When Do Stereotypes Come to Mind? (2003) PDF

Psychological Bulletin 2003, Vol. 129, No. 4, 522–544 When Do Stereotypes Come to Mind and When Do They Color A Goal-Based Theoretical Framework for Stereotype Activation Application The authors describe a of a stereotyped group stereotype activation hension and self-enhancement the strength of one’s inhibit stereotype activation influence the intensity application. This approach application. As one interacts with a member of a stereotyped group—for example, an African American, a woman, a doctor—the stereotype of that group may or may not come to mind. When it does, it may or may not color one’s impression of that person. These two processes have been labeled stereotype activation, that is, the extent to which a stereotype is accessible in one’s mind, and stereotype application, that is, the extent to which one uses a stereotype to judge a member of the stereotyped group. We ex- amine what determines the activation and the application of ste- reotypes and how these two processes relate to one another. We propose that both the activation and the application of stereotypes depend on three kinds of goals: comprehension goals, self- enhancement goals, and the motivation to avoid prejudice.1 We explain how these goals can promote and inhibit the activation and the application of stereotypes, and describe situational and chronic factors that can trigger these goals and enable their execution. Conceptualizing and Assessing Stereotype Activation and Stereotype Application There has always been a clear theoretical distinction between the accessibility of knowledge structures such as stereotypes and their use in judgment. Category accessibility was conceived as a state of perceptual readiness that makes the category available for This research was supported by Grants 410-2002-0964 from the Social Sciences and Humanities Research Council of Canada and Grant 138483-98 from the Natural Sciences and Engineering Research Council of Canada. We thank the stereotype discussion group at the University of Waterloo for many helpful contributions to the development of these ideas, and Christian Jordan and Mark Zanna for earlier version of this article. Correspondence concerning this article should be addressed to either Ziva Kunda or Steven J. Spencer, Department of Psychology, Univer- sity of Waterloo, Waterloo, Ontario N2L 3G1, Canada. E-mail: zkunda@ watarts.uwaterloo.ca or [email protected] & Fazio, 1997; Plant & Devine, 1998). 522 ACTIVATION AND APPLICATION 523 can sometimes be aware that a stereotype is on their mind, and can intentionally retrieve a stereotype when asked to (e.g., Devine, 1989). The inhibition of stereotype activation may also result from either automatic or controlled processes. People can intentionally suppress unwanted stereotypic thoughts (Wegner, 1994), but may also do so without intention or awareness (Moskowitz et al., 1999). Stereotype application, too, can occur either with or without in- tention and awareness. People may use applicable stereotypes intentionally to understand, explain, or predict the behaviors of others, but may sometimes form stereotypic impressions that they are not aware of (Wilson, Lindsey, & Schooler, 2000). Even when people are aware of the contents of such impressions, they may be unaware of the stereotypic influences that produced them (Green- wald & Banaji, 1995; Nisbett & Wilson, 1977). The distinction between stereotype activation and application should also not be confused with the distinction between associa- tive thinking, which draws automatically on associations based on similarity, experience, and contiguity, and rule-based thinking, which is goal driven and draws on symbolic rules, causal analysis, and logical inference (Sloman, 1996; E. R. Smith & DeCoster, 2000). Stereotype activation and application can both result from either mode. Stereotypes can spring to mind automatically through well-learned associations to stereotypic cues (Bargh, 1999), but they can also be retrieved through goal-driven inference that highlights their potential to satisfy perceivers’ goals (Sinclair & Kunda, 1999). Similarly, stereotypes can be applied through a process of automatic, association-based assimilation (Devine, 1989), but can also be applied as a result of rule-based motivated reasoning (Fein & Spencer, 1997). A Goal-Based Framework for Stereotype Activation and Application members of Most theories of stereotyping have assumed that stereotypes will be on the minds of people dealing with members of stereotyped groups, and focused on understanding when these activated ste- reotypes will be applied. However, although early research sug- gested that anyone who encounters a member of a stereotyped group activates that group’s stereotype spontaneously (cf. Bargh, 1999; Devine, 1989), it is now clear that such spontaneous stereo- type activation is neither inevitable nor universal (cf. Blair, 2002); it depends on perceivers’ prejudice, goals, cognitive resources, and learned associations (Gilbert & Hixon, 1991; Kawakami, Dovidio, Moll, Hermsen, & Russin, 2000; Lepore & Brown, 1997; Sinclair & Kunda, 1999). More interesting, even if stereotypes are activated at the start of an interaction with a stereotyped individual, this activation can dissipate as the interaction continues (Kunda, Davies, Adams, & Spencer, 2002). In more than half