On the Processing of Functionally-Relevant Consumer Information: Another Look At Source Factors

Kenneth G. DeBono, Union College
ABSTRACT - Two studies are reported which suggest that attitude functions may interact with a message's source to influence processing strategy. In particular, individuals for whom attitudes serve a social-adjustive function appear to systematically process the message of a socially or physically attractive source whereas individuals for whom attitudes serve a value-expressive function appear particularly likely to systematically process the message of an expert source. Implications for advertising and consumer behavior are discussed.
[ to cite ]:
Kenneth G. DeBono (1989) ,"On the Processing of Functionally-Relevant Consumer Information: Another Look At Source Factors", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 312-317.

Advances in Consumer Research Volume 16, 1989      Pages 312-317

ON THE PROCESSING OF FUNCTIONALLY-RELEVANT CONSUMER INFORMATION: ANOTHER LOOK AT SOURCE FACTORS

Kenneth G. DeBono, Union College

ABSTRACT -

Two studies are reported which suggest that attitude functions may interact with a message's source to influence processing strategy. In particular, individuals for whom attitudes serve a social-adjustive function appear to systematically process the message of a socially or physically attractive source whereas individuals for whom attitudes serve a value-expressive function appear particularly likely to systematically process the message of an expert source. Implications for advertising and consumer behavior are discussed.

The decade of the 1980's has witnessed a renewal of interest in persuasion, social influence, and persuasion-related phenomena. One consequence has been a rekindling of interest in the role that source -factors play in the persuasion process. Traditionally, source factors have been thought to play an important role in determining the effectiveness of a persuasive communication, although their specific role seems to shift as a function of the predominant theoretical orientation of the times. For example, earlier, more stimulus-response models (e.g., Hovland, Janis, & Kelley, 1953) portrayed source factors as serving as incentives for attitude change, whereas more recent, cognitive, models of persuasion (e.g., Chaiken, 1987; Petty & Cacioppo, 1986a, 1986b) have suggested that source factors play a multifaceted role.

Specifically, cognitive models suggest that under conditions of low personal involvement, source factors have a direct impact on the effectiveness of a persuasive message. In the interest of cognitive economy, individuals tend to more heuristically process the message, relying on source factors and not on the content of the persuasive message to decide whether the attitude position advocated is valid (Chaiken, 1987; Petty & Cacioppo, 1986a, 1986b). Thus, for example, under conditions of low personal involvement, individuals are more likely to agree with an attractive source than a non-attractive one regardless of the quality of the arguments offered (Chaiken, 1980).

By contrast, when an issue is more personally involving, source factors are believed to have little, if any, direct impact on the persuasiveness of a message. Individuals tend to more systematically process the message, often basing agreement on the cogency and strength of the arguments used (Chaiken 1987; Petty & Cacioppo, 1986a, 1986b). Thus, under these conditions, individuals are more likely to agree with a source who is armed with convincing arguments than one who has more specious arguments regardless of the source's physical attractiveness or expertise (e.g., Petty, Cacioppo, & Goldman, 1981).

Lastly their are times when the source factors themselves affect people's motivation to engage in effortful processing. In particular, under conditions of moderate involvement, individuals are more likely to process systematically the message of an attractive or an expert source than the message of an unattractive or non-expert source (e.g., Puckett, Petty, Cacioppo, & Fisher, 1983).

Importantly, the re-emerging functional perspective on attitudes {Katz, 1960; Smith, Bruner, & White, 1956) has the potential to further our understanding of the role that source factors play in the persuasion process. In particular, Kelman (1961) has suggested that the source factors of attractiveness and expertise can, under specified conditions, make a persuasion situation more personally involving than it otherwise might be. He suggests that for individuals who define themselves through their relationships with others, who are concerned that they play roles appropriate for their circumstances, and who are concerned that they maintain satisfying relationships with others, a persuasion situation involving an attractive source may be particularly motivating and involving. Similarly, he proposed that when individuals are concerned that their actions and beliefs are congruent with their important values, a persuasion situation involving an expert source becomes particularly motivating and personally involving.

This characterization of the influence process suggests that when the interpersonal needs of an individual can be satisfied by a message's source, source factors, because of the increased involvement they engender, may motivate a recipient to process more systematically a persuasive message. That is, if source factors do indeed make persuasion situations differentially involving for different people, then individuals should be differentially motivated to analyze and scrutinize the content of the persuasive message as a function of the message's source.

