Familiarity impacts person perception

We investigated the effects of familiarity on person perception. We predicted that familiarity would increase non-analytic processing, reducing attention to and the impact of individuating information, and increasing the impact of category labels on judgments about a target person. In two studies participants read either incriminating or exculpatory individuating information about a defendant in a criminal case and made judgments of guilt. In Study 1, participants were subliminally exposed to the defendant’s photo, another matched photo, or no photo before seeing the evidence. Participants familiar with the defendant’s photo both processed and used the individuating information less. In Study 2, participants were subtly made familiar or not with the incriminating and exculpatory information itself, and the defendant was described either as a priest or as a skinhead. Familiarity with the information reduced attention to its content and also tended to increase reliance on category information in guilt judgments. Copyright # 2006 John Wiley & Sons, Ltd. Imagine that as you browse the morning paper, an article on a local court case catches your eye. A brief article summarizing the arguments made by the prosecution and defense is accompanied by a small photo of the accused. Having never before seen the face before you go ahead and read the information, drawing your own conclusions. Now imagine instead that as you turned to read the facts of the case the face looked vaguely familiar. How might that vague glimmer of familiarity affect any judgment you make about the individual? We have argued that the experience (whether conscious or not) of such familiarity regulates information processing, making analytic processing less likely (Garcia-Marques & Mackie, 2000, 2001). We thus take a dual process view that human information processing involves two distinct modes of computation (see Abelson, 1994; Sloman, 1996; Smith & DeCoster, 2000, for reviews): nonanalytic processing (characterized by the mere activation of a way to deal with a focal stimulus), and stituto Superior de Psicologia Aplicada, Rua Jardim do Tabaco, 34, 1149-041 s, Ltd. Received 24 October 2005 Accepted 27 June 2006 840 Teresa Garcia-Marques and Diane M. Mackie analytic processing (in which the particulars of a situation are carefully and systematically analyzed). When stimulus situations match memory representations, initial bottom-up processing occurs with an ‘‘ease’’ or ‘‘fluency’’ that results in an (implicit) feeling of ‘‘similarity,’’ ‘‘recognition,’’ or ‘‘familiarity’’ (Eich, 1982; Fiske, 1982; Gillund & Shiffrin, 1984; Hintzman, 1988; Humphreys, Bain, & Pike, 1989; Jacoby & Dallas, 1981; Murdock, 1982). This familiarity in turn depresses analytic processing, saving limited analytic processing resources for novel situations rather than situations that have been dealt with before (Johnston & Hawley, 1994). Research findings in several areas are consistent with these arguments. Research on problem solving, for example, has demonstrated that a ‘‘feeling of knowing’’ regulates the process underlying problem solving (Reder & Ritter, 1992; Schunn, Reder, Nhouyvanisvong, Richards, & Stroffolino, 1997). Individuals’ quick judgments about whether they felt they could retrieve an answer to a problem or whether they had to compute it were independent of actually knowing the answer but closely dependent on the familiarity of the situation. Familiar situations gave participants a ‘‘feeling’’ that they ‘‘knew’’ the answer, and thus promoted less effortful, top-down, retrieval strategies. Unfamiliar situations, in contrast, triggered more effortful bottom-up computational strategies. Thus, a ‘‘feeling of knowing’’ caused by familiarity acted as a critical signal to switch the cognitive system between nonanalytic and analytic processing modes. We have also shown that prior exposure to a persuasive appeal decreases the likelihood of its elaboration (Claypool, Mackie, Garcia-Marques, McIntosh, & Udall, 2004; Garcia-Marques & Mackie, 2001). Using a typical dual processing paradigm, we were able to show that novel arguments were systematically processed as evidenced by differential agreement with strong and weak arguments. In contrast, however, people responded to previously encountered messages with less elaboration, eliminating the differential acceptance of strong and weak messages. Of most relevance, we (Smith, Miller, Maitner, Crump, Garcia-Marques, & Mackie, 2006) demonstrated in two experiments that repeated supraliminal exposure to a target person increased the occupational stereotyping of that target. In these experiments, participants were repeatedly exposed to the photos of some targets about whom they subsequently learned an occupational label and some mildly counter-stereotypic information. Participants later rated the targets on several traits, some of which were related to the occupational stereotype. Participants made more stereotypic ratings of targets to whom they had previously been exposed. In the second experiment, drawing attention to the possibility of multiple exposure undermined this effect, implicating the role of familiarity. The results were thus consistent with the idea that familiarity functioned as a regulator of processing mode, such that information about familiar objects received less analytic processing. All these approaches suggest that familiarity with a situation triggers non-analytic rather than analytic processing. In this research we extend our claim to the impression formation or person perception literature adding new insights about how familiarity might be elicited. According to dual process models of person impression, judgments of a target can be based either on analytic individuation, the piecemeal processing and combination of individual pieces of information about the target, or on non-analytic categorical processing, the drawing of inferences about target characteristics based on category membership (Brewer, 1988; Fiske & Neuburg, 1990; Neuberg & Fiske, 1987). Individuation is more effortful than categorybased processing, as evidenced by the fact that participants who engage in piecemeal processing of a target’s attributes spend more time looking at and take longer to read and rate the information (Fiske, Neuberg, Pratto, & Allman, 1986, cited in Fiske & Neuburg, 1990; Fiske, Neuberg, Beattie, & Milberg, 1987; Neuberg & Fiske, 1987) than do other participants. Also consistent with this idea is the finding that individuation is undermined by cognitive capacity constraints (Bodenhausen, 1990). Copyright # 2006 John Wiley & Sons, Ltd. Eur. J. Soc. Psychol. 37, 839–855 (2007)