a dozen studies, we have found no trace of stereotype activation in participants who had observed or interacted with a Black or an Asian individual for about 10 min (Kunda, Davies, Hoshino-Browne &, Jordan, 2003). We also found, however, that various events that occur at this later point in an interaction can bring the stereotypes back to mind. We propose that events influence stereotype activation and application during interaction through their impact on perceivers’ shifting goals. We suggest that both the activation and the application of stereotypes at a given time depend on the extent to which perceiv- ers are driven at that time by comprehension goals, self- KUNDA AND SPENCER Self-Enhancement Goals motivational Self-enhancement goals encompass the needs to maintain, pro- tect, and enhance self-esteem (Taylor & Brown, 1988; Tesser & Campbell, 1983), affirm self-worth (Steele, 1988), and preserve valued aspects of self-concepts (Aronson, 1968); these relate to the ego-defensive function of attitudes (D. Katz, 1960). Self- enhancement goals can motivate people to form particular impres- sions of others; one may wish to think highly of someone who has praised one and to disparage someone who has criticized one. People may activate and use applicable stereotypes that can sup- port such desired impressions (Fein & Spencer, 1997; Sinclair & Kunda, 1999; Spencer et al., 1998; for a review, see Kunda & Sinclair, 1999), much like they activate and use other beliefs that can bolster their desired conclusions (Gollwitzer & Moskowitz, 1996; Kruglanski, 1996; Kunda, 1990). Social identity theory (Tajfel & Turner, 1986) also predicted that one may derogate members of out-groups to establish the superiority of the in-group and so boost one’s self-worth. However, most research derived from that theory did not address stereotype activation or applica- tion because it examined reactions to minimally defined groups with no preexisting stereotypes, or assessed overall judgments about real-life groups rather than about individuals who belonged to them (for a review, see Brown, 2000). When stereotypes can interfere with self-enhancement goals, people may suppress them. People may inhibit stereotypes that might undercut their desired impression of a person; people mo- tivated to form a positive impression of someone may inhibit applicable negative stereotypes that contradict this impression (Sinclair & Kunda, 1999) much like they inhibit other information capable of interfering with goal completion (Gernsbacher, 1997; Zacks & Hasher, 1994). function The Motivation to Avoid Prejudice The motivation to avoid prejudice may inhibit stereotype acti- vation and suppress the application of activated stereotypes. Atti- tudes toward ethnic minorities and other stigmatized groups are often fraught with conflict and ambivalence. Thus, White Ameri- cans may subscribe to symbolic racism, that is, the belief that African Americans challenge core values of the Protestant Ethic and, at the same time, also endorse egalitarian values that prohibit prejudice (I. Katz & Hass, 1988; Kinder & Sears, 1981). Their motivation to avoid prejudice may stem from an internally driven desire to maintain an egalitarian identity or from an externally driven desire to comply with egalitarian social norms (Dunton & Fazio, 1997; Plant & Devine, 1998). The internal motive relates to the value-expressive function of attitudes (M. B. Smith et al., 1956) and encompasses the needs to express, verify, and act in accordance with central tenets of the self (Aronson, 1968; Swann & Read, 1981). The theory of aversive racism (Gaertner & Dovidio, 1986) focused mostly on internal motivation to control prejudice: It assumed that people hold egal- itarian values that prevent them from openly reacting in ways that could be construed as prejudiced, but that their negative feelings and beliefs about certain groups may nevertheless color their reactions to members of these groups when these reactions may also be construed in alternative, nonprejudiced ways. Put differ- ently, because of the motivation to maintain an egalitarian self- ACTIVATION AND APPLICATION 525 supporting this explanation, the stereotype’s activation increases. Thus, stereotypes that can facilitate goal attainment may be acti- vated and used for this purpose. If one repeatedly uses stereotypes to satisfy a goal, this strategy may come to be triggered automat- ically whenever that goal is activated in the presence of stereo- typed individuals (Spencer et al., 1998). The contents of the stereotype can determine its goal relevance, because such relevance is established through causal reasoning about the implications of stereotypic attributes for goal attainment. For example, a negative group stereotype may corroborate a neg- ative impression of a group member, but undermine a positive impression. Therefore, it may be activated and used when self- enhancement goals create the subgoal of disparaging a group member, but inhibited when the subgoal is to esteem a group member (Sinclair & Kunda, 1999). Similarly, when one is moti- vated by comprehension goals to predict a person’s attributes, an applicable stereotype can contribute to this only if it contains information deemed relevant; people may activate and use the stereotype of Asian Americans as interested in science to predict an Asian student’s major (Kunda