A Causal Test of the Stereotype Content Model

Group Processes & Intergroup Relations 2009 Vol 12(2) 147–155

Social Structure Shapes Cultural Stereotypes and Emotions: A Causal Test of the Stereotype Content Model

Peter A. Caprariello University of Rochester Amy J. C. Cuddy Harvard University Susan T. Fiske Princeton University

The stereotype content model (SCM) posits that social structure predicts specifi c cultural stereotypes and associated emotional prejudices. No prior evidence at a societal level has manipulated both structural predictors and measured both stereotypes and prejudices. In the present study, participants (n = 120) responded to an immigration scenario depicting a high- or low-status group, competitive or not competitive, and rated their likely stereotype (on warmth and competence) and elicited emotional prejudices (admiration, contempt, envy, and pity). Seven of eight specifi c predictions are fully confi rmed, supporting the SCM’s predicted causality for social structural effects on cultural stereotypes and emotional prejudices. keywords competence, competition, emotions, status, stereotypes, warmth

Theorists have long noted that social structural while perceived group status causes their stereo- relationships among groups are responsible for typic competence. Furthermore, systematic com- observed patterns of stereotypes, prejudice, and binations of warmth and competence predict discrimination (Allport, 1954; Sherif, Harvey, affective reactions to members of outgroups White, Hood, & Sherif, 1961). Two social struc- deﬁ ned by those dimensions (Cuddy et al., 2007; tural variables in particular, perceived group status Fiske et al., 2002). and competition, can form the foundation for ingroup members’ responses to outgroup mem- Author’s note bers at a societal level. According to the Stere- Address correspondence to Peter A. Caprariello, otype Content Model (SCM; Cuddy, Fiske, & University of Rochester, Clinical & Social Glick, 2007; Fiske, Cuddy, & Glick, 2007; Fiske, Sciences in Psychology, Meliora Hall, RC Cuddy, Glick, & Xu, 2002), perceived intergroup Box 270266, Rochester, New York 14627, USA competition lowers groups’ stereotypic warmth, [email: [email protected]]

1147-155_GPI_101053.indd47-155_GPI_101053.indd 114747 112/17/20082/17/2008 3:31:383:31:38 PMPM PProcessrocess BlackBlack Group Processes & Intergroup Relations 12(2)

The infl uences of social structure on subse- warmth (or lack thereof). It is equally important quent stereotypes and emotions are presumed to verify that a group’s actual status in relation to to be causal in nature, but the assumption of other groups causes perceptions of competence causality has yet to be tested. A number of studies and drives emotional reactions. Status and com- have explored relationships between the three petition might catalyze group bias for a number sets of variables, though for the most part the of reasons. By taking into account structural relationships have been tested correlationally relationships between groups (i.e. perceived (see Figure 1). Two experimental studies (Cuddy competition–cooperation and group status dif- et al., 2007) confi rmed causal relationships ferential), a target’s group membership provides between stereotypes, emotions, and behaviors. the information necessary to quickly answer two However, the social-structural hypothesis that questions that help guide interaction. First, per- relative status and interdependence determine ceivers need to know if a social entity intends to the content of stereotypes and emotions re- help or harm, and, second, if the entity can carry mains untested. The present research aims to out its intent. If one assumes that social perception fi ll that gap. operates in the service of interaction goals Status and competition consistently correlate (Fiske, 1992), then it follows that understanding in the expected direction with competence and an outgroup’s intentions and capabilities of warmth stereotypes, as well as with specifi c emo- carrying out its intentions will be a primary tions and behaviors (Cuddy et al., 2007; Eckes, motive in perceiving social entities. The warmth 2002; Fiske, Xu, Cuddy, & Glick, 1999; Fiske and competence dimensions appear to be et al., 2002), even cross-culturally (Cuddy, Fiske, universal and fundamental features of social Kwan, et al., in press). Adopting an experimental perception (Cuddy, Fiske, & Glick, 2008; Fiske approach to test the causal link between social et al., 2007). structure and stereotypes is necessary. Theor- Competition answers the question of intent. etically, it is important to verify that a group’s Competition pits the desired resources of one competitive tendencies drive perceptions of social group against the other, and in order to

Support for A. Support for B. Support for C. Fiske, Cuddy, & Glick (1999) Fiske et al. (2002) Cuddy, Fiske, & Glick (2007) – correlational data – correlational data – correlational data Fiske et al. (2002) Cuddy, Fiske, & Glick (2007) – experimental data – correlational data – correlational data Eckes (2002) – experimental data – correlational data Cuddy et al. (in press) – cross-cultural correlational data

A. B. C.

Societal Stereotypes Prejudice Discrimination Structure

This causal link remains untested. Figure 1. Correlational support for the hypothesized causal links between societal structure and group bias.

