Advances in Consumer Research Volume 9, 1982 Pages 428-430
SEXUAL CONTENT AND ADVERTISING EFFECTIVENESS: COMMENTS ON BELCH ET AL. (1981) AND CACCAVALE ET AL. (1981)
Robert S. Baron, The Univesrity of Iowa
The relationship between sex ant advertising is examined from four perspectives: the impact of sexual content on (a) attention, (b) affect, (c) arousal, and (d) information processing. The section on attention stresses the distinction between voluntary attention and the orienting response as well as the need to use more than one ad in a treatment condition. The section on affect points out that negative affect may impair advertising effectiveness. The section on arousal reviews recent research on misattribution of arousal and points out that under the correct circumstances sexual excitement may be misinterpreted as product interest and attraction. Finally, the section on information processing points out that sexual content may trigger a shallow processing of advertising copy that will detract from a high-quality logical message while improving the impact of emotional and/or implausible copy.
As I prepared for this session on sex and advertising, I was struck by the relative scarcity of research on the topic. For example, in my own field, social psychology, there are only a handful of studies which investigate the impact that sexual stimuli have on persuasive messages. In the area of consumer behavior, research on the relationship between sexual material and advertising impact is a bit more common (than in social psychology), but even here the number of relevant studies is quite low. This is fairly surprising since the use of sexually suggestive material is an obvious and frequent strategy in advertising. Venkatesan and Losco (1975), for example, found that 65% of print ads in a 1969-71 sample of ads used women as sexual objects D One possible reason for this scarcity of research is that it is due to a theoretical emptiness. That is, it could be that the theoretical issues relevant to sex and advertising are obvious and common-sensical ones that have not required or elicited extensive research. I don't feel this is so, however. Indeed, I feel there are a number of intriguing and nonobvious research issues concerning the relationship between sexual stimuli and advertising, and I'd like to briefly discuss a few of these issues today in hopes that it may encourage research on some interesting conceptual points that presently have been ignored. My discussion will focus on four different mechanisms that may determine how sexual material affects advertising impact. The first two mechanisms are fairly straightforward, the second two are less obvious.
SEX AND ATTENTION
It seems quite likely that sexual stimuli in an ad will elicit attention to it. At the very least, blatant sexual content should elicit an orienting response (OR)--i.e., a short-term increase in attention--since explicit sexual content has been relatively unusual in the mainstream media (at least up until recently). Beyond acting as an initial attentional lure, it seems plausible that sexual material could hold one's attention for a longer period given that the models are attractive, the scene is amusing or-pleasant, etc. It also seems likely that advertising effectiveness will be heightened less by a momentary OR than by a more prolonged attentional response. One danger in using only electrodermal measures to assess attention to ads which are briefly presented is that it is hart to differentiate between a momentary OR and a more lasting voluntary attentional response when looking exclusively at electrodermal response. For example, it is possible that an individual could emit a strong but momentary OR to a sexual ad (as reflected in skin conductance) and yet experience little change in cognition or preference. One way to minimize this problem is to obtain cognitive measures as well as electrodermal ones. This, of course, is the strategy used by Belch, Holgerson, Belch, and Koppman (1981) in one of the papers we just heard. This study is generally consistent with the notion that sexual ads elicit more attention (as indicated by GSR) and a stronger cognitive response than nonsexual ads. A number of comments seem warranted regarding this paper. First, the reader would be helped by more detail regarding procedure (how was GSR measured, where were electrodes placed, etc.) and the nature of the statistical tests. Secondly, the row differences in Tables l and 2 need to be subjected to statistical test. While there is little doubt that many row differences reported in these tables will be significantly different from each other and from baseline, it is hard to fully assess this paper without more complete statistical analyses. Third, while the authors should be applauded for their use of actual ads, a real problem with the use of in vivo advertising stimuli culled from the media is that it is possible that the degree of sexuality of an ad may be subtly confounded with other powerful but nonsexual variables. For example, a fully clothed image of Mrs. Olsen in a Folger's commercial differs in innumerable ways from Meredith Baxter Birney's image in the L'Oreal commercials. Only a few of these differences concern the fact that Meredith's appearance and her "I'm worth it" line are sexually suggestive; age is different, accent is different, product is different, the use of close-up photography is different, etc. I admit after seeing the ads Belch et al. used, I'm reassured that the stimuli look to be equated for such things as age, attractiveness, and degree of close-up use, but a far superior strategy is to employ several ads in each experimental category. Thus, I would recommend using, say, four female nude ads rather than one. This use of a sampling drops the probability that all ads in the category share some unplanned confound.
Finally, the use of cognitive measures in conjunction with GSR must be viewed as a strength of this study. While the cognitive data are not as consistent as the GSR data, there is some basis to infer that at least some sexual ads (female nudes and semi nudes) have an impact at the cognitive as well as physiological level.