et al., 2003), but are unlikely to use the stereotype of fat people for such predictions. Although the contents of a stereotype can determine whether it is activated or inhibited, a stereotype may also be inhibited simply because it is a stereotype, regardless of its contents. Comprehen- sion goals can prompt such inhibition when the presence of other information about a person convinces perceivers that any stereotype-based inference will disrupt comprehension (Leyens et al., 1992). The motivation to avoid prejudice can also prompt such inhibition, when perceivers believe that any stereotyping may indicate prejudice (Gaertner & Dovidio, 1986). asso- Overview Leirer, The observation of a member of a stereotyped group can prompt the activation of that group’s stereotype, if the perceiver pays attention to the person’s group membership (cf. Fiske & Neuberg, 1990). Stereotype activation can be increased if the perceiver is motivated to achieve comprehension or self-enhancement goals that can be satisfied by stereotyping the person. Such goals can also increase the application of the activated stereotype to the person. However, an activated stereotype can also trigger the motivation to avoid prejudice, which can dampen stereotype acti- vation and can undercut and even reverse stereotype application. These influences are shown in Figure 1. The mere activation of a stereotype can influence judgments about a stereotyped individual through automatic assimilation, much like activated stereotypes color impressions of individuals who do not belong to the stereotyped group (Devine, 1989; Lepore & Brown, 1997). Such automatic assimilation is depicted by the arrow leading directly from stereotype activation to stereotype application, and is not the focus of our discussion. Rather, we focus on the other route, whereby a group stereotype is applied to an individual because of the individual’s membership in the group, and stereotyping is increased or prevented because of its implica- tions for perceivers’ goals. Our assumption that stereotypes can be applied through automatic assimilation and through rule-based inference is shared by other models of stereotyping (Bodenhausen & Macrae, 1998; Devine, 1989) and of social cognition (Chaiken & Trope, 1999). We also assume that the two kinds of processes KUNDA AND SPENCER arrow with a plus sign leading from one component to another the second, whereas a thin arrow with a minus sign indicates that the first on context (e.g., Brewer, 1988) did not elaborate on this, most likely because of the dearth of relevant research at the time. With the recent explosion of research on stereotype activation, it is now clear that stereotype activation, like application, can vary from one situation to another. We therefore focus on both processes. We assume that the perceiver can entertain multiple goals si- multaneously, and that these can jointly determine stereotype activation and application. In contrast, most previous theories that assumed that goals could influence stereotyping focused on only one goal at a time. Theories that assumed that stereotypes often dominate impressions paid little attention to the pressure to sup- press stereotyping that can be exerted by the motivation to avoid prejudice (Brewer, 1988; Fiske & Neuberg, 1990). And theories about the inhibitory role of the motivation to avoid prejudice paid little attention to the countervailing pressures to stereotype that can be exerted by comprehension and self-enhancement goals (Devine, 1989; Fazio et al., 1995; Gaertner & Dovidio, 1986; Monteith, Sherman, & Devine, 1998). Our approach is distinctive in its emphasis on situational influ- ences on stereotyping. Others have focused predominantly on enduring influences such as perceivers’ chronic levels of motiva- tion to avoid prejudice (Dunton & Fazio, 1997; Plant & Devine, 1998) or to self-enhance (Tajfel & Turner, 1986), or the fit be- tween the features of a target and a stereotype (Fiske & Neuberg, 1990). We acknowledge such enduring influences but emphasize the impact of temporary, situation-driven goals and provide a detailed analysis of diverse situations that can intensify or weaken perceivers’ goals and, thereby, affect stereotype activation and application. We differ from earlier theories in our assumptions about how goals influence stereotyping. Key theories of stereotyping (Brewer, 1988; Fiske & Neuberg, 1990) assumed that goals influence ste- ACTIVATION AND APPLICATION 527 that each factor exerts its impact by influencing the amount of attention that the perceiver devotes to the target person’s member- ship in the stereotyped group or by influencing the intensity of the perceiver’s self-enhancement goals, comprehension goals, or mo- tivation to avoid prejudice. We typically infer that a goal has been influenced by a given factor on the basis of other research relating that factor to that goal or on the basis of a conceptual analysis of the factor’s impact; direct measures of goal intensity have not been common in this research. The factors are organized according to the goals they are assumed to influence, and the direction of their impact on stereotype activation and application is shown. We assume that factors that influence goal intensity exert comparable impact on stereotype activation and application because stereo- types are activated (or inhibited) to facilitate