In three investigations we examined the evaluative and behavioral reactions of high and low self-monitoring individuals to two advertising strategies: appeals to a product's image and claims about a product's quality. High self-monitoring individuals reacted more favorably to image-oriented advertisements, were willing to pay more for products if they were advertised with an image orientation, and were more willing to try a product if it was marketed with an image appeal. By contrast, low self-monitoring individuals reacted more favorably to product-quality-oriented ads, were willing to pay more for products if they were advertised with a quality orientation, and were more willing to try a product if it was marketed with a quality claim. Theoretical and practical implications of these findings for advertising strategies, as well as theoretical implications of these findings for the nature of attitudes, are explored.

Male undergraduates high and low in self-monitoring listened to either an expert or attractive male source deliver a counterattitudinal message supported by either strong or weak arguments. As expected, high self-monitoring individuals agreed with the expert source regardless of the quality of the arguments presented but agreed with the attractive source only when he delivered strong arguments. By contrast, low self-monitoring individuals agreed with the attractive source regardless of the quality of the arguments presented but agreed with the expert source only when he delivered strong arguments. Cognitive response and recall data suggested that high-self monitoring individuals were systematically processing the attractive source's message and were heuristically processing the expert source's message, whereas low self-monitoring individuals were systematically processing the expert source's message and were heuristically processing the attractive source's message. We discuss the role of source variables in persuasion settings, the determinants of an information-processing strategy, and the functional underpinnings of attitudes.

In Study 1, individuals for whom attitudes serve a primarily social-adjustive function (i.e., highself-monitoring individuals) and individuals for whom attitudes serve a primarily value-expressive function (i.e., low-self-monitoring individuals) were identified. As expected, high-self-monitoring individuals experienced more attitude change after exposure to a message said to address a socialadjustive function, and low-self-monitoring individuals experienced more attitude change after listening to a message presumably directed at a value-expressive function. Moreover, subjects tended to generate proportionally more message-relevant thoughts in response to, and tended to recall better, functionally relevant messages. In addition, recall tended to be particularly consistent with post message attitudes when the message was functionally relevant. A second study suggested that the attitude change obtained in Study 1 occurred via peripheral route processes. Results are discussed in terms of the usefulness of adopting a functional approach and its implications for persuasion, information processing, and memory for attitudinally relevant information. Additionally, methods for studying attitudinal functions are discussed. This article is based on a doctoral dissertation submitted to the graduate school of the University of Minnesota.I would like to thank Eugene Borgida. Mark Snyder, and three anonymous reviewers for their comments on an earlier version of this article, and Shirley DeBono and Allen M. Omoto for their assistance in the design of this study. Many thanks also go to Candace and Tamara Polin and Rick Hamish for serving as experimenters and for coding data.

To examine the relation between degree of involvement in a task and the complexity of strategy a subject applies to the task, we randomly assigned 48 female university volunteers to either a dating condition (high-involvement) or one of two (low-involvement) control conditions. These subjects performed a covariation judgment task for which the likelihood of their using simple or complex strategies was calculated. High-involvement subjects used more complex strategies and tended to be more accurate. These data are discussed in terms of the functionality of human information processing, heuristic analyses of inference strategies, and the importance of considering level of personal involvement in analyses of task performance.

In Study 1, high and low self-monitors either listened to music on a cassette tape or sampded a cola product after observing either an image- or a quality-oriented advertisement for the product. High self-monitors rated the quality of the products higher than low self-monitors after suing the image-oriented advertisements, and low self-monitors rated the quality of the products higher than high self-monitors after seeing the quality-oriented advertisements. In Study 2, high and low self-monitors -sorted ads that varied in the degree to which they represented image- or quality-based appeals into self-relevant and non-self-relevant categories. The more image oriented the ads were, the more self-relevant high self-monitors perceived the ads to be, and the more quality oriented the ads were, the more self-relevant low self-monitors perceived them to be. In Study 3, 1 week after viewing image-and quality-oriented ads, high and low self-monitors saw additional image and quality ads (some new, some originals) and were asked whether the ads were old or new. High self-monitors were more accurate at identifying image ads as old or new, whereas low self-monitors were more accurate at identifying quality ads as old or new.