A Comparison of Advertisement Effects: Sexy Female Communicator Vs Non-Sexy Female Communicator

Gordon L. Patzer, Virginia Polytechnic Institute and State University
ABSTRACT - Behavioral science research strongly suggests that the appearance of an individual affects the perception of and reactions to the individual. This study attempts to determine the influence of one aspect of appearance--sexiness--upon marketing communications. Advertisement mock-ups with a female communicator were presented to male and female receivers. Attitude measures of advertisement effectiveness, perceptions of the product, and perceptions of the communicator were assessed. The results are discussed in terms of several theoretical explanations and implications for the marketing practitioner.
[ to cite ]:
Gordon L. Patzer (1980) ,"A Comparison of Advertisement Effects: Sexy Female Communicator Vs Non-Sexy Female Communicator", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 359-364.

Advances in Consumer Research Volume 7, 1980     Pages 359-364


Gordon L. Patzer, Virginia Polytechnic Institute and State University

[The author thanks Kent B. Monroe for his advice, comments, and encouragement.]


Behavioral science research strongly suggests that the appearance of an individual affects the perception of and reactions to the individual. This study attempts to determine the influence of one aspect of appearance--sexiness--upon marketing communications. Advertisement mock-ups with a female communicator were presented to male and female receivers. Attitude measures of advertisement effectiveness, perceptions of the product, and perceptions of the communicator were assessed. The results are discussed in terms of several theoretical explanations and implications for the marketing practitioner.


The use of a "sexy female" for a communicator in advertisements is being employed increasingly. This technique is apparent in all the media and with a large number of products and companies. Little marketing research has addressed the influence of a sexy female communicator upon advertisement effectiveness. As a consequence, the rationale for the use of sexy posed and sexy dressed females is based on rules of thumb and hunches developed over a period of years in some ad hoc manner. This relationship between a sexy female communicator and an advertisement's effectiveness is the focus of this paper.

Marketing Research

Sexy females in advertisements have caused much discussion about the ethical and societal issues involved; yet, among all the discussions and "best-guesses" little is known about the effects of sexiness in advertising (Peterson and Kerin, 1977). The first attempt to review and research literature in this area of sexy female communicators and advertising effectiveness was presented by Wilson and Moore (1979) at the 1978 conference of the Association for Consumer Research. Rather than duplicate the efforts of Wilson and Moore, and to conserve editorial space, the reader is referred to the Wilson and Moore article for a review of the research literature in this area.

Current Study

Both the psychological and marketing research investigating the effects of communicator sexiness upon persuasive communication effectiveness have severe limitations. Probably, one of the most serious criticisms of both the marketing and psychology research is the lack of theoretical understanding. Though Wilson and Moore (1979) propose several theoretical considerations in their review of the research, prior researchers have been neither motivated by theory, nor have they used theory to explain their research findings. The purpose of this current study is to expand the knowledge of the effect of sexiness within marketing communications and within a theoretical framework.

This study investigates the influence of (dressed) sexy communicators on advertisement effectiveness, as measured by attitudinal measures. The independent variables are (1) the sexy and non-sexy conditions of a female communicator, and (2) the sex of the receivers (male and female). The independent variable is a number of measures of recall and receiver's perceptions of and attitudes toward the advertisement, the product, and communicator. The test of advertisement effectiveness is based on three attitudinal components that have been successfully used by companies to test advertising copy (Baker and Churchill, 1977). The three components and the respective measures of each are: (1) cognitive (believable, informative, and clear); (2) affective (interesting, appealing, impressive, attractive, and eye-catching); and (3) conative (try product, buy product, and seek-out product).

To best provide a theoretical context for this study, a number of theories were used to both formulate the hypotheses and explain the results. The theories selected are not competing, but rather are complementary when applied to this research area. The following hypotheses were tested:

H1:  An advertisement with a sexy female communicator will be more effective (for each attitudinal component) then with a non-sexy communicator.

The theoretical support for this hypothesis is distraction theory (Festinger and Maccoby, 1964). According to distraction theory, a receiver generates internal counterarguments when presented with a persuasive communication. But when persuasive communication is combined with a distraction, the ability of the receiver to generate internal counterarguments is reduced and the persuasive attempt is more effective than when no such distraction is presented. In this study the sexiness of the communicator is expected to serve as a distraction such that the persuasive communication effectiveness in the sexy communicator condition will be greater then in the non-sexy communicator condition.