148

1147-155_GPI_101053.indd47-155_GPI_101053.indd 114848 112/17/20082/17/2008 3:31:463:31:46 PMPM PProcessrocess BlackBlack Caprariello et al. causal test of the stereotype content model

compete, one must intend to maximize one’s successfully in their careers, indicating posi- resources over the others. Perceiving a competitor tive competence-based expectations (Aquino, (or a competing group) and knowing their Stewart, & Reed, 2005). In an experiment that potentially confl icting goals aids the evaluation manipulated status by varying the cost of a target of a group’s intentions as warm or not warm, in person’s house, occupants of more expensive order to recognize a potential threat to resources houses were perceived as more competent (but (Fiske, 1993). Knowing that a group intends not warm) than occupants of less expensive to compete for resources suggests that group houses (Oldmeadow & Fiske, 2007). In an members have negative intentions toward others experiment that manipulated both status and (cold, unfriendly, and untrustworthy), while competition between individuals, perceived com- knowing that a group intends to cooperate sug- petence and warmth appeared, as the SCM gests positive intentions toward others (warmth, would predict (Russell & Fiske, in press), but the friendliness, and trustworthiness). setting was again individual. Status answers the question of capability. Status Group-level experiments have shown that some indicates an ability to derive resources from low-status groups are considered high in sociabil- others. Groups with high status typically have high ity and low in competence, whereas some high- power as well, indicating the ability to control status groups are considered low in sociability and provide resources. Both status and power, and high in competence (Betancor, Rodriguez, however, are defi ned by their abilities to regu- Rodriguez, Leyens, & Quiles, 2005). Furthermore, late resources, and therefore recognition of working mothers’ stereotypic low competence status is inherently linked to perceived com- (Cuddy, Fiske, & Glick, 2004) apparently stems petence (Fiske, 1993). Knowing that a group from the diminished status associated with not only intends to derive resources but has the juggling careers and motherhood (Ridgeway & ability to do so suggests that group members Correll, 2004). Therefore, experimental evidence are competent, while a lack of ability suggests suggests that changes in status are causal factors the opposite. in interpreting an individual group member’s abilities or a specifi c group’s abilities. Status, competition, and stereotypes Competition has also been manipulated, for of warmth and competence many years, to gauge its effects on stereotypes, starting with the Robber’s Cave study (Sherif et al., The SCM posits causal relationships among 1961). In more recent examples, competition social structure, stereotypes, and prejudice. Cor- has been manipulated as the real and symbolic relational data indicate that social structure threat posed by a Rwandan minority subgroup; predicts stereotypes, which in turn predict emo- both types of threats increased the respondents’ tions (Cuddy et al., in press; Fiske et al., 1999, negative attitudes towards the group (Stephan, 2002). However, only a few studies to date have Renfro, Esses, Stephan, & Martin, 2005). Likewise, manipulated social structural variables so as to experiments operationalized competition as infer causal infl uence on subsequent stereo- the ability of foreigners to take away jobs from types and emotions. Some of these studies mani- nationals, and they found that highly competitive pulated the status of individuals as members foreigners elicited greater discrimination than of groups. For example, in an experiment that low-competition foreigners (Falomir-Pichastor, manipulated the status and dress of a hypothet- Munoz-Rojas, Invernizzi, & Mugny, 2004). In ical female employee, a woman in a high-status other experiments, inter-group competition pro- position who dressed in a sexy manner was per- moted stereotyped impressions between com- ceived to be less competent than when dressed petitors, whereas inter-personal competition in business attire (Glick, Larsen, Johnson, & promoted individuated impressions (Ruscher, Branstiter, 2005). In another experiment, job Fiske, Miki, & Van Manen, 1991). Therefore, com- candidates who occupied higher status jobs were petition causally diminishes perceptions of expected to perform better and progress more an outgroup’s likeability and individuality.

