reflection

http://gpi.sagepub.com/ Relations

Group Processes & Intergroup

http://gpi.sagepub.com/content/14/3/311 The online version of this article can be found at:

  DOI: 10.1177/1368430210392932

2011 14: 311 originally published online 17 March 2011Group Processes Intergroup Relations Jared B. Kenworthy, Melisa A. Barden, Steven Diamond and Alejandro del Carmen

task no shoot decision−Ingroup identification as a moderator of racial bias in a shoot

   

Published by:

http://www.sagepublications.com

can be found at:Group Processes & Intergroup RelationsAdditional services and information for      

  http://gpi.sagepub.com/cgi/alertsEmail Alerts:

 

http://gpi.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

http://gpi.sagepub.com/content/14/3/311.refs.htmlCitations:  

What is This?  

- Mar 17, 2011 OnlineFirst Version of Record  

- May 3, 2011Version of Record >>

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

Group Processes & Intergroup Relations 14(3) 311 –318

© The Author(s) 2011 Reprints and permission: sagepub.

co.uk/journalsPermissions.nav DOI: 10.1177/1368430210392932

gpir.sagepub.com

Article

G P I R

Group Processes & Intergroup Relations

At least since the 1999 Bronx shooting (Fritsch, 2000) of unarmed Amadou Diallo by New York City police officers, social psychologists have had an interest in the dynamics of decisions to shoot or not shoot potentially dangerous suspects. Of key theoretical interest in subsequent research has been the confluence of split-second decision- making and the effects of racial stereotypes on response bias against Black targets. Our aim in this article is to replicate and extend prior research on this topic by testing the moderation of the Black– White decision bias by ingroup identification.

Extant research literature illustrates a pro- nounced bias in the perception of weapons when participants complete a simulation of the

decisions a police officer may be forced to make. In the standard experimental paradigm, partici- pants are presented with images of Black and White targets who are holding either a weapon (e.g., gun or knife) or a benign object (e.g., cell phone, soda can). Their task is to identify, as quickly and as accurately as possible, which type of object the target is holding. They “shoot” by

Article

392932 XXX10.1177/1368430210392932Kenworthy et al.Group Processes & Intergroup Relations

1University of Texas at Arlington 2Walsh University

Corresponding author: Jared B. Kenworthy, Department of Psychology, University of Texas at Arlington, Texas, USA. Email: [email protected]

Ingroup identification as a moderator of racial bias in a shoot–no shoot decision task

Jared B. Kenworthy,1 Melisa A. Barden,2 Steven Diamond1 and Alejandro del Carmen1

Abstract In this experiment, group identification is examined as a potential moderator of the documented racial bias in a shoot–no shoot paradigm. Target race (Black or White) was manipulated within-subjects, and (White) participants’ strength of racial ingroup identification was employed as a continuous predictor. Using signal detection analyses, it was found that the criterion to shoot Black targets decreased as an increasing function of racial ingroup identification. This relationship held even after controlling for gender, social dominance orientation, a measure of cultural stereotyping, and the amount of contact with Blacks. Theoretical implications are discussed.

Keywords ingroup identification, racial bias, shoot–no shoot

Paper received 10 April 2009; revised verion accepted 05 November 2010.

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

312 Group Processes & Intergroup Relations 14(3)

pressing one button, and they indicate a “don’t shoot” response with another (Correll, Park, Judd, & Wittenbrink, 2002, 2007; Correll et al., 2007). One way to measure the racial shooter bias is by testing for a statistical interaction between the ethnicity of the target and the type of object held by the target. Participants tend to be more likely to shoot an unarmed Black target than an unarmed White target.

Invoking an automatic-processes explanation, Correll et al. (2002) noted that the racial shooter bias may be caused by similar mechanisms to a Stroop Task. In a Stroop Task, participants are instructed to ignore automatic tasks (i.e., reading the word ‘RED’) and focus instead on more effortful tasks (i.e., the color of ink in which the word is written) which are often incongruent. Likewise, for the shoot–no shoot task, partici- pants must identify the type of object in a target’s hand (i.e., an effortful task) in the face of auto- matically processed information (viz., target race) which is linked to stereotypical inferences.