goal attainment by promoting or preventing stereotype application. However, for many factors, there is currently evidence of impact on only one of these processes. We first review the factors shown to influence stereotype activation, and then turn to stereotype application. that stereotype and Stereotype Activation Attention to the Target’s Category and Stereotype Activation its activa- Perceivers who encounter a member of a stereotyped group will activate that group’s stereotype only if they are aware of the person’s category membership and have the cognitive resources needed to process this information. Any factor that diverts the perceiver’s attention from the person’s category membership can prevent the stereotype from getting activated in the first place, or cause its initial activation to dissipate. Initial exposure. When one encounters a new person, the person’s salient category membership can capture one’s immediate attention (cf. Brewer, 1988; Fiske & Neuberg, 1990). This can serve comprehension by permitting rapid classification. Classify- ing a person as a member of a social group can entail the activation of that group’s stereotype. There is considerable evidence that stereotypes can be activated spontaneously on exposure to a ste- reotyped individual (for a review, see Bargh, 1999). In several studies, a very brief video of an Asian or a Black person prompted the activation of the relevant stereotype (Gilbert & Hixon, 1991; Kunda et al., 2002; Macrae et al., 1995; Pendry & Macrae, 1996; Sinclair & Kunda, 1999). Such stereotype activation can be auto- matic; it can occur even when the exposure is subliminal or too brief to permit controlled processing. Thus, subliminal exposure to photographs of Black faces led non-African Americans to activate the African American stereotype (Spencer et al., 1998), to behave aggressively toward a White individual (Bargh et al., 1996; Chen & Bargh, 1997), and to misidentify tools as weapons (Payne, 2001). Similarly, subliminal or very brief exposure to Black faces led many White participants to activate negative affect (Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; Fazio et al., 1995). Such effects were especially pronounced for highly proto- typical Black faces (Livingston & Brewer, 2002). Perceivers’ level of prejudice may determine whether they spon- taneously activate group stereotypes on encountering a group member. When primed with affectively neutral stereotypic asso- ciates, participants high in prejudice activated stereotypes whereas those low in prejudice did not (Kawakami, Dion, & Dovidio, 1998; KUNDA AND SPENCER Influence Stereotype Activation and Application by Influencing to Avoid Prejudice Studies in which the factor influenced Studies in which the factor influenced stereotype activation stereotype application (⫹) Hoshino-Browne & Kunda, 2000 (⫹) Hoshino-Browne & Kunda, 2000 (⫺) Hoshino-Browne & Kunda, 2000 (⫹) Kunda et al., 2002 (⫺) Jordan & Kunda, 2000 (⫺) Locksley et al., 1980c Rasinski et al., 1985 (⫺) Macrae et al., 1995 (⫹) Neuberg & Newsom, 1993 (⫹) Sinclair & Kunda, 1999 (⫹) Sinclair & Kunda, 1999 (⫺) Sinclair & Kunda, 1999 (⫹) Spencer et al., 1998 (⫹) Fein & Spencer, 1997 Greenberg et al., 1990 (⫺) Fein & Spencer, 1997 Muklincer & Shaver, 2001 (⫺) Moskowitz et al., 1999 (⫺) Fein et al., 2003 (⫺) Fein et al., 2003 (⫺) Macrae et al., 1998 (⫺) Macrae et al., 1998 (⫺) Fein et al., 2003 (⫺) Fein et al., 2003 (⫹) Darley & Gross, 1983 Yzerbyt et al., 1994 Gaertner & Dovidio, 1977 Snyder et al., 1979 (⫹) Spencer & Fein, 2000 stereotype activation or application; (⫺) ⫽ the cited study found the factor to decrease stereotype which explains its negative impact on stereotype application. b This factor likely also to its impact. c This finding was obtained in many other studies as well. For review, see immediate exposure to stereotyped individuals on stereotype acti- vation is moderated by prejudice. Passage of time. Initially activated stereotypes can dissipate with time. Their activation may decay rapidly as one’s attention is drawn to other incoming stimuli, as is the case for semantic priming, whose effects can be eliminated by the mere interposition of several unrelated words between a prime and a target word (Neely, 1991). In one set of studies, participants viewed a video- tape of a Black or White student discussing campus life (Kunda et al., 2002). Their activation of the Black stereotype was assessed reactions after they had been watching the video for either 15 s or 12 min. Participants assessed within 15 s of their initial exposure to the Black target showed stereotype activation, but participants as- activated sessed after 12 min of exposure did not. The initially activated stereotypes had dissipated with time. Several additional studies also found that stereotypes were not activated for perceivers who had been interacting with or observing a video of a Black or Asian individual for 10 –15 min (Kunda et al., 2002, 2003). Thus, even though applicable stereotypes can be activated spontaneously as soon as one encounters a person, their activation may fade within less than 15 min of further exposure to the person. As time unfolds, one’s attention shifts from the person’s category membership to individuating information or to the demands of the tasks at hand. ACTIVATION AND APPLICATION 529 prompt the activation of relevant