149

1147-155_GPI_101053.indd47-155_GPI_101053.indd 114949 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Group Processes & Intergroup Relations 12(2)

But these studies manipulated only status, not correspond with the four combinations of high both status and competition simultaneously, and and low warmth and competence will predict they did not measure their respective infl uence the same patterns of emotions. on competence and warmth specifi cally. The present research Status, competition, and specifi c Our predictions suggest causal effects of social emotional prejudices structure on societal stereotypes and emotions, Perceiving other groups as warm versus cold and but evidence to support causality is lacking. No competent versus incompetent affects people’s study to date has simultaneously manipulated emotional responses to those groups. The SCM both relative status and competition at the inter- describes the systematic, differentiated patterns group level and observed changes in warmth– of emotional prejudices that tend to result competence stereotypes and emotional preju- from stereotypes of high and low warmth and dices. The goal of the present study, therefore, competence (Fiske et al., 2002). Four emotions— is to extend previous research by manipulating admiration, contempt, envy, and pity—result from the social structural variables of competition and the four combinations of high and low warmth status in order to observe subsequent changes in and competence stereotypes. Group-based emo- trait stereotypes and affective reactions, thereby tions are often activated by situational appraisals testing the proposed causal structure of the SCM. of the potential harm or benefi t the other group To do so, we conducted an experiment using poses (Mackie, Devos, & Smith, 2000).Theoret- hypothetical scenarios. Hypothetical scenarios ical support for the relationships among these allow us to fully cross structural variables and particular emotions and the four combinations of observe causal changes as they pertain to an competence–warmth stereotypes can be derived unfamiliar ethnic group, without the risk of from social comparison-based approaches to interference due to group relations constrained emotion (e.g. Smith, 2000). The uni-valent com- by actual historical and current circumstances. binations of warmth-competence stereotypes Thus, differences in competence–warmth lead to univalent emotional responses. High- traits should result directly from the causal in- warmth/high-competence stereotypes elicit fl uence of structural variables, devoid of covari- admiration, an emotion that follows from assi- ation with actual group relationships, a crucial milative (i.e. not competitive), upward (i.e. high test of the SCM. The unfamiliar ethnic groups status) social comparisons (e.g. Fiske et al., described in the scenarios varied systematic- 2002). In the opposite corner, low-warmth/low- ally in both their competition and status, in competence stereotypes elicit contempt, an order to test the hypothesis that, all else being emotion that follows from contrastive (i.e. com- equal, cultural stereotypes toward members of petitive), downward (i.e. low status) compari- outgroups result from social structural rela- sons (e.g. Dijker, Koomen, van den Heuvel, & tionships between groups. Specifi cally, we test Frijda, 1996). The ambivalent com-binations of the following: warmth-competence stereotypes lead to more ambivalent emotional responses. Low-warmth/ • High group competition will lower rated high-competence stereotypes elicit envy, an warmth. emotion that stems from contrastive, upward • Group competition will have no effect on social comparisons (e.g. Smith, Parrott, Ozer, & rated competence. Moniz, 1994). High-warmth/low-competence • High group status will increase rated com- stereotypes elicit pity, an emotion that stems petence. from downward, assimilative social comparisons • Group status will have no effect on perceived (e.g. Cuddy et al., 2007). We predict that by warmth. extension, the four combinations of high and • Low-competition, high-status groups will low intergroup competition and status that elicit admiration.

150

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115050 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Caprariello et al. causal test of the stereotype content model