Supporting the automatic link to ethnic stere- otypes, Correll et al. (2007) utilized the shoot–no shoot paradigm to investigate whether the acces- sibility of the stereotype—Blacks as dangerous— would further exacerbate the racial shooter bias. In one study, participants were primed with news- paper articles featuring White or Black criminals. Bias to shoot the Black targets was increased after the participants were primed with stories about Black criminals. In the two other studies reported, the frequency of armed Black and White targets was manipulated within the experimental ses- sions. As expected, a greater frequency of Black armed targets made the stereotype more accessi- ble and exacerbated the racial shooter bias.

Other factors that may potentially play a role in producing the shooter bias have also been explored. The perception of cultural stereotypes was examined as a moderator of the shooter bias (Correll, Urland, & Ito, 2006). There was a pronounced shooter bias for the participants who indicated that American society is more likely to link violence to Blacks than Whites. Correll et al. (2002) examined a variety of explicit prejudice scales and found no significant

correlations with the Shooter Bias. One variable that was found to influence the Shooter Bias was the amount of contact the participant had with African-Americans (Correll et al., 2002). Surprisingly, however, the effect was in the opposite direction than what was expected. The more contact the participants reported, the more pronounced was the exhibited shooter bias. Peruche and Plant (2006) expanded this concept and differentiated between positive and negative contact. They found that officers who had reported positive personal contact with Black people were better able to eliminate the shooter bias with simulation training.

As illustrated above, a variety of moderators have been examined in relation to the shooter bias. One important factor that has been ignored is ingroup identification. Social Identity Theory (Tajfel & Turner, 1986) and related theoretical perspectives (e.g., Self-Categorization Theory; Turner, Hogg, Oakes, Reicher, & Wetherell, 1987) argue for the role of ingroup identification in many group and intergroup processes. These theoretical perspectives posit that when we cate- gorize ourselves as group members, our thoughts and behaviors become more depersonalized and prototypically ingroup-like.

One’s level of ingroup identification indicates the degree to which group membership is subjec- tively important and salient, thereby making group-favoring responses more likely. Ingroup identification has been shown to predict or mod- erate many group and intergroup phenomena. Compared to low ingroup identifiers, those who identify more strongly report more positive atti- tudes toward the ingroup (e.g., Cairns, Kenworthy, Campbell, & Hewstone, 2006; Jetten, Spears, & Manstead, 1997), more negative attitudes toward the outgroup (e.g., Jetten et al., 1997; Lindeman, 1997), see themselves as more prototypical of the ingroup (e.g., Barreto & Ellemers, 2000; Pickett, Bonner, & Coleman, 2002), punish ingroup devi- ants more severely (e.g., Coull, Yzerbyt, Castano, Paladino, & Leemans, 2001; Hutchison & Abrams, 2003), and perceive more homogeneity within both the ingroup and the outgroup (e.g., Doosje, Ellemers, & Spears, 1995; Ellemers,

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

Kenworthy et al. 313

Spears, & Doosje, 1997; Wann & Branscombe, 1995). In the present experiment, we examine whether ingroup identification can impact rela- tively uncontrolled, automatic intergroup proc- esses in the same way that it does more conscious, self-report and behavioral responses.

For the present research, we employed the shoot–no shoot paradigm, as described above, and recruited a sample of White participants. Racial identification was operationalized as the degree to which participants identify with their own race (viz., White). We submitted the responses to analyses derived from signal detec- tion theory. Specifically, of interest was the degree to which the criterion (see note 1) for shooting Black versus White targets would shift as a function of racial ingroup identification. Specifically, we hypothesized that as racial ingroup identification increased, the criterion values for shooting Black (but not White) targets would decrease.

To conduct a conservative test of our key hypothesis, we included a number of control var- iables. These included gender, social dominance orientation (Pratto, Sidanius, Stallworth, & Malle, 1994), cultural stereotypes of Blacks as aggressive (Correll et al., 2002; Correll et al., 2007), and amount of contact with Blacks (Correll et al., 2002). Stereotypes and contact have been found to correlate with some measures of the shooter bias in previous research (see Correll et al., 2002), whereas gender and social dominance orientation were included to add to the discriminant validity of the measures and findings.

Method

Participants, design, and procedure Seventy-four White undergraduates (54 female, 20 male) from a large, public university in the southern United States participated in exchange for course credit. The design was a 2 (Target Race: Black vs. White) x 2 (Object Type: gun vs. no gun) within-subjects factorial. A questionnaire was administered to assess participant demo- graphics and the following constructs. Racial

ingroup identification was employed as a continu- ous predictor in the analyses.