stereotypes. Gender may come to mind when one needs to predict a person’s ability to carry heavy weights, and profession when one wishes to determine a person’s income. In one set of studies, participants took part in a structured interview with an Asian or White confederate during which the confederate asked the participant stereotype-irrelevant questions (Hoshino-Browne & Kunda, 2000). Following 10 –15 min of such interaction, half of the participants were given the goal of forming an impression of their interaction partner’s personality and likely career choice. Controls were given, instead, the goal of elaborating on the contents of their discussion. As may be expected from the finding that stereotype activation can dissipate by the end of such a lengthy encounter (Kunda et al., 2002), controls interacting with an Asian confederate showed no activation of the Asian stereotype. In contrast, participants given the task of forming an impression of their Asian partner did activate the Asian stereotype. Most likely, they recruited the stereotype so as to inform their impressions of this person. This finding sheds new light on research on stereotype applica- tion. One may be tempted to conclude that participants who applied a stereotype to a stereotyped individual had that stereotype on their minds throughout their exposure to this individual. How- ever, this conclusion may be unwarranted because measures of stereotype application typically request participants to judge the target on stereotypic dimensions. Such requests may suffice to prompt the activation of stereotypes that might have otherwise remained inactive. Therefore, the widespread evidence of stereo- type application (for a review, see Fiske & Neuberg, 1990) may have led to a considerable exaggeration of the extent to which stereotypes are routinely on perceivers’ minds. Sensitive discussion topic. When issues sensitive to a stereo- typed group arise during interaction with a group member, the stereotype becomes relevant to predicting that person’s likely reactions, and may be activated for that purpose. A gentile inter- acting with a Jew may be reminded of that person’s Jewishness when Nazi war crimes come up, and a White American interacting with an African American may be reminded of that person’s ethnicity when affirmative action comes up. In one study, White Canadians took part in a 10 –15-min struc- tured interview with an Asian or a White confederate (Hoshino- Browne & Kunda, 2000). On each round of the interview, the participant and the confederate were given 1 min to think about a newly introduced discussion topic. The confederate then inter- viewed the participant about the topic. The first few topics were stereotype unrelated, but the final one was either a topic assumed to be sensitive to Asian Canadians—Canada’s immigration pol- icy— or a neutral topic—Canada’s health policy. After participants had thought about this topic, but before they had a chance to discuss it, their activation of the Asian stereotype was assessed. Participants expecting to discuss the neutral topic with an Asian person showed no stereotype activation, but participants expecting to discuss the sensitive topic with an Asian person did activate the Asian stereotype. The thoughts listed by participants revealed that they relied on their partner’s ethnicity to tailor their own reac- tions—participants expressed much less negativity toward immi- gration when they expected to discuss it with an Asian than with a White person; most likely, they were trying to avoid any appear- ance of prejudice. It appears that perceivers who are no longer cognizant of the stereotype of their interaction partner’s group may KUNDA AND SPENCER individual may inhibit the activation of the stereotype, much like people inhibit other irrelevant information (Zacks & Hasher, 1994). In one set of studies, participants watched a video in which a Black or a White student discussed stereotype-unrelated issues for 12 min before mentioning that he had either failed a recent test (stereotype-consistent information) or excelled at it (stereotype- inconsistent information; Jordan & Kunda, 2000). Controls who had viewed the Black student but received no information about his test performance did not activate or inhibit the Black stereo- type, but participants who had received diagnostic information about the Black student inhibited the activation of the Black stereotype, and did so regardless of whether this information was consistent or inconsistent with the Black stereotype. Thus, the acquisition of diagnostic stereotype-related information about a stereotyped person during an encounter with this person can high- light the irrelevance of the stereotype to the person and prompt its inhibition. Alternative stereotype. An applicable stereotype may also be inhibited if it comes to seem irrelevant when a different applicable stereotype becomes more salient. In one study, participants watched a brief silent video of a Chinese woman behaving in a manner that highlighted either her ethnicity (eating with chop- sticks) or her femininity (applying makeup; Macrae et al., 1995). Participants activated the more salient stereotype but inhibited the competing one. Every person belongs to multiple stereotyped groups; highlighting one of these group memberships can prompt the inhibition of stereotypes associated with others. In