• High-competition, high-status groups will group arrive, to what extent will people here elicit envy. be likely to feel each of the following emotions • High-competition, low-status groups will elicit toward them?’ and rated their emotional reac- pity. tions (see below). • Low-competition, low-status groups will elicit contempt. Warmth–competence stereotypes Participants rated their perceptions of the unfamiliar ethnic Method group’s warmth (comprising good-natured and warm, ␣ = .93) and competence (comprising Participants competent and capable, ␣ = .87), using 7-point One hundred and twenty undergraduates from scales (1 = extremely unlikely to 7 = extremely likely). a private northeastern university were recruited All warmth-competence adjectives came from from a psychology participant pool in exchange prior SCM studies (see Cuddy et al., 2007; Fiske for course credit. Participants completed the fol- et al., 2002). lowing materials as part of a larger, separate battery of questionnaires. Of the 120 participants, Emotional prejudices Using 7-point scales 31 were male (26%) and 89 were female (74%). (1 = extremely unlikely to 7 = extremely likely), par- The mean age was 20.1 years (SD = 1.6). Ethni- ticipants rated the likelihood that they and cities were not recorded. others would feel admiration (comprising admiration and pride, ␣ = .52), envy (comprising envy Materials and jealous, ␣ = .94), pity (comprising pity and sympathy, ␣ = .74), and contempt (comprising Vignettes The questionnaires depicted an contempt and disgust, ␣ = .61) toward the incom- unfamiliar ethnic group said to be immigrat- ing group.1 All emotion words came from prior ing to our country in the near future. In a 2 × 2 SCM studies (see Cuddy et al., 2007; Fiske between-subjects design, the questionnaires et al., 2002). varied the competition (high, low) and status (high, low) of the group in its home country and Results asked participants to report their perceptions of the group. Effects of competition and status on warmth Participants read: and competence stereotypes To test the ﬁ rst hypothesis, we submitted warmth Due to political and economic circumstances, demo- ratings to a 2 (Competition: high, low) × 2 graphers predict waves of immigration in the next few years from an ethnic group outside our borders (Status: high, low) ANOVA. As predicted, there called Wallonians. In their home country, members was a main effect of competition on warmth, of this group typically have prestigious jobs, and are such that competitive groups were rated as less well educated and economically successful [low- warm (M = 3.41) than non-competitive groups ␩ 2 status jobs, and are uneducated and economically (M = 3.99), F(1, 115) = 6.67, p < .02, p = .06. unsuccessful]. However, they also take power and Also as predicted by Hypothesis 2, there was no resources from [share power and resources with] main effect of status on warmth (F < 1.0) and members of other groups. When members of this no status × competition interaction (F < 1.0). ethnic group arrive here, to what extent will people Table 1 presents the mean competence and here be likely to view incoming group members in warmth ratings by condition. the following ways? To test the third hypothesis, we submitted Bolded text was bolded in the original; variant competence ratings to a 2 (Competition: high, indicated in bracketed text. Participants then low) × 2 (Status: high, low) analysis of variance rated the unfamiliar ethnic group on competence (ANOVA). As predicted, there was a main effect and warmth adjectives (see below). Next, par- of status on competence, such that members of ticipants read, ‘When members of this ethnic high-status groups were rated as more competent

151

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115151 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Group Processes & Intergroup Relations 12(2)

(M = 4.71) than low-status groups (M = 2.99), assigning a weight of +3 to the cluster predicted ␩ 2 F(1, 115) = 55.51, p < .001, p = .33. Also as to be high on the given emotion (e.g. low- predicted by Hypothesis 3, there was no main competition/low-status for pity) and weights of effect of competition on competence (F = 2.08), –1 to each of the clusters predicted to be low on and no signifi cant competition × status interac- that emotion. Table 2 presents the mean emotion (F < 1.0). tion ratings by condition. As predicted, members of low-competition, high-status groups were Differentiation of emotional prejudices rated as eliciting signifi cantly more admiration Next, we tested Hypotheses 4–8, that combina- and pride (M = 0.52) than members of all other tions of competition and status are differentially groups (M = –0.18), t(116) = 3.59, p < .001. associated with unique emotional reactions. We Furthermore, members of high-status, high- fi rst standardized the means of the emotions competition groups elicited more envy (M = 0.65) to account for main effect differences in the over- than members of other groups (M = –0.24), all endorsement rate of one emotion over an- t(116) = 5.44, p < .001. Likewise, members of other (e.g. the average admiration rating across low-competition, low-status groups elicited more groups was 2.88 on a 7-point scale, whereas the pity and sympathy (M = 0.65) than other group average contempt rating was 4.17. See Table 2). members (M = –0.20), t(116) = 4.60, p < .001. We then submitted the emotions ratings to a However, contrary to predictions, members 2 (Competition: high, low) × 2 (Status: high, low) of high-competition, low-status groups did × 4 (Emotion: admiration, contempt, envy, pity) not differ from the average of the other three mixed ANOVA with repeated measures on the groups in ratings of contempt and disgust Emotion factor, which revealed the predicted (Ms = –0.06 and 0.02, respectively, t < –1.0). Competition × Status × Emotion interaction In fact, group members were granted less ␩ 2 F(3, 345) = 4.59, p < .01, p = .03. contempt than members of high-competition/ We conducted separate contrast analyses for high-status groups and low-competition/low- each emotion to test more focused predictions, status groups. The only group members who

Table 1. Mean competence and warmth ratings by condition

Status Competition Competence Warmth High High 4.58 (1.39) 3.47 (1.26) High Low 4.83 (1.35) 4.13 (1.44) Low High 2.80 (1.03) 3.35 (0.95) Low Low 3.21 (1.21) 3.84 (1.20) Note: Bolded means signiﬁ cantly differ from other column means at p < .05. Standard deviations are in parentheses.

