In 3 experiments on impression formation, participants received categorical (occupation) and individuating information (behavior) about target persons presented either in random order (mixed presentation) or in blocks (blockwise presentation). Presentation mode should be a metainformational cue from which judges infer the purpose of the task. Mixed presentation is a cue for differentiation between categories, and the integration of a category-specific typical case leads to reduced differentiation within categories, assimilation, and, hence, enhanced differentiation between categories. Blocked presentation is a cue for differentiation within categories, and consideration of category boundaries as a frame of reference leads to enhanced differentiation within categories and a concomitant contrast effect. The findings of Experiments 1 and 2 supported this reasoning. Experiment 3 showed, in addition, that an explicit task instruction and the metainformational cue lead to equivalent results. Judgments and impressions about persons can be based on diverse information such as their behavior, traits, group membership, physical appearance, and so forth. These diverse types of information are usually classified into individuating and categorical information. Individuating information refers to a person's behavior or traits, or both, whereas categorical information refers to his or her membership in groups. Judges may use both types of information when forming impressions (for reviews, see Brewer, 1988; Fiske & Neuberg, 1990; Hamilton & Sherman, 1994; Kunda & Thagard, 1996). When they do not have enough diagnostic individuating information or when they do not use this information accurately, the impact of the categorical information is high. Categorical information may influence judgments in a number of ways. An assimilation effect occurs when people are rated as being more similar to their category mean than when assessment is based on individuating information alone (cf. Krueger & Rothbart, 1988; Kunda & ShermanWilliams, 1993; Nelson, Acker, & Manis, 1996; Nelson, Biernat, & Manis, 1990). A contrast effect, on the other hand, occurs when people are rated to be less similar to their category mean than when judged on the basis of individuating information alone (Deaux & Taynor, 1973; Linville & Jones, 1980; Manis, Nelson, & Shedler, 1988; Manis, Paskewitz, & Cotler, 1986; Pratto & Bargh, 1991). These effects have long been treated as distortions that are due to abbreviated and inadequate information processing (the cognitive miser; e.g., Markus & Zajonc,
Read full abstract