Racial ingroup identification To measure racial ingroup identification (with respect to being White), the following nine items were assessed on 7-point scales (1 = I strongly disagree, 7 = I strongly agree): “Being a member of my race/ ethnicity is an important reflection of who I am”, “In general, being a member of my race/ethnicity is an important part of my self-image”, “I see myself as a member of my race/ethnicity”, “Being a member of my race/ethnicity is central to my sense of who I am”, “I value being a member of my race/ethnicity”, “Overall, being a member of my race/ethnicity has very little to do with how I feel about myself ” (reversed), “I feel proud to be a member of my race/ethnicity”, “Being a member of my race/ethnicity is unimportant to my sense of what kind of person I am” (reversed), and “I feel strong ties to other people of my race/ ethnicity”. The reliability of this scale was good (Cronbach’s α = .94), and the items were averaged (after appropriate reversals) into an index of ingroup identification for which higher scores indicate stronger ingroup identification.

Social dominance orientation We measured social dominance orientation, using the 16-item form, as found in Pratto et al. (1994). The reliability of this scale was good in this study (α = .90).

Cultural stereotypes of Blacks as aggressive As in Correll et al. (2002), we measured partici- pants’ perceived degree of American cultural stereotypes of Blacks as violent. The five items were measured by having participants mark a blank line (approximately 3.5 inches) between the endpoints of each scale, and then by measuring the location of the mark. The endpoints were not at all violent–very violent, not at all harmless– very harmless (reversed), not at all aggressive–very aggressive, not at all dangerous–very dangerous, and not at all hostile–very hostile. After the appropriate reversal, this scale exhibited good reliability (α = .89).

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

314 Group Processes & Intergroup Relations 14(3)

Table 1. Intercorrelations among key variables

1 2 3 4 5 6 7

1. Criterion (c) – Black – 2. Criterion (c) – White .07 – 3. Racial Ingroup Identification −.31** −.06 – 4. Gender −.25* −.06 −.09 – 5. Cultural Stereotype of Blacks .06 .15 .37** −.38** – 6. Quantity of Contact with Blacks −.03 .23* −.07 .02 −.08 – 7. Social Dominance Orientation −.16 .02 .21^ .19 .09 −.06 – 8. Difference – White (c) minus Black (c) −.66** .57** .20^ .08 −.05 .18 .10

Notes: N = 74. Gender is coded as 1 = female, 2 = male. ^p < .10, * p < .05, ** p < .01.

Amount of contact with Blacks Participants were asked to indicate the amount of contact that they have with African-Americans, with the following question stem (see Tausch, Tam, Hewstone, Kenworthy, & Cairns, 2007): “Think- ing of social contacts – whether at home, or at work, or somewhere else – how much contact do you have with African Americans…?” This stem was followed by two separate items, measured on 4-point scales (1 = a great deal, 2 = some, 3 = a little, 4 = none at all): “just chatting to people” and “over all social situations”. These items exhibited good reliability (α = .75). These items were reversed, so that a higher score on this index indicated more contact.

Shoot–no shoot task For the shoot–no shoot task, we employed a computer methodology sim- ilar to the Correll et al. (2002) paradigm, but used E-prime as the software platform. Several differ- ent Black and White actors were used, holding one of two guns (silver or black), or one of two benign objects (soda can, cell phone). Control scenes had no people in them, and instructions dictated no response at all for control scenes. There was an equal number of each type of tar- get (Black vs. White), counterbalanced with an equal number of object type (gun vs. no gun). All images were presented in a random order, and participants had an upper time limit of 800 milli- seconds to respond (see Correll et al., 2002; Correll et al., 2007).

Results

Signal detection analyses Responses to the targets were submitted to signal detection analysis. Of primary concern was the criterion score, c, or the tendency for a shoot ver- sus a no-shoot response.1 We computed a c value for Black and White targets separately for each participant. We then conducted regression analy- ses in which the criterion scores were regressed onto the control variables (step 1; including gen- der, coded as 1 = female, 2 = male), and then onto racial ingroup identification (step 2). The intercorrelations among key study variables are presented in Table 1.