H2:  The perceived sexiness of the communicator will be negatively correlated with perceived physical attractiveness for the female receives and positively correlated for the male receivers.

The primary purpose of this hypothesis is to yield an answer to an empirical question which is indirectly suggested both empirically and theoretically. Cohen and Saine (1977) report that males form more positive impressions of females than males, and females form more positive impressions of males than females. Though their results are based on a study unrelated to the present variables of interest, they do indicate the hypothesized effects. The theoretical basis for hypothesis two is mediated generalization which occurs in learning theory (Lott, 1955). According to this concept of learning theory the attitude which subjects hold toward others in "everyday life" will generalize to be the same to others who are communicators in advertisements.

H3:  Perceived communicator sexiness will be positively correlated with perceived communicator credibility, perceived expensiveness of product, and recall of advertising copy details.

Learning theory (e.g., Tarpy and Mayer, 1978, pp. 17-50) provides the base for hypothesis three. First, the positive correlation predicted between sexiness and communicator credibility is suggested by classical conditioning, in which the receivers will associate females with the advertised product (body soap). The result is that a sexy female may be perceived as more credible about such a product as body soap than a non-sexy female. Second, based on past experiences the receivers are predicted to associate the sexiness of a person with expensiveness of items surrounding such a person. In this situation, it is a product promoted by a sexy female communicator. Third, the recall of copy details will be greater with a communicator of higher credibility because the receivers will pay more attention. Therefore, following from the first part of this hypothesis regarding source credibility, there will be greater recall of copy details with the sexy rather than the non-sexy communicator.

H4:  Perceived communicator sexiness will be positively correlated with perceived communicator intelligence, trustworthiness, and expertise.

This fourth hypothesis is based on consistency theory (e.g., Festinger, 1957). Hypothesis three predicted that communicator sexiness will be positively associated with communicator credibility. According to consistency theory, if the receiver holds positive attitudes about the communicator's credibility, the receiver is likely to hold similar attitudes about other characteristics regarding the communicator. The result is that the receivers will be consistent, such that, if perceptions of high communicator credibility are held, then the communicator will also be perceived to possess higher intelligence, be more trustworthy, and to have greater expertise than communicators of low credibility. Following from hypothesis three the communicators of higher and lower credibility will be the sexy and non-sexy communicators, respectively.

A secondary purpose of this fourth hypothesis is to test the general hypothesis suggested by the source credibility literature (e.g., McGuire, 1969, pp. 136-314). Generally, the literature proposes that a communicator's credibility is primarily dependent on, or certainly highly correlated with, the trustworthiness and expertise of the communicator. This fourth hypothesis combined with the third hypothesis provides an indirect approach to the empirical issue concerning the influence of source trustworthiness and source expertise on source credibility.


An experimental research design was used to test the hypotheses. The justification for an experimental design to investigate the effects of communicator sexiness on advertisement effectiveness is suggested by Baker and Churchill (1977): (1) the causal nature of appearance has been well documented, (2) appearance can be manipulated without affecting internal or external validity, (3) interactions can be analyzed, and (4) people are not aware or are not willing to admit the influence that appearance has in their responses (e.g., Miller and Rivenbark, 1970; Bryne, 1971), and an experiment allows the research manipulation of appearance or sexiness to be disguised.

Design.  The design was a 2 X 2 factorial with two independent variables: communicator's appearance (sexy and non-sexy) and respondent's sex (male or female). Respondents were randomly assigned to one of the two experimental conditions, controlling for equal cell sizes. The communicator was a female with experience as a model in advertising. Her photos were from a modeling agency composite, which resulted in a professional and typical advertisement look--increasing the authenticity of the advertisement mock-ups. To control for inter-person differences of physical characteristics the same person was used in both communicator appearance conditions.

Manipulation of the appearance condition was achieved by use of two black-and-white photos (2 2" X 4"). The two conditions of appearance were determined by a Delphi method where at least six modeling agency and advertising professionals defined each photo as either sexy or non-sexy. To determine if the appearance manipulation in this study was successful a manipulation check was included on the response form. The communicator's appearance may be defined in terms of dress, pose, and stance, which were liberal, seductive, and suggestive for the sexy condition and moderate but fashionable, relaxed, and happy for the non-sexy condition, respectively.

