The three target articles composing this symposium argue persuasively that social inference is influenced by people's goals and needs. Dunning (this issue) describes how the desire maintain a positive self-image leads the development and maintenance of positive self-schemata. According Kunda and Sinclair (this issue), stereotypes regarding a target are activated, applied, or inhibited depending on the target's behavior has positive, negative, or neutral implications for the self. In the third target article, Murray (this issue) shows how people construe their romantic partners in ways that sustain the perception that one's partner has positive attributes and the relationship will continue. In an earlier article, Kunda (1990) classified motivated cognition into two major categories: those in which the motive is to arrive at an accurate whatever it may be, and those in which the motive is arrive at a particular conclusion (p. 480). The articles in this symposium focus on the second category. Regardless of which category of motivated cognition we consider, a common element is the nature of the question posed in selfand other-evaluation. In Dunning's research, the implicit or explicit question is Do I possess positive attributes? For Kunda and Sinclair's stereotyping research, the question is the evaluator who praises or derogates the self is credible or uncredible. The relationship studies described by Murray focus on whether my romantic partner has positive attributes that will help maintain our In each case, a goal frames the question such that information is retrieved from memory or beliefs are constructed that are consistent with the goal. A hot question (using Abelson's hot-cold cognition dichotomy) steers a cold, cognitive process. This follows from reasonably well-understood processes used in lay hypothesis testing: People rely on a positive test strategy whereby they seek out instances in which the hypothesized property is known or expected be present rather than absent (Klayman & Ha, 1987; see also Kunda, 1990). For example, the question, Am I a capable leader? prompts a biased memory search for instances attesting my leadership skills and not those situations when my leadership capabilities were lacking. After receiving negative feedback, Kunda and Sinclair's research participants should want derogate the source. In essence, they are implicitly asking, What kind of person gave me such negative feedback? Consequently, they recruit relevant beliefs that would discredit the feedback. In contrast, if the feedback that participants received was positive, then they frame the question differently because now the self is bolstered if the source is credible. Murray's participants want be part of a committed and sustaining relationship. The question here is Is my partner loving and committed?, which should encourage the retrieval of positive memories and beliefs about the partner. In each of these cases, the evaluative question is framed in a way that increases the retrieval of hypothesis-consistent evidence. The nature of the evaluative question is also important in understanding how comparisons with other people influence feelings and behavior (Smith, 1981). In contemporary social comparison theory (Suls & Wills, 1991), the two categories of motivated cognition are evident. Self-evaluation involves people making social comparison for the purpose of obtaining accurate assessments of their abilities and opinions. According Festinger (1954), self-evaluation is best achieved by comparing with similar others, although he was ambiguous about what constituted similarity. Over the last 40 years, the similarity hypothesis has received mixed support although a reformulation of the theory (Goethals & Darley, 1977) has been more successful. A second motive, self-enhancement, prompts the selection of comparisons that increase or maintain positive affect, particularly if the self has been threatened. Here, a directional conclusion, borrow Kunda's (1990) phrase, is preferred. The self-enhancement approach social comparison has several variants, but the most influential was proposed by Wills (1981), who argued that people under threat compare with other persons who are worse off feel better. Indeed, in several studies with populations under threat, such as medical patients, participants reported downward comparisons that appeared be used as a means cope with their predicament (e.g., Wood, Taylor, & Lichtman, 1985). There is considerable evidence of social comparisons made in accord with the self-evaluation and self-enhancement motives. However, the empirical literature shows much inconsistency in selection and impact of comparison information (Kruglanski & Mayseless, 1990; Suls & Miller, 1977). Similar others are not always preferred assess one's level of ability. In the case of self-enhancement, there are several em-
Read full abstract