Contrast effects in stereotype formation and change: the role of comparative context

Two experiments investigated the way in which the presence of a comparative or inter-group context during stereotype formation affects stereotype change, induced by subsequent disconfirming information. Participants learned about a focal group, after learning about one of the two context groups. After reporting their stereotypes about both groups, participants learned additional information about the focal group. This information described new group members who either confirmed or disconfirmed the group stereotype. Consistent with previous research, participants formed more extreme stereotypes about the focal group on dimensions that distinguished it from the context group (i.e., a contrast effect). In response to the subsequently presented disconfirming group members, a greater stereotype change was observed on dimensions that distinguished the focal group from the context group than on dimensions it did not. We argue that these effects are due to differences in perceived typicality of disconfirming group members. 2002 Elsevier Science (USA). All rights reserved. Attempts by social psychologists to identify ways in which group stereotypes can be effectively changed have met with mixed success. Most of these attempts have explored the conditions under which stereotype-inconsistent information (in the form of behaviors or traits shown by groupmembers) reduces stereotypic views of the group (e.g., Johnston & Hewstone, 1992; Kunda & Oleson, 1995; Rothbart, 1996; Rothbart & John, 1985; Weber & Crocker, 1983). The focus in this literature has largely been on characteristics of the stereotype-inconsistent information that facilitate or hinder stereotype change. To gain a better understanding of stereotype change, we argue in this paper that it is important to better understand the factors that affect stereotype formation and content in the first place. That is, we argue that change in response to stereotype-inconsistent information may importantly depend on the nature and content of the stereotype that one is attempting to change. Group stereotypes have been shown to be contextdependent. Beliefs about a particular group depend heavily on the comparative context in which those beliefs are salient. Our central thesis is that stereotype change in response to stereotype-inconsistent information also depends in predictable ways upon this comparative context. To develop the rationale for this hypothesis, we begin by reviewing what is known about the role of context in stereotype formation. In light of these effects, we then outline some ways in which these context effects may influence responses to stereotype-inconsistent information. Finally, we present the results of two studies designed to test these ideas. Context effects on group stereotypes The presence of a salient comparative context during stereotype formation has been shown to have two different, albeit related, effects. First, learning about a social category in the presence of a comparative category leads to the accentuation of inter-group differences. This effect is a robust one that numerous studies have shown. Second, learning abut a social category in the presence of a comparative category leads to the accentuation of Journal of Experimental Social Psychology 38 (2002) 443–458 www.academicpress.com Journal of Experimental Social Psychology Corresponding author. E-mail address: [email protected] (N.A. Wyer). 0022-1031/02/$ see front matter 2002 Elsevier Science (USA). All rights reserved. PII: S0022 -1031 (02 )00010-0 intra-group homogeneity. As we discuss below, the empirical evidence in support of this second effect is considerably more mixed. The combination of these two effects has been used to define what theorists in the Social Identity tradition refer to as the meta-contrast ratio (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987): ‘‘People use characteristics that differentiate between groups as a crucial piece of information in categorization—people are considered to be members of the same group to the extent that they are different from other groups, but are similar to each other, on some relevant dimensions.’’ These theorists have argued that the presence of a comparative context group leads to more polarized stereotypes (Oakes, Haslam, Turner, Worchel, & Morales, 1998). This hypothesis has been empirically supported in work by Oakes, Haslam and co-workers (Doosje, Haslam, Spears, Oakes, & Koomen, 1998; Haslam & Turner, 1995; Oakes, Haslam, Reynolds, Abrams, & Hogg, 1999). How do these contrast effects come about? Thinking about a focal group in the presence of another comparative focal group seems to affect inter-group differentiation at a number of different levels. First, at a relatively early stage, more attention is paid to information that differentiates the target and context groups from each other (Trope & Mackie, 1987). This differential attention means that group stereotypes are more likely to consist of attributes that differentiate a group from its comparative context (see Judd & Park, 1993; McCauley & Stitt, 1978). Ford and Stangor (1992) reported that when participants learned about two novel groups that differed on one trait dimension but not on another, they were more likely to spontaneously mention the differentiating dimension than the non-differentiating one when later describing the groups. In addition, Babey (1999) showed that participants drew trait inferences about a group more quickly and based on less information if they learned about that group in the context of a second one than if they did not have that context. As a result of greater attention to differentiating dimensions, the presence of a context may lead to more extreme or stereotypic judgments of a group on those dimensions. This accentuation of stereotypic differences has been shown both with purely perceptual categories (e.g., Corneille & Judd, 1999; Tajfel & Wilkes, 1963) and with more complex social categories (e.g., Eiser, 1971; Ford & Tonander, 1992; Krueger & Rothbart, 1990; Schell, 1997). As we suggested above, the evidence for accentuation of intra-group homogeneity in the presence of a comparative context is considerably more mixed. For example, the classic study of Tajfel and Wilkes (1963) found no evidence that categorization increased intracategory similarity. More recently, Krueger and Rothbart (1988) also failed to find such an evidence. On the other hand, McGarty and Penny (1988) showed that opinion statements within a category were judged as more similar to each other when a categorization was made salient to participants. Rothbart, Davis-Stitt, and Hill (1997) reported similar results in research that involved more socially relevant stimuli. Somewhat surprisingly, recent work by Livingston, Andrews, and Harnad (1998) reported evidence for intra-group accentuation of homogeneity in the absence of inter-group differentiation. Context and stereotype change The most typical approach to changing people’s stereotypes about a social group is to expose them to group members who disconfirm the stereotype. This has been done either by exposing perceivers to actual group members in inter-group contact settings (e.g., Brown, Vivian, & Hewstone, 1999; Wilder, 1993) or to written descriptions of hypothetical group members (e.g., Johnston & Hewstone, 1992; Kunda & Oleson, 1995; Weber & Crocker, 1983). Although contact with members of derogated outgroups can sometimes lead to more positive attitudes towards those groups (Pettigrew & Tropp, 2000), change in the group stereotype seems to be a much more elusive result. For disconfirming information to lead to a stereotype change, a number of important conditions seem to be necessary. Most importantly for the present argument, the disconfirming group member must actually be seen as a relevant group member. They must be typical of the group as a whole in most ways, even while disconfirming the group stereotype to some extent (Rothbart & John, 1985). Otherwise, the disconfirming member is likely to be ‘‘fenced off’’ (Allport, 1954) or subtyped (Johnston & Hewstone, 1992; Maurer, Park, & Rothbart, 1995; Rothbart & John, 1985; Weber & Crocker, 1983). A subtyped individual is one who is not considered a ‘‘real’’ group member and consequently tends to be excluded from consideration when reporting the group stereotype. Thus, stereotype change in response to a disconfirming group member is unlikely to occur if the individual is subtyped. What factors influence whether or not a disconfirming group member is subtyped? There is ample evidence to show that an important factor is the degree to which they disconfirm the group stereotype. For instance, Weber and Crocker (1983) found that participants who learned about group members who were partially atypical of the group stereotype changed their beliefs about the group. On the other hand, participants who learned about group members who were entirely atypical of the group did not change their stereotypes at all. Thus, the more extreme the stereotype disconfirmation, the less 444 N.A. Wyer et al. / Journal of Experimental Social Psychology 38 (2002) 443–458