Subjects.  Subjects were 30 males and 30 females randomly selected from the university main library and student center. To approximate a naturalistic setting the subjects were selected from the magazine reading rooms of each. The product advertised was selected because it is a product normally bought by the subjects (college students).

Materials.  The materials consisted of a sealed box and a folder with three pages: (1) an introduction, (2) an advertisement mock-up, and (3) a response form. The box, with a small opening, was used by the respondents to slip their response forms into. The introduction page, typed on departmental stationery, thanked the subject for participating, gave instructions, and assured total anonymity of the subject's responses. The advertisement mock-up page presented a photo of the communicator combined with typical copy for a new (non-existent) body soap. The response page used a seven point semantic differential scale to assess: (1) affective, cognitive, and conative attitudinal effects of each advertisement condition, (2) perceptions of the communicator's sexiness, physical attractiveness, credibility, trustworthiness, intelligence, and expertise, and (3) perceptions of relative price of the product. Four aided questions were used to assess recall of details stated in the advertisement copy.

Procedure.  Potential subjects were approached and asked if they would mind taking a few minutes to participate in a marketing communications study. Upon approval the subject was handed the folder with the three pages and instructed to read the first page. If, after the first page, there were no questions the person in charge left for a short period of time, returning later to pick up the materials.

Analysis.  Consistent with earlier studies (e.g., Baker and Churchill, 1977), the multiple measures for each attitudinal component were summed and averaged to attain a single score for each component. Analysis of variance was performed separately for each set of affective, cognitive, and conative component scores, and on the sexiness manipulation check. T-tests were performed on each measure to investigate differences between groups. In addition, Pearson Product-Moment Correlations were conducted to determine correlations between perceived sexiness with measures of physical attractiveness, intelligence, credibility, trustworthiness, expertise, perception of product price, and recall of advertisement details stated in the copy.




Manipulation Check

Analysis of variance was performed to determine the success of the manipulation of the sexiness variable. When the data was analyzed without regard to sex of receivers the results were significant (F=14.4, p=.001). However, a more meaningful analysis was achieved by analyzing the data according to sex of subjects. Analysis of variance (Table 1) showed strong main effects for the sexiness conditions, the sex of the receivers, and interaction effects.



Figure 1 indicates that the overall group effect shown in the analysis of variance was due to a large impact on male receivers (t=6.25, p=.000) and no impact on the female receivers (t=-.70, p=.489) by the communicator. This difference between the male and female receivers illustrates the strong interaction effects indicated in the two-way analysis of variance. Because of this substantial difference, all further results are presented for each sex individually, rather than for the respondents as a whole.

Hypothesis One

The first hypothesis states that a sexy female communicator will be more effective than a non-sexy female communicator. Support for this first hypothesis is mixed. The data for the male receivers are supportive of this hypothesis, but the data for the female receivers are not. Possible, this lack of effect upon females was due to their failure to perceive a difference of the communicator's sexiness between conditions. Regardless of the non-significant differences for the female receivers perception of sexiness, the sexiness manipulation (sexy and non-sexy conditions) did have an impact on both the males and females; consequently, the data for both sexes are reported.

Affect.  Analysis of variance (Table 1) performed with the affective component scores indicate significant main effects for the sexiness conditions and the sex of the receivers, and interaction effects between conditions and receivers. The mean scores of the affective attitudinal component are reported in Figure 2. The affective measures of the males were significantly higher (t=2.83, p=.009) for the sexy versus the non-sexy communicator. The affective scores for the females were in the same direction as the males but did not approach significance (t=.71, p=.486).

Cognitive.  Analysis of variance (Table 1) indicated the sexiness manipulation caused mixed results for the cognitive attitudinal component. The sexy and non-sexy conditions were not significantly different while the sex of the receivers was significant, resulting in significant interaction effects of the conditions by the receivers. The cognitive component for the sexy conditions was significantly higher (t=2.18, p=.038) for the males than the non-sexy condition. In contrast, the effect was opposite for females, where the cognitive component was significantly lower (t=4.40, p=.000) for the sexy condition than the non-sexy condition. These effects upon the cognitive component are presented in Figure 3.





Conative.  The results of the conative attitudinal component are comparable to the cognitive results. The sexiness conditions were not significantly different while the difference between the sexes was significant, resulting in significant interaction effects. The sexy condition produced significantly higher conative measures for the males (t=2.73, p=.011) and significantly lower measures for the females (t=-3.12, p=.004), than the non-sexy condition. The mean scores for the sex of subjects and interaction effects are presented in Figure 4.

