correlations declined substantially, most remained significant. What this result indicates is that messages perceived as graphic and colorful are, in fact, persuasive; but, above and beyond this, when people perceive a message-as colorful and graphic, they believe that they are persuaded and that others would be persuaded to a greater extent than is actually the case. We will return to this hypothesis shortly, but before doing so, let us assess the status of the fifth hypothesis.
The fifth hypothesis is that vivid information only affects attitudes when it must compete with other, less vivid information, for selective attention. This hypothesis was spawned by-a perplexing paradox, namely, that a phenomenon conceptually very similar to vividness does produce robust effects. In studies examining the phenomenon of salience, subjects' attention is differentially directed to one person in a small group rather than to others. Typically, this is accomplished either by instructions to attend to a certain individual, or because that individual possesses some attention-getting attribute, such as bright clothing (e.g. McArthur and Post 1977). A large number of these studies (see McArthur 1981 and Taylor and Fiske 1970 for reviews) shows that these attended-to individuals are remembered better, their attributes are rated more extremely, and their causal role is exaggerated, relative to others who received less attention. Why would salience effects be so strong when vividness effects are so weak? The key may be differential attention: laboratory investigations of vividness explore vividness effects under conditions of absolute attention, whereas salience research creates conditions of selective attention (i.e. one attends to one individual at the "expense" of others). Very possibly, when people are able to devote their full attention to a message, the degree to which it is vivid (or salient) may make no difference. Full attention favors processing of the information regardless of whether it is vivid or pallid and dull. However, under conditions of selective attention, information that is attention-getting (salient or vivid) may have an advantage. In normal everyday life, conditions of informational competition requiring selective attention clearly exist. Many things simultaneously vie for our attention. While riding on the subway, one may hear a radio, catch the headlines of a newspaper, or glance at a magazine, all simultaneously. Under such conditions, vividly presented information may win out over more pallid and dull information: It may compete more successfully for attention, and hence, be more fully processed than non-vivid information (Taylor and Fiske 1978). Thus, this hypothesis has the greatest potential not only to explain why the laboratory shows no evidence for vividness effects, but why so-called salience studies work and vividness studies do not.
Our most recent experimental efforts have examined this hypothesis, and unfortunately, we do not have any definitive information at this point. However, we can describe the paradigm and share several observations on it. The paradigm for examining the selective attention hypothesis involves giving the subject a task to perform, namely filling out a questionnaire, and having him or her simultaneously overhear messages that are presented either in a vivid or non-vivid fashion; again, vividness is operationalized as colorful, concrete language with short, clear sentences. After subjects have completed the questionnaire and overheard the tape playing in the background, they are asked to complete an attitude questionnaire, and their recall of the communications is assessed. The selective attention hypothesis predicts that under these conditions of competition for attention, only the vivid messages will be fully processed, and the more pallid and dull communications will simply not be noticed.
Our observation is that juggling the parameters of this paradigm has proven to be fairly tricky. Extremely engrossing tasks lead subjects to tune out vivid and non-vivid messages alike. Alternatively, rather boring tasks are abandoned by subjects in favor of all communications, whether vivid or non-vivid. The fact that the conditions necessary to test the selective attention hypothesis are themselves somewhat difficult to create suggests that they cannot in themselves create a substantial vividness effect. Let us reiterate that this impression must be considered tentative. Nonetheless, it does appear that if the hypothesized selective attention effects operate, they do so within a fairly narrow band of task constraints.
What, then, is the current status of the vividness effect, at least as our lab currently assesses it? In our opinion, the vividness effect is a weak and fragile one. It takes a confluence of carefully-constructed conditions to demonstrate its existence. If a vivid ad is more successful than a non-vivid ad, in our judgement, it may occur because of the statistical power afforded by the large number of people-exposed to the ad, and not because the effect itself is large. The weakness of the effect has not discouraged belief in its existence, however. This misplaced faith, we would argue, stems from the Carl Sagan effect: When people encounter a communication which they find to be colorful, graphic and entertaining, they erroneously assume that it has persuaded both themselves and others. Our faith in the vividness effect is indeed making a mountain out of a molehill.
"A Gift for Vividness." Time, October 20, 1980, 68.
Borgida, E. (1979), "Character Proof and the Fireside induction." Law and Human Behavior, 3, 189-202.
Gottlieb, D. E., Taylor, S. E., and Ruderman, A (1977), "Cognitive Bases of Children's Moral Judgements," Developmental Psychology, 13, 547-556.
Knower, F H. (1935), "Experimental Studies of Changes in Attitudes: I. A Study of the Effect of Oral Argument on Changes of Attitudes.' Journal of Abnormal and Social Psychology, 6, 315-347.
Knower, F H. (1936), 'Experimental Studies of Changes in Attitudes: II. A Study of the Effect of Printed Argument on Changes in Attitudes Journal of Abnormal and Social Psychology, 30, 522-532.
McArthur, L. Z. (1981), "What Grabs You? The Role of Attention in Impression Formation and Causal Attribution," in E. T Higgins, C. P. Herman, and I. P. Zanna (Eds.), Social Cognition: The Ontario Symposium (Vol. l), Hillsdale, N. J.: Erlbaum
McArthur, L. Z. and Post, D. (1977), "Figural Emphasis and Person Perception," Journal of Experimental Social Psychology, 13, 520-535.
Miller, W. C. (1969), "Film Movement and Affective Response and the Effect on Learning and Attitude Formation," Audio-Visual Communication Review, 17, 172-181.
Nisbett, R., and Ross, L. (1980), Human Inference: Strategies and Shortcomings of Social Judgement, Englewood Cliffs. N. J.: Prentice-Hall.
Taylor, S. E., and Fiske, S. T. (1978), "Salience, Attention, and Attribution: Top of the Head Phenomena," in L. Berkowitz (ed.) Advances in Experimental Social Psychology (Vol. 11), New York: Academic Press.
Taylor, S. E. and Thompson, S. C. (1982), "Stalking the Elusive Vividness Effect," Psychological Review, 89, 155-181.
Taylor, S. E., Wood, J. V. and Thompson, S. C. (work in progress), "Nailing Down the Elusive Vividness Effect."
Winkler, J. W., Taylor, S.E. and Falcone, H. T. (1979), "Persuasive Scenarios and Risk Perception Following A Nuclear Accident," unpublished manuscript, Harvard University.
Winkler, J. D., Taylor, S. E., Tebbetts, R. F., Jemmott, J. B. and Johnson, J. (1979), "Information Types, Cognitive Heuristics and Political Persuasion," paper presented at the Annual Meeting, American Psychological Association, New York.