August 1978. Experiment 2 ba sed cognitions in mediating opinion change, was partially funded by a grant from the Humanities whereas a heuristic view de-emphasizes deand Social Sciences Committee of the Research tailed information processing and focuses on Board of the University of Toronto. the role of simple rules Qr cognitive heuristics Requests for reprints should be sent to Shelly jn mediati persuasion.

High- and low-task-importance Ss read a strong or weak unambiguous message or an ambiguous message that was attributed to a high- or low-credibility source. Under low task importance, heuristic processing of the credibility cue was the sole determinant of Ss' attitudes, regardless of argument ambiguity or strength. When task importance was high and message content was unambiguous, systematic processing alone determined attitudes when this content contradicted the validity of the credibility heuristic; when message content did not contradict this heuristic, systematic and heuristic processing determined attitudes independently. Finally, when task importance was high and message content was ambiguous, heuristic and systematic processing again both influenced attitudes. Yet, source credibility affected persuasion partly through its impact on the valence of systematic processing, confirming that heuristic processing can bias systematic processing when evidence is ambiguous. Implications for persuasion and other social judgment phenomena are discussed.

and Kardes (1986) demonstrated that Ss were able to evaluate adjectives more quickly when these adjectives were immediately preceded (primed) by attitude objects of similar valence, compared with when these adjectives were primed by attitude objects of opposite valence. Moreover, this effect obtained primarily for attitude objects toward which Ss were presumed to hold highly accessible attitudes, as indexed by evaluation latency. The present research explored the generality of these findings across attitude objects and across procedural variations. The results of 3 experiments indicated that the automatic activation effect is a pervasive and relatively unconditional phenomenon. It appears that most evaluations stored in memory, for social and nonsocial objects alike, become active automatically on the mere presence or mention of the object in the environment. The research we report in this article concerns the extent to which attitudes may be activated automatically on mere observation of the attitude objects to which they correspond. This phenomenon is central to Fazio's (1986, 1989, 1990) attitude accessibility model, a theory of the process by which attitudes guide behavior. Consistent with the attitude accessibility model, we demonstrate that attitudes are capable of automatic Editor's Note. Two of the authors are Associate Editors of this journal; therefore, this article was subjected to a special review process to avoid any potential conflict of interest. I asked Anthony Greenwald, a former Editor of JPSF to handle the article. He solicited the reviews and made the recommendations that the article be published. I am grateful to him for taking on this chore.

In this study we tested the framing hypothesis that a pamphlet stressing the negative consequences of not performing breast self-examination (BSE) would be more persuasive than a pamphlet emphasizing BSE's positive consequences. College-aged female subjects were exposed to a loss-frame pamphlet, a gain-frame pamphlet, or a no-arguments pamphlet, or they received no pamphlet describing the importance of and the techniques for performing BSE. Attitudes toward BSE and intentions to perform BSE were assessed immediately after this intervention and again 4 months later. The followup also assessed subjects' postexperi mental BSE behavior. Consistent with predictions, subjects who read a pamphlet with arguments framed in loss language manifested more positive BSE attitudes, intentions, and behaviors than did subjects in the other three conditions. The greater impact of the loss pamphlet could not be attributed to greater fear arousal, better memory for pamphlet content, greater perceived susceptibility to breast cancer, or stronger beliefs in BSE's efficacy on the part of the loss subjects. Only measures of perceived self-efficacy in performing BSE were differentially affected by the framing manipulation, with loss subjects reporting the greatest levels of self-confidence. The results are discussed in terms of prospect theory's framing postulate and a simpler negativity-bias conceptualization, and underlying mechanisms such as differential salience and vividness are considered. Clinical implications of the findings are also explored. Failure to comply with recommended health-care behaviors is a major contributor to death and disability in this country (e.g., Belloc, 1973; Stachnik, 1980). Research indicates that nonadherence rates are often extremely high, particularly for discretionary preventive and diagnostic behaviors such as quitting smoking, getting exercise, and performing monthly breast self-examination (BSE; see Ley, 1982; Masur, 1981). In the present study, we investigated an intervention designed to increase the performance of BSE among college-aged women. This health behavior was chosen for a number of reasons, including the high prevalence of breast cancer among American women (American Cancer Society, 1983), the relatively high survival rates associated with this disease when it is diagnosed at an early stage (American Cancer Society, 1983), and the potential effectiveness of monthly BSE as an aid in the early detection of malignant breast lumps (e.g., Foster & Costanza, 1984).'

Ss received consensus information that was either congruent or incongruent with the valence of persuasive message content. In Experiment 1 Ss believed that their processing task was either important or unimportant whereas in Experiment 2 all Ss believed that their task was unimportant. In accord with the heuristic-systematic model's sufficiency principle, high-task-importance Ss exhibited a great deal of systematic processing regardless of congruency, whereas low-importance Ss processed systematically only when they received incongruent messages; in the congruent conditions heuristic processing dominated. Attitude data generally reflected these processing differences and confirmed the additivity and attenuation assumptions of the model. The utility of the sufficiency principle for understanding motivation for elaborative processing and the relevance of the findings to understanding the processing and judgmental effects of expectancy disconfirmation are discussed.

No abstract

Subectsfor whom a health threat was relvant or irrelevant were recruited and matched on prior beliefs in the health threat. Following exposure to either a low- or a high-threat message, high-relvance subjects were less likely to believe in the threat. Consistent with earlier work, no evidence was found to suggest that defensive inattention to the messages mediated subjects' final beliefs. Instead, processing measures suggested that highrelevance subects processed threatening parts of both messages in a biased fashion. The relationship between biased judgment and biased processing is discussed, as are the difficulties in documenting the latter

No abstract