Black targets For these scores, a regression model was conducted in which control variables were entered in step 1. These variables included the c score for White targets, gender, social domi- nance orientation, cultural stereotypes of Blacks as aggressive, and quantity of contact with Blacks. In step 2, racial ingroup identification was entered as a continuous predictor. As shown in Table 2, only gender approached significance in step 1. How- ever, racial ingroup identification significantly pre- dicted criterion scores for Black targets in step 2, over and above the effects of the control variables. This means that as racial ingroup identification increased, participants displayed a lower criterion, or a stronger bias to shoot Black targets.2

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

Kenworthy et al. 315

In a related set of analyses, we substituted false alarm rates for criterion scores in regression mod- els to those described here above. That is, with the same control variables, we regressed false alarm rates for Black targets onto all control variables and hit rates for Black targets in step one, then onto racial ingroup identification in step 2. This model yielded parallel results. Specifically, hit rates for Black targets negatively predicted false alarm rates for Black targets (β = −.35, p < .01), and gender was again a marginal predictor (β = .20, p = .09). Over and above the contribution of these variables, racial ingroup identification significantly predicted false alarm rates for Black targets (β = .34, p < .01), indicating that as ingroup identifica- tion increased, so too did false alarm rates. A reversed model, wherein hit rates for Black targets were regressed onto all variables, including false alarm rates as a control variable, yielded a weaker pattern. Specifically, only false alarm rates were a significant predictor of hit rates (β = −.39, p < .01); racial ingroup identification became a mar- ginal predictor (β = .25, p = .06). Together, these analyses indicate that although racial ingroup identification was associated with both hit rates and false alarms for Black targets, the stronger association was between ingroup identification and false alarm rates. Thus, as ingroup identifica- tion increased, so too did the tendency to shoot unarmed Black targets.

White targets A similar set of regressions was conducted to predict criterion scores for White targets. This time, however, the c score for Black targets was entered as a control variable in step 1. Neither step 1 nor step 2 of this analysis yielded a significant overall model. However, quantity of contact with Blacks was a significant predictor (β = .25, p = .03) of the criterion score for White targets. This means that, consistent with the zero- order correlations in Table 1, as contact increased, so too did the criterion for shooting White targets.

Additional analyses As can be seen in Table 1, we also computed a difference score between the criterion values for White and Black targets. This was done by sub- tracting the Black c score from the White c score for each participant. A higher score on this meas- ure indicates a greater disparity in responses between the White and Black targets. As shown, a greater difference score was negatively associated with the Black c score and positively associated with the White c score. It was also marginally pos- itively associated with racial ingroup identifica- tion. This indicates that as racial identification increased, so too did the difference between White and Black c scores, indirectly supporting the main hypothesis for this study.

Table 2. Regression results for predicting criterion scores for Black targets

B (SE) β t-test ΔR2

Model 1 Criterion Score – White Targets .07 (.11) .08 .61 Gender −.31 (.17) −.23 −1.78^ Social Dominance Orientation −.07 (.07) −.11 −.94 Cultural Stereotype of Blacks −.004 (.02) −.03 −.23 Quantity of Contact with Blacks −.05 (.12) −.05 −.44 Model Summary: F(5, 68) = 1.17, p = .33 .08 Model 2 Racial Ingroup Identification −.15 (.05) −.36 −2.92** Model Summary: F(6, 67) = 2.51, p = .03 .10**

Notes: Standard errors are in parentheses. Model 2 contains all control variables from Model 1, in addition to Racial Ingroup Identification. Gender is coded as 1 = female, 2 = male. ^p = .08, ** p < .01.

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

316 Group Processes & Intergroup Relations 14(3)

In a set of regressions analogous to those pre- sented above (i.e., including all control variables), racial ingroup identification emerged as a mar- ginal predictor (β = .24, p = .06) of this difference score index, indicating that as ingroup identifica- tion increased, the discrepancy between responses to White and Black targets tended to increase.

Discussion Supporting our primary hypothesis, the decision to shoot Black, but not White, targets became more likely as racial ingroup identification increased. Note that this finding derives from the analyses of the criterion scores, which consist of both hit rates and false alarm rates. However, in a supplementary analysis, we found that the false alarm rates (controlling for hit rates) were more strongly predicted by racial ingroup identification than were hit rates (controlling for false alarm rates). These results were found even after con- trolling for gender, social dominance orientation, stereotyping, and contact. For White targets, the only significant predictor of the criterion score was the contact measure. Specifically, as self- reported quantity of contact with African- Americans increased, the criterion scores for White targets increased. On the surface, this find- ing appears to be a failure to replicate previous work (see Correll et al., 2002; Correll et al., 2007), which has shown the cultural stereotype of Blacks as aggressive to be associated with the shooter bias. Upon closer examination, however, this is not the case. Their bias measure was an index of ingroup-minus-outgroup response times to certain trials, whereas our dependent measure was the criterion value for making a shoot decision when presented with Black tar- gets. Thus, no direct comparison can be made.3 A similar argument can be made for the failure to replicate the association between contact with African-Americans (see Correll et al., 2002) and the criterion score.