sum, stereotyping can sometimes facilitate comprehension goals and can sometimes interfere with such goals. When, during interaction, perceivers experience an increase in the intensity of comprehension goals that can be achieved by using an applicable stereotype, they may activate it. And when they encounter infor- mation that undercuts the relevance of an applicable stereotype to comprehending their interaction partner, they may inhibit it. Self-Enhancement Goals and Stereotype Activation Self-enhancement goals can sometimes be satisfied by dispar- aging or esteeming another person. Events that strengthen such goals may prompt the activation of stereotypes that support the desired impression and the inhibition of stereotypes that under- cut it. Receiving praise or criticism from the target. When one is praised, one may become motivated to think highly of the evalu- ator so as to maximize the self-enhancing potential of the praise. But when one is criticized, one may become motivated to dispar- age one’s evaluator so as to diffuse this personal threat. Perceivers may activate those applicable stereotypes that can support such a desired impression of their evaluator, and may inhibit those that can interfere with it. Motivated stereotype inhibition was found in three studies in which participants observed a brief video of a Black or a White person delivering positive or negative feedback on their performance on a leadership test (Sinclair & Kunda, 1999). Participants praised by a Black person inhibited the activa- tion of the Black stereotype, most likely because of its potential to undermine their desired positive impression of this person. For similar reasons, participants motivated to disparage their evaluator inhibited an applicable positive stereotype: Participants criticized ACTIVATION AND APPLICATION 531 women when primed with female faces, but participants with chronic egalitarian goals did not. A follow-up study revealed that this lack of stereotype activation by chronic egalitarians was not due to their lack of associations between female cues and the stereotype of women but, rather, to their inhibition of that stereo- type. Thus, people with chronic egalitarian goals can inhibit ste- reotype activation in the service of these goals when they encoun- ter stereotyped individuals. Challenge to impartiality. People may sometimes inadver- tently behave in a manner that can be construed as racist or sexist. If bigotry runs contrary to their personal standards or to their understanding of societal norms, they may find such challenges to their impartiality disturbing and become motivated to avoid being or appearing prejudiced (Monteith, 1993; Sherman & Gorkin, 1980). To satisfy this goal, they may try not to think about others in stereotypic terms, and inhibit applicable stereotypes. In one set of studies, participants first received a challenge to their impar- tiality: they were told that a test they had taken previously sug- gested that they were racist (Fein et al., 2003). Controls received racism-neutral feedback. Participants then read an article about a Black or White lottery winner, and performed a word-fragment- completion task in which words were preceded by subliminally presented Black or White faces (the race of the lottery winner and of the primes was always the same). The word-fragment- completion task assessed their activation of the Black stereotype. All participants were cognitively busy during the study. Like cognitively busy participants in other studies, recipients of neutral feedback primed with Black individuals did not activate or inhibit the Black stereotype (Gilbert & Hixon, 1991; Spencer et al., 1998). In contrast, participants who had been told that they were racist inhibited the Black stereotype when primed with Black individuals. Similar results were obtained in studies where participants’ impartiality was challenged by their own seemingly bigoted be- havior. In one study, participants learned that they had failed to realize that a surgeon could be a woman (Sherman & Gorkin, 1980); in another, participants were forced to choose among sets of responses that included only sexist options (Moskowitz et al., 1999); and in another study, participants were led to believe that they had judged a man negatively because he was gay (Monteith, 1993). These studies found that threats to impartiality that stem from one’s own behavior can make people feel uncomfortable and increase their motivation to avoid prejudice (Monteith, 1993; Mos- kowitz et al., 1999; Sherman & Gorkin, 1980). Such threats may prompt people to purge their minds of stereotypic thoughts, but there is no evidence yet for stereotype inhibition in such circumstances. Self-focus. When attention is focused on the self, personal and societal norms become salient (Carver & Scheier, 1981). A height- ened self-focus in the presence of a stereotyped person may therefore highlight antiprejudice norms and increase the motiva- tion to adhere to them. This may prompt attempts to inhibit applicable stereotypes (Macrae et al., 1998). One relevant study was based on the assumption that when one deliberately tries to suppress certain thoughts, these thoughts become hyperaccessible later when one is no longer trying to suppress them (Wegner & Erber, 1992). Suppressed stereotypes should thus rebound once perceivers stop trying to suppress