To test this hypothesis however, it is first necessary to identify individuals for whom the source factors of expertise and attractiveness should be differentially involving. We believe the differing functional bases of the attitudes of high and low self-monitoring individuals make them prime candidates in this search. High self-monitoring individuals (as identified by their relatively high scores on the Self-Monitoring scale, Snyder, 1974; Snyder & Gangestad, 1986) typically strive to be the type of person called for in each situation in which they find themselves. They are concerned about, and are adept at, tailoring their behavior to fit social and interpersonal considerations of appropriateness. Importantly, the attitudes of high self-monitoring individuals appear to serve a social adjustive function. High self-monitoring individuals are concerned with what important reference groups believe and will readily change their attitudes if they discover that their attitudes are at odds with a significant majority of their peers (DeBono, 1987). Thus, one could argue that the social-adjustive nature of their attitudes may lead high self-monitoring individuals to find a situation involving an attractive source particularly involving, which may in turn lead them to more systematically process what the source has to say.

In contrast, low self-monitoring individuals (as identified by their relatively low scores on the Self-Monitoring scale) typically do not attempt to mold their behavior to fit situational and interpersonal considerations of appropriateness. Rather, these individuals tend to guide their behavioral choices on the basis of relevant inner sources such as values, feelings, and dispositions (cf, Snyder, 1987). Moreover, the attitudes of low self-monitoring individuals may play a value-expressive function. Low self-monitoring individuals are concerned that their attitudes reflect accurately their important values and they will readily change their attitudes if informed that their present attitudes are not congruent with important values (DeBono, 1987). This value-expressive orientation may lead low self-monitoring individuals to find a situation in which an expert source is delivering a message to be particularly involving, which may lead them to more systematically process the source's message. - Do the differing functional orientations of high and low self-monitoring people lead them to process the messages of expert and attractive sources differently? In an attempt to answer this question, DeBono & Harnish (in press) conducted a partial replication of Kelman's (1961) classic experiment on the three processes of attitude change. In particular, we had one-hundred male Michigan State University undergraduates listen to a counter-attitudinal message delivered by either an expert or socially attractive source. The attitude topic revolved around a current controversy on the campus of Michigan State University. The women's pom-pom squad, in a fund raising effort, produced a calendar in which members, rather revealingly dressed (e.g. in bikinis), posed with various members of the campus community (e.g., the University president). Most students did not question the propriety of the calendar (although a small, but vocal, minority did), so the message students heard delineated reasons why the calendar should be banned from campus.

Those subjects randomly assigned to the expert condition believed the message came from a nationally known, well-published research psychologist who specializes in the effects of the print media on attitudes and belie&. Those assigned to the attractive source condition believed that the message came from an honors student at MSU who is very active in student government and is presently the chair of the MSU chapter of the Student Poll. Subjects in this condition were informed that although the student would be speaking in the first person, the views he expressed were based on the consensus opinion of MSU students as assessed by a recent student poll.

Within each source condition, participants were randomly assigned to hear one of two versions of the message. In the strong argument condition, the speaker said that the calendar promoted sexist attitudes toward women, further reinforced the stereotype of women as sex objects, and tarnished the progressive and enlightened image of the University. In the weak argument condition, the speaker said that the calendar gave free advertising to the swimsuit manufacturers, would lead parents to wonder if their investment in their daughter's education was worthwhile, and would draw too much attention away from athletes at sporting events as attention would now be focused on the pom-pom squad.

After hearing the speaker, we asked students for their personal attitudes toward the calendar (1 = worthless, 7 = valuable), we asked them to list all the thoughts they had while listening to the message and to code them in terms of the favorableness toward the speaker's arguments, and, after a filler task, we asked them to recall as many of the speaker's arguments as possible.

The results for post-message attitudes are displayed in Table 1.

A self-monitoring x source x argument strength ANOVA revealed a significant three-way interaction, F(1, 92) = 32.72, p < .001. As can be seen, high self-monitoring individuals were responsive to the quality of the message arguments only when the speaker was socially attractive, forming negative attitudes toward the calendar only after hearing the strong arguments. In contrast, high self-monitoring individuals formed relatively negative attitudes toward the calendar when the source was an expert, regardless of argument quality. These data suggest that high self-monitoring individuals were systematically processing the message of the attractive source, but were more heuristically processing the message of the expert source.

This pattern of data is reversed for low self monitoring individuals. They appeared to be sensitive to argument quality only when the source was an expert; they agreed with the attractive source regardless of argument quality. Thus, they appeared to systematically process the message of the expert source, but seemed to more heuristically process the message of the expert source.