Hypothesis Two

The second hypothesis states that the perceived sexiness of the communicator will be negatively correlated for female receivers and positively correlated for male receivers. The data for the males supported this hypothesis while the data for the females did not. The data were analyzed according to sex by using Pearson Product-Moment Correlations. Perceived sexiness and physical attractiveness of the communicator were not correlated for females (Pearson r=-.14, p=.466), but was positively correlated for males (Pearson r=.53, p=.003).



Hypothesis Three

The third hypothesis states that perceived sexiness of the communicator will be positively correlated with perceived communicator credibility, perceived expensiveness of the product, and recall of copy details. No significant correlations resulted for females. For males, the perceptions of sexiness was significantly correlated with communicator credibility (Pearson r=.49, p=.007), and perceived expensiveness of the product (Pearson r=.65, p=.001), but was not significant for recall (Pearson r=-.32, p=.089).

Hypothesis Four

The fourth hypothesis states that the perceived sexiness of the communicator will be positively correlated with perceived communicator intelligence, trustworthiness, and expertise. Again, no significant correlations were found for the female receivers. For males, the correlations between sexiness and communicator intelligence (Pearson r=.49, p=.006), trustworthiness (Pearson r=.41, p=.024), and expertise (Pearson r=.63, p=.001) were all significant.

The secondary, empirical, issue of this hypothesis concerning correlations for measures of trustworthiness and expertise with measures of source credibility is confirmed. For female receivers, trustworthiness was highly correlated with source credibility (Pearson r=.99, p=.001), and the correlation of expertise with credibility was also significant (Pearson r=.84, p=.001). Male receivers, almost identical to females, displayed highly significant correlations for source trustworthiness with source credibility (Pearson r=.96, p=.001), and source expertise with source credibility (Pearson r=.83, p=.001).


Summary of Results

General Comments.  The results did not completely support the hypotheses. The analysis of the data showed perceived female sexiness to be a determinant of advertisement effectiveness for male receivers but not for female receivers. This greater impact of a sexy female communicator upon male receivers than female receivers was not hypothesized but it is consistent with other studies (Miller and Rivenbark, 1970; Berscheid et. al., 1971). The fact sexiness did not have similar effects on female and male receivers may be due to same-sex and opposite-sex effects. Baker and Churchill (1977) investigating the effects of physical attractiveness in advertising found the communicators of opposite sex to the receivers were more effective than same sex communicators. Cohen and Saine (1977) have also reported that males form more positive impressions of females than of other males, and females, in turn, form more positive impressions of males than they do of other females. Future research must address this issue by using a sexy male communicator with both male and female receivers.

Specific Comments.  First, the manipulation check produced results that suggest males and females employ different standards to rate the sexiness of another female. But yet, even though the female receivers appeared not to perceive the communicator differently in the sexy and non-sexy conditions, they, as well as the males, responded differently to the sexy communicator than to the non-sexy communicator.

It is interesting to ponder if the opposite-sex-same-sex effects obtained for the affective component of hypothesis one would result with a sexy male communicator. In other words, would male subjects fail to perceive the difference between a sexy and a non-sexy male communicator, while female subjects perceive and respond very positively to the sexy male communicator. Additional comments regarding hypothesis one can be made in response to the results from the cognitive and conative attitudinal components. Apparently, males find advertisements more believable, informative, and clear while females find advertisements more unbelievable, uninformative, and confusing when a sexy female communicator is used than when a non-sexy communicator is used. According to the conative attitudinal component the behavior of males is positively influenced by the sexiness of a female communicator while the behavior of females is negatively influenced. The implications of each of these attitudinal components are consistent--female receivers are negatively affected and male receivers are positively affected by a sexy female communicator.

The data of hypothesis two indicate that males equate female sexiness with physical attractiveness. The data, also, indicate that females do not equate the sexiness of another female with physical attractiveness. As suggested earlier, this relationship between sexiness and physical attractiveness may be dependent upon same-sex and opposite-sex dyads.