them. Therefore, if perceivers who have an experience that can prompt stereotype inhibition later KUNDA AND SPENCER tivated stereotypes to a stereotyped individual (Gilbert & Hixon, 1991; participants busy only during application phase). On other occasions, perceivers do not apply activated stereotypes (Sinclair & Kunda, 1999; detached observers). On yet other occasions, perceivers engage in counterapplication of activated stereotypes (Fein & Spencer, 1998). We assume that perceivers will be more likely to apply an activated group stereotype to a group member when stereotyping this person can further their comprehension and self-enhancement goals (cf. Hamilton & Sherman, 1994). There- fore, events that trigger these goals may increase stereotype appli- cation, whereas events that decrease preoccupation with these goals or undercut the extent to which stereotyping can satisfy them may reduce stereotype application. However, perceivers are likely to avoid applying activated stereotypes when they are driven by the motivation to avoid prejudice (cf. Devine, 1989). Therefore, factors that increase this motivation may reduce stereotype appli- cation, whereas factors that decrease it may increase stereotype application. However, controlling prejudice is an effortful process that can be disrupted when the perceiver lacks sufficient resources (Gilbert & Hixon, 1991). Therefore, the impact of the motivation to avoid prejudice on stereotype application may be moderated by the availability of cognitive resources. When one’s goals have conflicting implications for stereotype application, their relative strength may determine whether one applies the stereotype. People who would usually refrain from stereotyping because they are motivated to avoid prejudice may nevertheless apply a stereotype if they are sufficiently driven by self-enhancement goals that can be satisfied by stereotyping (Fein & Spencer, 1997); and people who would usually apply a stereo- type to further comprehension may nevertheless refrain from ap- plying it when they become motivated to avoid prejudice (Hoshino-Browne & Kunda, 2000). In most of the studies reviewed in this section, stereotype application was assessed from ratings of target individuals on stereotypic attributes. We also review some studies that assessed attitudes toward the target from behaviors such as offering the target help (Gaertner & Dovidio, 1977) or sitting near the target (Snyder, Kleck, Strenta, & Mentzer, 1979) because the application of attitudes and of beliefs about a group can be governed by the same factors. For example, giving people an excuse for seemingly prejudiced reactions can increase the application of both stereo- types (e.g., Yzerbyt et al., 1994) and attitudes (e.g., Gaertner & Dovidio, 1977). Motivation to Avoid Prejudice and Stereotype Application exerted The motivation to avoid prejudice can lead perceivers to inhibit the activation of stereotypes applicable to individuals they encoun- ter. This could decrease stereotype application by reducing auto- matic assimilation of the target’s behavior to the stereotype (see Figure 1). But even when a stereotype’s activation has not been quashed, the motivation to avoid prejudice can decrease its appli- cation through a more deliberative process; perceivers can actively avoid stereotyping others if they have the necessary cognitive resources (Devine, 1989; Fazio & Towles-Schwen, 1999). Several studies that assessed both stereotype activation and application have found that perceivers may refrain from applying an activated group stereotype to the group member whose very presence had provoked its activation (Gilbert & Hixon, 1991, never busy par- ACTIVATION AND APPLICATION 533 causes. In recognition of this, contemporary scales of prejudice such as the Modern Racism Scale (McConahay, 1986) and the Symbolic Racism Scale (Kinder & Sears, 1981) exclude explicit questions that focus directly on beliefs and feelings about nega- tively stereotyped groups, and attempt to tap negativity instead through indirect questions that ask, for example, about attitudes toward busing students to avoid segregation. The underlying as- sumption is that people who would not express explicit negativity toward a negatively stereotyped group such as African Americans would nevertheless be willing to express negative views of this group if these could be attributed to nonracist beliefs. The same holds for reactions to individual group members. People often consider it inappropriate to apply group stereotypes to individuals, but may nevertheless do so if such stereotypic reac- tions can also be attributed to other sources of information. In one study, participants viewed a brief video of a girl whose socioeco- nomic background was either high or low (Darley & Gross, 1983). Participants given only this social-class information did not apply it to their judgments of the girl’s intelligence, but participants shown an additional video of the girl performing ambiguously on an intelligence test did; they judged her as more intelligent if her social class was high than if it was low. The test performance provided an excuse for using a stereotype whose use would oth- erwise seem inappropriate—stereotype-based judgments could now be attributed to a more legitimate source of information, the girl’s performance. This possibility gains support from a