SEX AND AFFECT
Another straightforward hypothesis is that sexual material may have an impact by eliciting affect. A positive affective reaction to a sexual at (such as lust or attraction) should increase persuasive impact via classical conditioning, with the opposite occurring if the ad elicits negative feelings (such as disgust, embarrassment, or uneasiness). [A closely related notion is that individuals are more persuadable if they are in a good mood and less persuadable if they are in a bad mood.]
A study by Peterson and Kerin (1977) can be interpreted in this light. In this study total female nudity in an ad lowered product evaluations. Indeed, just suggestive clothing on the female model also lowered product evaluations provided that sexual content was completely irrelevant to the product. In both these cases, the sexual content seems likely to have elicited negative affect such as surprise, disapproval, or shock. However, a suggestively clothed female model increased product evaluations when sexual themes were somewhat relevant to the product category. An unpublished study by Zimbardo, Ebbesen, and Fraser (described in Zimbardo & Ebbesen, 1969) makes a related point. In this study, undergraduate male subjects were distracted with slides as they listened to a message. The affective nature of the slides were varied. Positive slides depicted attractive female semi-nudes, while the most negative slides were gruesome medical photos. In accord with the conditioning hypothesis, positive slides heightened persuasive impact, while negative slides reduced the persuasiveness of the message (compared to a neutral control group).
The research and writing on subliminal implants is closely related to this classical conditioning perspective. Key (1976), for example, argues that one reason subliminal sexual implants heighten an ad's impact is that they generally elicit an emotional response which becomes associated with the product particularly after repeated exposure. This associative process, of course, is just a description of a classical conditioning procedure. While the power of classical conditioning is hardly a controversial issue, Key's notion that subliminal implants involving sex will generally elicit an emotional response which, in turn, will increase an ad's effectiveness must be viewed as speculation. The paper we just heard by Caccavale, Wanty, and Edell obviously offers little support for Key's arguments.
As the authors note, their study does have some limitations. For example, we are not sure whether the subliminal implants were too subliminal, especially since the ads were not presented repeatedly. Similarly, there is some question whether the implants and the sexually suggestive copy were "sexy" enough. Finally, the use of specially created "non-natural" ads introduces questions regarding external validity. In rebuttal, however, the obscene word implants used here are quite similar to those alluded to by Key, and the use of experimental ads permits a high degree of control over possible confounds. Perhaps the best strategy would be a convergent approach in which artificial ads were used in one study and actual ads, perhaps culled from Key's own examples, were used in subsequent research. If both types of stimuli produced similar data patterns, one's confidence in the data would be substantially greater.
Clearly, more research is required. No matter how fanciful Key's arguments appear to experienced researchers, assertions regarding the power and insidious influence of subliminal implants have continued to capture the public's curiosity. As a result, social scientists will be increasingly embarrassed if they cannot offer some opinions that are based on empirical fact.
SEX AND MISATTRIBUTION OF AROUSAL
A third and less obvious reason sexual material may affect ad effectiveness is that it is possible that people may misinterpret physiological activity caused by sexual content. A good deal of research in social psychology in the last five years supports the Schachterian prediction that individuals on occasion will misinterpret physiological activity, thereby experiencing increases in such reactions as anger, aggression, sexual arousal, humor, and attitude change. For example, Zillmann and his associates have found insulted subjects are angrier and more aggressive if they were aroused either by exercise or by an obscene film just prior to being insulted (Zillmann, 1971; Zillmann & Bryant, 1974). Cantor, Bryant, and Zillmann (1974) have also found that subjects have a greater reaction to humorous stimuli if they first have read highly erotic or aggressive passages designed to heighten excitement. Pittman (1975) found that attitude change caused by essay writing was greater if subjects were aroused by threat of shock before writing the essay. Similarly, Cooper, Zanna, and Taves (1978) found that increasing physiological activity of subjects with amphetamines also led to more attitude change after essay writing. Finally, Cantor, Zillmann, and Bryant (1975) found that males thought an obscene film to be more erotic if they were first aroused by cycling. These studies are all consistent with the view that physiological activity created by one source can be perceived as being caused by another source. In addition, they indicate this misattribution of arousal can elevate attitude change. Given that sexual advertising material will at least occasionally cause increases in physiological activity, this research has clear implications for research on sex and advertising. When conditions favor misattribution of arousal (cf. Cantor, Zillmann, & Bryant, 1975; Cooper, Zanna, & Taves, 1978), any physiological activity created by sexual stimuli may be interpreted as interest in or attraction to the product. Indeed, sexual arousal created on page l may contribute to the appeal of a product advertised on page 5. As of yet, I know of no research investigating these precise issues, but given the recent support for closely related phenomena in social psychology, there is ample basis to make such predictions with a good deal of confidence.