This interpretation is buttressed by an examination of the cognitive response and recall data. To examine the nature of the thoughts people had while listening to the message, we calculated the proportion of favorable thoughts relative to the total number of favorable and unfavorable thoughts. These are displayed in Table 2.

A self-monitoring x source x argument strength ANOVA revealed a significant three-way interaction, F(1, 92) = 6.33, p < .02. As can be seen, the thoughts of high self-monitoring individuals appeared to be a function of argument quality only when the source was an attractive one, further suggesting that they were systematically processing that source's message. By contrast, the thoughts of low self-monitoring individuals appeared to be responsive to argument quality only when the source was described as an expert, thus suggesting that they were only systematically processing his message.

Lastly, an examination of the recall data revealed self-monitoring x source interaction. High self-monitoring individuals were able to recall more when the source was described as socially attractive than when describe as expert (M = 3.58 vs. M =3,27), low self-monitoring individuals were able to recall more when the source was described as expert than when described as socially attractive (M = 3.29 vs. M = 2.42).

TABLE 1

MEAN POST-MESSAGE ATTITUDE SCORES

TABLE 2

MEAN PROPORTIONS OF FAVORABLE THOUGHTS

Taken together, these data suggest that different sources can indeed make persuasion situations differentially involving for different people and a result of this differential involvement is a differential processing of the persuasive information. When a source can fulfill the interpersonal needs of an individual, it appears as though that individual becomes willing to take the time and energy to process systematically what the source is saying. As such, the individual becomes responsive to argument quality and is only persuaded by strong arguments. By contrast, when the source can not fulfill the interpersonal needs of an individual, that individual, not caring to exert much cognitive energy, more heuristically processes the message, using the source as a persuasion cue and agreeing with the source regardless of argument quality.

Before discussing the implications of these findings for advertising and consumer behavior, it is important to address the generality of these findings. When we speak of source attractiveness, we can mean one of two things. We can mean social attractiveness, as in the study above, or we can mean physical attractiveness. Given that the physical attractiveness of others and of objects seems to be important for high self-monitoring individuals (DeBono & Snyder, in press), DeBono & Telesca (1988) set out to discover if a physically attractive source has the same information-processing effects on high and low self-monitoring individuals as did the socially attractive source.

TABLE 3

MEAN POST-ADVERTISEMENT ATTITUDE SCORES

In this study, we informed sixty-nine Union College undergraduates that a new sun tan lotion, "Savage Tan", was being -introduced to New York's Capital District and that they would be asked to preview and evaluate one of the advertisements for the product. The advertisement contained two parts. The first, the visual, was a slide of a darkly tanned, physically attractive woman described to participants as a local model. She was pictured standing in front of a painting of a sunset holding a container of "Savage Tan". The second part of the advertisement was a voice-over played for participants on a cassette recorder. For some of the participants the voice over contained rather strong arguments for using the product. They were told that the lotion was scientifically designed and proven to protect the user from harmful ultraviolet rays and contained special ingredients to help prevent premature aging and wrinkling of the skin. For the remaining participants the voice over contained weak, rather specious arguments in favor of the product. Participants were told that the lotion was silky smooth and easy to apply and it came in a newly-designed container that had an easy-to-grasp handle.

Following exposure to the advertisement, we asked participants to evaluate the product (on two seven point scales: 1 = weak, 7 = strong, and 1 = worthless, 7 = valuable), write down all the thoughts that they had about the product and its presentation while they were observing the advertisement, code their thoughts in terms of their favorableness toward the product, and finally, recall three things about the presentation: who promoted the product, and the two arguments that the spokesperson used.

Post-message attitude toward the product are displayed Table 3. A self-monitoring x argument strength ANOVA revealed a two-way interaction, F(1, 65) = 10.43, p < .01, such that the attitudes of high self-monitoring individuals were affected by argument quality, becoming favorable toward the product only after hearing strong arguments, whereas the attitudes of low self-monitoring individuals became favorable regardless of argument quality. This suggests that high self-monitoring individuals may have been systematically processing the model's message whereas as low self-monitoring may have been more heuristically processing the message.

A significant self-monitoring x argument strength interaction for the proportion of favorable cognitive responses (See Table 4) tells a similar story. Only the thoughts of high self-monitoring individuals were affected by argument quality: they tended to have a higher proportion of favorable thoughts in response to the strong arguments than to the weak whereas there was no difference in the proportion of favorable thoughts by low self-monitoring individuals as a function of argument quality. Moreover, high self-monitoring individuals were able to recall significantly more of the message than were low self-monitoring individuals, (M = 2.24 vs. M = 1.68, p < .05).