Table 2. Standardized emotion ratings by condition

Status Competition Admiration ␺ Envy ␺ Pity ␺ Contempt ␺ High Low 0.52 (.82) +3 0.55 (.85) –1 –0.43 (.78) –1 –0.32 (.91) –1 High High 0.16 (.91) –1 0.65 (.94) +3 –0.57 (.87) –1 0.13 (.90) –1 Low Low –0.57 (.82) –1 –0.44 (.71) –1 0.65 (.91) +3 0.25 (1.15) –1 Low High –0.13 (1.13) –1 –0.82 (.55) –1 0.40 (.89) –1 –0.06 (.99) +3 Unstandardized Mean 2.88 (1.15) 3.43 (1.85) 3.54 (1.40) 4.16 (1.20) Note: Cell values represent mean z-scores within emotion. Bolded means signiﬁ cantly differ from other column means at p < .001, according to contrast weights associated with each analysis. The ␺ column refers to contrast weights used in analyses. Standard deviations are in parentheses.

152

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115252 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Caprariello et al. causal test of the stereotype content model

elicited lesser contempt ratings were members these groups seem to be less-than-human, in that of low-competition/high-status groups. Post hoc sense. But these groups do activate the insula, contrasts revealed that the difference between an area reliably implicated in disgust (Harris & members of this group and the other groups was Fiske, 2006). Third, actual ratings of immigrant signifi cant, t(115) = –2.02, p < .05. That is, while groups seen to be low-status and exploitative our prediction regarding who would elicit the (e.g. undocumented migrants) do elicit low- most contempt was not supported, we did fi nd warmth and low-competence attributions, as that members of low-competition/high-status here (Lee & Fiske, 2006). And anecdotal evi- groups (typically the ingroup or societal default dence would suggest that the American public group) elicited the least. currently views such migrants with much contempt and disgust, seeking to exclude them from Discussion the country. The present study may suffer the external val- Results soundly support the SCM framework, idity shortcomings inherent to most scenario shedding light specifi cally on the causal relation- studies. Using hypothetical scenarios allowed ship of society-level structures among groups us to fully cross the structural variables— in determining society-level responses to out- competition and status—and to measure two groups. Further, outgroup response covers both important outcome variables—stereotypes and cognitive and affective variables, all rooted from emotions—without any confounding variables the common catalyst of perceived competition due to group histories. Hypothetical scenarios and status. These seven out of eight effects are also helped us avoid alternative explanations demonstrated at the societal level, as partici- due to expectancies or social desirability. For pants were asked for their perceptions of how example, by using an unfamiliar outgroup, we society in general will respond to the unfamiliar are able to impose theoretical characteristics group. One prediction was not fully supported. on the group, independent of participants’ Participants were mostly neutral on contempt preconceptions of actual groups. Furthermore, and disgust (hovering around the midpoint, participants need neither avoid appearing pre- more than for the other emotions), apparently judiced toward a hypothetical scenario (Kunda & reluctant to report contempt and disgust for any Spencer, 2003) nor differentiate between stereo- of the immigrant groups; perhaps this reluctance types and personal beliefs (Devine & Elliot, is not surprising, given the extreme negativity 1995). Instead we demonstrate causal effects of these emotions. Indeed, all the emotional at the societal level, by asking participants how prejudice ratings fell in the bottom half of the they feel other group members will respond in scale (in the ‘unlikely’ range), suggesting a reluc- general. For similar reasons, other intergroup tance to commit to any of them. Most of the researchers have also used scenario designs in effects resulted from less reluctance for the pre- their experiments (e.g. Alexander et al., 1999; dicted emotion-group pairing than for the others. Castano & Giner-Sorolla, 2006). It is notable that In the case of contempt and disgust, the null Wallonians, the French-speaking Belgians, are result does not undermine the model’s predic- not vivid in the minds of geography-deprived tions because of three kinds of prior data: First, American undergraduates. Thus, despite its the student samples and national representative potential weaknesses, the hypothetical scenario sample survey (Cuddy et al., 2007; Fiske et al., served its purpose, to show how social structure 2002) showed clear reports of disgust/contempt can in this circumstance produce results con- toward low-warmth, low-competence groups, also sistent with the causal predictions of the SCM. seen as low-status and high-competition (or at Out of eight predicted SCM structural effects, least exploitation). Second, neuro-imaging data seven were fully supported here. These experi- show that these groups uniquely fail to elicit mental data support a causal interpretation of activation in the medial prefrontal cortex, an correlational results from prior student and area that reliably responds to social stimuli. So non-student samples in over a dozen countries,