Furthermore, the data of hypotheses three and four, again, support the notion that the sexiness of a communicator in an advertisement does have an impact on both product characteristics and communicator characteristics as perceived by the receiver. This influence is apparently more substantial for a male receiver than a female receiver; this at least true with a female communicator. Males appear to attribute very favorable characteristics to a sexy female communicator in an advertisement, to the extent that source (communicator) credibility is substantially higher for the sexy than the non-sexy communicator.

In summary, regardless of the perception of a female communicator's sexiness, as measured in this study, there is an impact on advertisement effectiveness. This impact is especially pronounced with the female receivers who showed no significant correlations between perceived sexiness and the measures taken, but, yet responded very negatively on the measures taken for the sexy female communicators. For male receivers the effects were consistently favorable to the sexy female communicator.

Theoretical Explanations

Learning Theory.  Classical conditioning (e.g., Tarpy and Mayer, 1978) explains the results of this study. Due to prior learning and/or some Freudian drive, males are attracted to sexy females in our society. Classical conditioning suggests that these males who value sexiness will also value objects associated with the valued (sexy) communicator. As expected, by learning theory, the valued (sexy) communicator received greater attention and was more effective than the non-sexy communicator. This learning theory explanation is consistent with the female receivers who failed to indicate a value for the sexiness of another female and who were also unaffected by the sexy (communicator) advertisement. The lack of value indicated by female receivers may be a learned response, due either to parental discouragement of sexiness for the female while growing up or it may now be due to perceptions of competitive threats.

Distraction Theory.  The different results between the sexed is expected according to the distraction theory which states that persuasive communications will be more effective when presented with a distraction than without (Festinger and Maccoby, 1964; Rosenblatt, 1966). Festinger and Maccoby's distraction theory (1964) proposes that a receiver generates internal counterarguments when confronted with a persuasive communication (advertisement), but when a distraction (e.g., a sexy female communicator) is also presented the receiver is unable to generate these internal counterarguments. This distraction theory certainly explains the difference between the male and female receiver's responses to the sexy advertisements, i.e., the sexy female communicator advertisement is more effective for male receivers than female receivers because it distracts the males more so than the females.

A number of researchers have opposed distraction theory by suggesting that if persuasive communication is to be effective than learning must first occur (Peterson and Peterson, 1959; McGuire, 1966; Haaland and Venkatesan, 1968). These researchers contend that distraction reduces learning and, consequently, advertisement effectiveness is reduced with distraction. The results of this current study are not supported by the distraction theory's opposition concerning reduced advertisement effectiveness due to distraction; however, the results do support the opposition's hypothesis of reduced learning. For males, the recall of advertisement details (representing learning) was negatively correlated with the sexy condition while the females (who were, apparently, not distracted) showed no such negative correlation.

Consistency Theory.  The results of this study are also explained by consistency theory (Heider, 1958; Festinger, 1957). Consistency theory, applied to this research, suggests that if a receiver holds a positive attitude toward a sexy female, then, other elements of the advertisement with the sexy communicator will be viewed similarly. The results are supportive because males perceived the sexiness of the communicator as positive, and, accordingly, their initial non-positive attitudes changed to be positive toward the elements of advertisement. The female receivers also exhibited consistent attitudes toward the sexy communicator and elements of the advertisement condition.

Dominant Cue Hypothesis.  Finally, the dominant cue hypothesis states that a cue which is immediately available serves a dominant information function (Sigall and Aronson, 1969; Landy and Sigall, 1974). The communicator's sexiness in this experiment was likely the dominant cue from which males received positive information and females received negative information. The results appear to be supported and explained by the theoretical suggestions of this dominant cue hypothesis.

Source Credibility

Source credibility, as defined by expertise and trustworthiness, is crucial for effective persuasive communication (McGuire, 1969; Britt, 1978). Consequently, the results of this current study are relevant to the marketing practitioner. The male receivers saw the sexy female as very trustworthy and high in expertise while the female receivers responded in exactly the opposite direction. This source credibility finding implies that an advertiser desiring to influence the male consumer will be more successful if a sexy female is the communicator than if a non-sexy female is used, but, the opposite effects will result if a sexy female is used in an attempt to influence female consumers.