What are the implications of these studies for advertising and consumer behavior? One general answer would be "tentative". We are fully aware that the methods we have chosen to conduct our research carry with them the baggage of all laboratory research: the limited scope of stimuli used and the inherent problems of external validity. Nonetheless, we believe that our research has the potential to contribute to an understanding of "real world" persuasion attempts. For example, an ostensibly interesting implication of these results is that, to make maximum use of source factors, it may be more effective to present a source that is not 'functionally relevant' to the individual. Overall, low self-monitoring individuals were more persuaded by the attractive sources than were high self-monitoring individuals and high self-monitoring individuals were more persuaded by expert sources than were low self-monitoring individuals. For example, it mattered little to low self-monitoring individuals what attractive sources said whereas high self-monitoring individuals were only persuaded when the source had strong arguments.

TABLE 4

MEAN PORPORTIONS OF FAVORABLE THOUGHTS

The soundness of this strategy, however, may be a function of the goals of the persuasion attempt. That the attractive and expert sources elicited differing processing strategies in high and low self-monitoring individuals may also have implications for the duration of the attitude change engendered and the likelihood that the attitude change will translate into behavior change. In particular, research has suggested that attitude change induced via systematic, or central route, processes tends to be of a longer duration than attitude change resulting from more peripheral route processes and is more likely to be related to behavior (cf. Petty dc Cacioppo, 1986b). Therefore, although, individuals were more persuaded by non-functionally relevant sources, this attitude change was most likely short-lived and not behaviorally-related. Hence, a strategy of using a non-functionally relevant source may be ultimately effective if one only desires relatively short-term attitude change. By contrast, to the extent that one is interested in more long-term, behaviorally-related attitude change, a functionally relevant source may be more effective, given that the source has convincing arguments.

REFERENCES

Chaiken S. (1980). Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology,39, 752-766.

Chaiken, S. (1987). The heuristic model of persuasion. In M. P. Zanna, J. M. Olson, & C. P. Herman (Eds.), Social influence: The Ontario symposium (Vol. 5, pp. 3-39). Hillsdale, NJ: Erlbaum.

DeBono, K. G. (1987). Investigating the social adjustive and vale expressive functions of attitudes: Implications for persuasion processes. Journal of Personality and Social Psychology, 52, 279-287.

DeBono, K. G., & Harnish, R. J. (in press). Source expertise, source attractiveness, and the processing of persuasive information: A functional approach. Journal of Personality and Social Psychology.

DeBono, K. G., & Snyder, M. (in press). Understanding consumer decision-making processes: The role of form and function in product evaluation. Journal of Applied Social Psychology.

DeBono, K. G., & Telesca, C. (1988). The role of source physical attractiveness in the persuasion process: A functional perspective. Unpublished manuscript, Union College, Schenectady, N. Y.

Katz, D. (1960). The functional approach to the study of attitudes. Public Opinion Quarterly, 24, 163204.

Kelman, H. C. (1961). Processes of opinion change. Public Opinion Quarterly, 25, 57-78.

Petty, R. E., & Cacioppo, J. T. (1986a). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 19, pp. 123-205). New York: Academic Press.

Petty, R. E., & Cacioppo, J. T. (1986b). Communication and persuasion: Central and peripheral routes to attitude change. New York: Springer-Verlag.

Petty. R. E., Cacioppo, J. T., & Goldman, R. (1981). Personal involvement as a determinant of argument-based persuasion. Journal of Personality and Social Psychology, 41, 847-855.

Puckett, J., Petty, R. E., Cacioppo, J. T., & Fisher, D. (1983). The relative impact of age and attractiveness stereotypes on persuasion. Journal of Gerontology, 38, 340-343.

Smith, M. B., Bruner, J. S., & White, R. W. (1956). Opinions and personality. New York: Wiley and Sons.

Snyder, M. (1974). The self-monitoring of expressive behavior. Journal of Personality and Social Psychology, 30, 526-537.

Snyder, M. (1986). Public appearances. private realities: The psychology of self-monitoring. New York: Freeman.

Snyder, M., & Gangestad, S. (1986). On the nature of self-monitoring: Matters of validity, matters of assessment. Journal of Personality and Social Psychology, 51, 125-139.

----------------------------------------