153

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115353 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Group Processes & Intergroup Relations 12(2)

for varying aspects of the structure-to-bias (stereo- Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2007). The types, prejudice) predictions (Cuddy et al., in BIAS map: Behaviors from intergroup affect press; Fiske et al., 2002). What is unique here and stereotypes. Journal of Personality and Social is the simultaneous experimental manipula- Psychology, 92, 631–648. Cuddy, A. J. C., Fiske, S. T. & Glick, P. (2008). tion of both competition and status, for a general- Warmth and competence as universal ized group of immigrants, rather than a specifi c dimensions of social perception: The social group (such as mothers or older adults), Stereotype Content Model and the BIAS Map. at the societal level. This provides a crucial piece In M. P. Zanna (Ed.), Advances in experimental of the SCM puzzle. social psychology (vol. 40, pp. 61–149). San Diego: Academic Press. Cuddy, A. J. C., Fiske, S. T., Kwan, V., Glick, Note P., Demoulin, S., Leyens, J-Ph., Bond, M. H., et al. (in press). Is the Stereotype Content Model 1. The alpha coeffi cients for admiration and culture-bound? Cross-cultural universalities and contempt are relatively low, according to differences of stereotyping principles. British Rosenthal and Rosnow (1991), but not Journal of Social Psychology. exceptionally so, given that both variables are Devine, P., & Elliot, A. (1995). Are racial composed of only two items (Cortina, 1993). In stereotypes really fading? The Princeton trilogy both cases, the items making up the variables revisited. Personality and Social Psychology Bulletin, correlate more strongly with each other than 21, 1139–1150. with any other emotion items. In addition, Dijker, A. J., Koomen, W., van den Heuvel, H., & decomposing admiration and contempt and Frijda, N. H. (1996). Perceived antecedents of re-running analyses with single items produces emotional reactions in inter-ethnic relations. identical results. British Journal of Social Psychology, 35, 313–329. Eckes, T. (2002). Paternalistic and envious gender References stereotypes: Testing predictions from the stereotype content model. Sex Roles, 47, 99–114. Alexander, M. G., Brewer, M. B., & Herrmann, R. K. Falomir-Pichastor, J. M., Munoz-Rojas, D., (1999). Images and affect: A functional analysis Invernizzi, F., & Mugny, G. (2004). Perceived of out-group stereotypes. Journal of Personality in-group threat as a factor moderating the and Social Psychology, 77, 78–93. infl uence of in-group norms on discrimination Allport, G. (1954). The nature of prejudice. Reading, against foreigners. European Journal of Social MA: Addison-Wesley. Psychology, 34, 135–153. Aquino, K., Stewart, M., & Reed, A. (2005). How Fiske, S. T. (1992). Thinking is for doing: Portraits social dominance orientation and job status of social cognition from Daguerreotype to infl uence perceptions of African-American laserphoto. Journal of Personality and Social affi rmative action benefi ciaries. Personnel Psychology, 63, 877–889. Psychology, 58, 703–744. Fiske, S. T. (1993). Controlling other people: Betancor, V., Rodriguez, A., Rodriguez, R., The impact of power on stereotyping. American Leyens, J.P., & Quiles, M.N. (2005). The Psychologist, 48, 621–628. effect of status on sociability and competence Fiske, S. T., Cuddy, A. J. C., & Glick, P. (2007). stereotypical dimensions. Psicothema, 17, 297–302. Universal dimensions of social perception: Castano, E., & Giner-Sorolla, R. (2006). Not quite Warmth and competence. Trends in Cognitive human: Infra-humanization as a response to Science, 11, 77–83. collective responsibility for intergroup killing. Fiske, S. T., Cuddy, A. J. C., Glick, P., & Xu, J. (2002). Journal of Personality and Social Psychology, 90, A model of (often mixed) stereotype content: 804–819. Competence and warmth respectively follow Cortina, J. M. (1993). What is coeffi cient alpha? from perceived status and competition. Journal of An examination of theory and applications. Personality and Social Psychology, 82, 878–902. Journal of Applied Psychology, 78, 98–104. Fiske, S. T., Xu, J., Cuddy, A. J. C., & Glick, P. (1999). Cuddy, A. J. C., Fiske, S. T., & Glick, P. (2004). (Dis)respecting versus (dis)liking: Status and When professionals become mothers, warmth interdependence predict ambivalent stereotypes doesn’t cut the ice. Journal of Social Issues, 60, of competence and warmth. Journal of Social 701–718. Issues, 55, 473–489.