SEX AND INFORMATION PROCESSING
A fourth way in which sexual material may affect advertising impact is by altering information processing strategies and attentional capacity. This argument assumes that sexual material will often act as a distraction drawing attention from the message of the ad. Stated differently, sexual material may absorb attentional capacity, making the ad's message a lower priority input than the sexual material. If so, this should trigger what Petty and Cacioppo (1981) have referred to as peripheral processing taking cognitive shortcuts in evaluating the message, i.e., reacting to such superficial things as attractiveness and credibility of the message source, number of arguments, audience reaction, and affective reactions. This process loosely resembles Zajonc's notion of noncognitive preference change alluded to in his talk earlier today. Peripheral processing is contrasted with central processing, i.e., carefully weighting the pros and cons of a message and deciding one's preference on the basis of this consideration. If sexual material makes peripheral processing more likely, this can increase persuasion if the message in fact is low quality. But if the message is high quality, distracting subjects with sexual content can draw attention from these positive message features and prevent them from thinking about and elaborating these points.
Petty, Wells, and Brock (1976) demonstrated just such an effect using a nonsexual distraction. When high-quality, logical arguments were used in a message, a mechanical distraction during the message decreased positive cognitive thoughts and lowered attitude change. If, however, the message contained weak and illogical arguments, distraction decreased negative (counterargument) thoughts and increased persuasion. In short, distraction led to less careful processing of the message. There is no reason to think this will be less true with sexual distractors. If so, ad copy embedded in sexual content is likely to be evaluated less assiduously and agreed with for more emotional and superficial reasons than is true in nonsexual ads. In addition, since preferences acquired through peripheral processing have fewer strong cognitive supports, there is reason to predict that such preferences will be less resistant to change. Again, I know of no research specifically addressing these hypotheses as they apply to sexual material and advertising. However, given the emerging support for Petty and Cacioppo's conceptual distinction (Petty & Cacioppo, 1981), it seems likely to prove a fruitful area for research.
In summary, I have outlined four mechanisms that may bear upon the impact that sexual material has on advertising effectiveness. Two of these mechanisms--heightened attention and heightened affect--have already elicited some research attention. Two other mechanisms--misattribution of arousal and alterations in information processing--have been largely ignored. Research on all these mechanisms seems more than warranted given the heavy reliance of advertisers on sexual stimuli to publicize and market their products.
Belch, Michael A., Holgerson, Barbro E., Belch, George E., and Koppman, Jerry (1981), "Psychophysiological and Cognitive Responses to Sex in Advertising," Advances in Consumer Research, Vol. IX.
Caccavale, John G., Wanty, Thomas C., III, and Edell, Julie A. (1981), "Subliminal Implants in Advertisements: An Experiment," Advances In Consumer Research, Vol. IX.
Cantor, Joanne R., Bryant, Jennings, and Zillmann, Dolf (1974), "Enhancement of Humor Appreciation by Transferred Excitation," Journal of Personality and Social Psychology, 30, pp. 812-821.
Cantor, Joanne R., Bryant, Jennings, and Zillmann, Dolf (1975), "Enhancement of Experienced Sexual Arousal in Response to Erotic Stimuli through Misattribution of Unrelated Residual Excitation," Journal of Personality and Social Psychology, 32, pp. 69-75.
Cooper, Joel, Zanna, Mark P., and Taves, Peter A. (1978), "Arousal as a Necessary Condition for Attitude Change Following Induced Compliance," Journal of Personality And Social Psychology, 36, pp. 1101-1106.
Key, Wilson B. (1976), Media Sexploitation, Englewood Cliffs, NJ: Prentice-Hall.
Peterson, R. A., and Kerin, R. A. (1977), "The Female Role in Advertising: Some Experimental Evidence," Journal of Marketing, 41 (3), pp. 59-63.
Petty, Richard E., and Cacioppo, John T. (1981), Attitudes and Persuasion: Classic and Contemporary Approaches, Dubuque, IA: William C. Brown.
Petty, Richard E., Wells, Gary L., and Brock, Timothy C. (1976), "Distraction Can Increase or Decrease Yielding to Propaganda," Journal of Personality and Social Psychology, 34, pp. 874-884.
Pittman, Thane S. (1975), "Attribution of Arousal as a Mediator in Dissonance Reduction," Journal of Experimental Social Psychology, 11, pp. 53-63.
Venkatesan, M., and Losco, J. (1975), "Women in Magazine Ads: 1959-1971," Journal of Advertising Research, 15, pp. 49-54.
Zillmann, Dolf (1971), "Excitation Transfer in Communication-Mediated Aggressive Behavior," Journal of Experimental Social Psychology, 7, pp,. 419-434.
Zillmann, Dolf and Bryant, Jennings (1974), "Effect of Residual Excitation on the Emotional Response to Provocation and Delayed Aggressive Behavior," Journal of Personality and social Psychology, 30, pp. 782-791.
Zimbardo, Philip G., Ebbesen, Ebbe B., and Fraser, Scott C. (1969), "Negative Emotional Distraction and Attitude Change: Horror, Disgust and Inhibition." Unpublished manuscript, Stanford University. Described in P. G. Zimbardo and E. B. Ebbesen, (1969) Influencing Attitudes and Changing Behavior, Reading, MA: Addison-Wesley.