different study in which participants received only an illusion of having individuating in- formation (Yzerbyt et al., 1994). Participants first listened to an uninformative interview with a person identified at the end as either a librarian or a comedian. They then had to shadow a voice heard in one ear while a different voice was being heard in the other ear. Half the participants were later informed (falsely) that the unattended voice had communicated individuating information about the target (Yzerbyt et al., 1994). These participants’ illusory belief that they had obtained individuating information freed them to rely on stereotypes that they would have otherwise avoided using—they rated the librarian as more introverted than the come- dian, whereas participants without this illusory belief did not. Thus, the actual or illusory presence of individuating information can provide an excuse for relying on stereotypes without appearing to violate antistereotyping norms. The presence of excuses for discrimination can also facilitate discriminatory behavior. One relevant study (Gaertner & Dovidio, 1977) built on an earlier finding that people who encounter a person in distress are less likely to offer help when they believe other witnesses are present (Latane & Darley, 1970). Another witness can provide a justification for one’s own inaction: One may assume that the other witness will handle the problem or, if the other fails to act, that the problem is not serious enough to warrant intervention. If the person in distress is Black, the avail- ability of such excuses may diffuse perceivers’ concern that their failure to help might be construed as reflecting prejudice. In a study based on this logic (Gaertner & Dovidio, 1977), White participants believed that they were listening to another partici- pant, who was either White or Black, over an intercom (in fact, they listened to prerecorded audiotapes). They also believed either that they were the only listener or that there were two additional listeners. When the speaker appeared to experience an emergency, KUNDA AND SPENCER series of studies (Monin & Miller, 2001), participants who were first given an opportunity to reject blatantly sexist statements were subsequently more likely to endorse subtly sexist statements. Peo- ple who have established their egalitarian credentials are also more willing to apply group stereotypes to individual group members. In one study, participants’ confidence in their egalitarian credentials was boosted by informing them that a test taken earlier revealed that they were low in racism. Others received no feedback about their racism (Spencer & Fein, 2000). Participants who had just been informed that they were low in racism, and who were therefore confident that their judgments would not be taken as prejudiced, subsequently judged a Black job candidate more 1979). negatively than did controls, and were less likely to prefer him to a stronger White candidate. Confident in their egalitarian monitor; credentials, these participants relaxed their efforts to avoid prejudice, and expressed negativity that they would have oth- erwise curtailed. Challenges to impartiality. When one comes under suspicion of prejudice, one may become especially motivated to ensure that one’s reactions to stereotyped individuals are unbiased. Therefore, people who inadvertently behave in a seemingly prejudiced man- ner may subsequently go out of their way to prove that they are not prejudiced. Indeed, inducing people to show sexism in one setting can cause them to express less sexism in another setting (Mos- kowitz et al., 1999; Sherman & Gorkin, 1980). Challenges to impartiality can also influence reactions to stereotyped individuals. In one study, White participants viewed a videotaped student discussing race-unrelated issues (Fein et al., 2003). In one condi- tion the student was White. In three other conditions he was Black, and mentioned either that he had or had not experienced racism on campus, or did not mention racism. When a Black student claims to be a victim of campus racism, this challenges the impartiality of any White student on campus. Indeed, participants so challenged rated the Black student considerably more positively than did participants in the other three conditions. Moreover, for partici- pants who had come under suspicion of prejudice, the motivation to avoid prejudice was powerful enough to prompt them to over- correct for possible stereotypic influences—they rated the Black student who had challenged their impartiality more positively than they rated the White student. The motivation to avoid prejudice partici- can not only undercut stereotype application, it can also reverse it, the defen- resulting in counterapplication. and so Sensitive topics. When a conversation with a member of a stereotyped group turns to a topic sensitive to that group, one may worry that one might inadvertently let slip comments that might be taken as prejudiced. Earlier, we described a study in which par- ticipants who expected to discuss a topic sensitive to Asian Cana- dians—Canada’s immigration policies—with an Asian Canadian activated the Asian stereotype (Hoshino-Browne & Kunda, 2000). The thoughts that these participants listed in preparation for this

When Do Stereotypes Come to Mind? (2003) PDF

Document Details

Tags

Related

Summary

Full Transcript