154

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115454 112/17/20082/17/2008 3:31:473:31:47 PMPM PProcessrocess BlackBlack Caprariello et al. causal test of the stereotype content model

Glick, P., Larsen, S., Johnson, C., & Branstiter, H. Russell, A. M., & Fiske, S. T. (in press). It’s all (2005). Evaluations of sexy women in low- and relative: Social position and interpersonal high-status jobs. Psychology of Women Quarterly, 29, perception. European Journal of Social Psychology. 389–395. Sherif, M., Harvey, O.J., White, B.J., Hood, W.R., & Harris, L., & Fiske, S.T. (2006). Dehumanizing Sherif, C.W. (1961). Intergroup conﬂ ict and co- the lowest of the low: Neuro-imaging responses operation: The robber’s cave experiment. Norman: to extreme outgroups. Psychological Science, 17, University of Oklahoma Press. 847–853. Smith, R. H. (2000). Assimilative and contrastive Kunda, Z., & Spencer, S. (2003). When do emotional reactions to upward and downward stereotypes come to mind and when do they social comparisons. In J. Suls, & L. Wheeler color judgment? A goal-based theoretical (Eds.), Handbook of social comparison: Theory framework for stereotype activation and and research (pp. 173–200). New York: Kluwer application. Psychological Bulletin, 129, 522–544. Academic/Plenum Publishers. Lee, T. L., & Fiske, S. T. (2006). Not an outgroup, Smith, R. H., Parrott, W. G., Ozer, D., & Moniz, A. but not yet an ingroup: Immigrants in the (1994). Subjective injustice and inferiority as stereotype content model. International Journal of predictors of hostile and depressive feelings of Intercultural Relations, 30, 751–768. envy. Personality and Social Psychology Bulletin, 20, Mackie, D. M., Devos, T., & Smith, E. R. (2000). 705–711. Intergroup emotions: Explaining offensive Stephan, W., Renfro, C. L., Esses, V., Stephan, C., & action tendencies in an intergroup context. Martin, T. (2005). The effects of feeling threatened Journal of Personality and Social Psychology, 79, on attitudes towards immigrants. International 602–616. Journal of Intercultural Relations, 29, 1–19. Oldmeadow, J., & Fiske, S. T. (2007). System-justifying ideologies moderate Paper received 9 January 2008; revised version accepted status = competence stereotypes: Roles for 29 September 2008. belief in a just world and social dominance orientation. European Journal of Social Psychology, 37, 1135–1148. Biographical notes Ridgeway, C. L., & Correll, S. J. (2004). peter a. caprariello, Motherhood as a status characteristic. Journal of Social Issues, 60, 683–700. amy j. c. cuddy, Harvard University Rosenthal, R., & Rosnow, R. (1991). Essentials of behavioral research: Methods and data analysis susan t. ﬁ ske, Princeton University (2nd edn.). Boston: McGraw-Hill. Ruscher, J., Fiske, S. T., Miki, H., & Van Manen, S. (1991). Individuating processes in competition: Interpersonal versus intergroup. Personality and Social Psychology Bulletin, 17, 595–605.

155

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115555 112/17/20082/17/2008 3:31:483:31:48 PMPM PProcessrocess BlackBlack Group Processes & Intergroup Relations 12(2)

1147-155_GPI_101053.indd47-155_GPI_101053.indd 115656 112/17/20082/17/2008 3:31:483:31:48 PMPM PProcessrocess BlackBlack