Future Research

The limited knowledge of the effects of communicator sexiness upon persuasive communication makes it imperative for additional research if the sexiness variable is to be understood. The experimental research design is appropriate to expand the communicator manipulation to include both sexy and non-sexy male communicators and combinations of sexiness levels with multiple communicators together. Also, a range of products and media need to he examined while investigating the characteristics of those receivers influenced and those not influenced by the communicator's sexiness

Several observations during this current experiment suggest additional measures may be appropriate. These observations were: (1) the males appeared more ready to evaluate the communicator than did the female receivers; (2) female receivers asked more questions about the advertisement than did the males; (3) the female receivers often hesitated to evaluate the personal characteristics of the communicator; (4) both males and females were more ready to judge the sexy female than the non-sexy; and (5) the sexy advertisement condition resulted in more post-study questions by both sexes as well as a tendency for the males to go back and re-read the advertisement mock-up after the exercise was completed. Measures to assess and tabulate these subjective responses should be employed. Also, objective measures of time spent reading and looking at the advertisement and time spent completing the response sheet will provide progress toward fully understanding consumer behavior in response to marketing communications. Finally, the practice begun in this study of applying applicable theory must be continued for this sexiness research, as well as all consumer behavior research, if the field is to advance to a meaningful level.


Baker, M. J. and Churchill, G. A. (1977), "The Impact of Physically Attractive Models on Advertising Evaluations," Journal of Marketing Research, 14, 538-55.

Berscheid, E., Dion, K. K., Walster, E., and Walster, G. W. (1971), "Physical Attractiveness and Dating Choice: A Test of the Matching Hypothesis," Journal of Experimental Social Psychology, 7, 173-89.

Britt, S. H. (1978), Psychological Principles of Marketing and Consumer Behavior, Lexington, Massachusetts: Lexington Books.

Bryne, D. (1971), The Attraction Paradigm, New York: Academic Press.

Bryne, D., London, O., and Reeves, K. (1968), "The Effectiveness of Physical Attractiveness, Sex, and Attitude Similarity on Interpersonal Attraction," Journal of Personality, 36, 259-71.

Cohen, M. M. and Saine, T. J. (1977), "The Role of Profanity and Sex Variables in Interpersonal Impression Formation," Journal of Applied Communications Research, 5, 45-51.

Festinger, L. (1957), A Theory of Cognitive Dissonance, Stanford, California: Stanford University Press.

Festinger, L. and Maccoby, N. (1964), "On Resistance to Persuasive Communications," Journal of Abnormal and Social Psychology, 68, 359-66.

Haaland, G. A. and Venkatesan, M. (1968), "Resistance to Persuasive Communication: An Examination of the Distraction Hypothesis," Journal of Personality and Social Psychology, 9, 167-70.

Heider, F. (1958), The Psychology of Interpersonal Relations, New York: Wiley.

Lott, B. E. (1955), "Attitude Formation: The Development of a Color-Preference Response Through Mediated Generalization," Journal of Abnormal and Social Psychology, 50, 321-326.

McGuire, W. J. (1966), "The Current Status of Cognitive Consistency Theory," in S. Feldman, ed., Cognitive Consistency, New York: Academic Press.

McGuire, W. J. (1969), "Nature of Attitudes and Attitude Change," in G. Lindzey and E. Aronson, eds., Handbook of Social Psychology, Reading Massachusetts: Addison-Wesley.

Miller, H. L. and Rivenbark, W. (1970), "Sexual Differences in Physical Attractiveness as a Determinant of Heterosexual Likings," Psychological Reports, 27, 701-2.

Peterson, R. A. and Kerin, R. A. (1977), "The Female Role in Advertisements: Some Experimental Evidence," Journal of Marketing, 41, 59-63.

Peterson, L. R. and Peterson, H. J. (1959), "Short-Term Retention of Individual Verbal Items," Journal of Experimental Psychology, 58, 193-98.

Rosenblatt, P. C. (1966), "Persuasion as a Function of Varying Amounts of Distraction," Psychonomic Science, 5, 85-6.

Sigall, H. and Aronson, E. (1969), "Liking for an Evaluator as a Function of the Physical Attractiveness and Nature of the Evaluations," Journal of Experimental Social Psychology, 5, 93-100.

Tarpy, R. M. and Mayer, R. E. (1978), Foundations of Learning and Memory, Glenview, Illinois, Scott, Foresman, and Co., 17-50.

Wilson, R. D., and Moore, N. D. (1979), "The Role of Sexually-Oriented Stimuli in Advertising: Theory and Literature Review," in W. L. Wilkie, ed., Advances in Consumer Research, 6, Miami: Association for Consumer Research, 55-61.