These findings suggest that for low racial ingroup identifiers, race (in this case, being White) is not a chronically salient or important social cat- egorization dimension. For those who identified

more strongly, however, it seems that race has impact even at the automatic level of processing. Participants were given an upper limit of less than one second to respond, and the shooting decisions made are consistent with the notion that ingroup identification plays a role even in relatively uncontrolled processing. To our knowl- edge, this is the first empirical demonstration that racial ingroup identification may be an indicator of chronic links to outgroup stereotype activa- tion, or of uncontrolled ingroup-favoring behav- ioral responses.

One recent study (Crisp, Stone, & Hall, 2006, Experiment 3), however, did show that high and low ingroup identifiers (with respect to British nationality) display different patterns of bias on an implicit attitudes measure (viz., IAT), and that high identifiers show an increase in both explicit and implicit bias in response to superordinate recategorization between the ingroup and the outgroup (in their case, the French). We point out also that across their studies, Crisp et al. found consistently that the locus of the effect was vari- ation in responses to the outgroup. In other words, shifts in bias between the ingroup and outgroup were found to be a function of decreas- ing positivity toward the outgroup for high (but not low) identifiers. Such a pattern parallels our findings with respect to differential biases in false alarm error rates. Specifically, as described above, ingroup identification predicted criterion scores for outgroup targets, but not to ingroup targets. Such findings are consistent with a social identity perspective, according to which concerns for intergroup differentiation increase as a function of ingroup identification.

Other research paradigms might also examine the potentially moderating role of racial ingroup identification. For example, the paradigm employed by Payne (e.g., 2001, 2006; see also Govorun & Payne, 2006) investigates the role of racial stereotypes in a weapons identification task. In this research, participants are presented with computer images of Black or White faces as primes. These primes are subsequently followed by an image of a gun or a tool (e.g., pliers, channel locks) and a visual mask. Participants must then

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

Kenworthy et al. 317

indicate whether they saw a gun or a tool. Following Black primes, participants consistently misidentify the tool as a gun at a higher rate than following White primes. If our present findings are generalizable, racial ingroup identification may moderate that effect as well, such that the rate of tool-as-gun errors (following Black, but not White, primes) should be most pronounced for high ingroup identifiers, and weaker or absent for low identifiers.

Future research might further explore the effect that we have reported here by experimen- tally manipulating ingroup identification via ran- dom assignment. A manipulation of ingroup identification (e.g., Leonardelli & Brewer, 2001) may evoke automatic tendencies for ingroup- favoring responses, even at the implicit level. Relatedly, discovering the sources and origins of high versus low identification should be attempted so that interventions may be designed and imple- mented to reduce or eliminate the pervasive effects of racial stereotyping and profiling in social life.

Notes 1. This criterion score is computed as c = −0.5*(zFA +

zH). FA = proportion of false alarms (relative to correct rejections), and H = proportion of hits (relative to misses). The z indicates z-score trans- formations of these proportions.

2. Similar analyses using the d ’ scores for Black tar- gets, which estimate the sensitivity to armed versus unarmed targets [d ’ = z(FA) – z(H); FA = propor- tion of false alarms (relative to correct rejections), and H = proportion of hits (relative to misses)], yielded no significant results apart from the expected association between d ’ for Black targets and d ’ for White targets (β = .55, p < .001). Ingroup identification did not significantly predict d ’ for Black targets (β = −.18, p = .12).

3. To our knowledge, the only previously published test of the relationship between cultural stereo- types and the criterion score occurs in Correll et al. (2002, Study 3). Although this relationship was sig- nificant, this test controlled for personal prejudice as well as the motivation to control prejudice (Dunton & Fazio, 1997). This latter construct was not measured here, and so a direct comparison is again precluded.

References

Barreto, M., & Ellemers, N. (2000). You can’t always do what you want: Social identity and self- presentational determinants of the choice to work for a low-status group. Personality and Social Psychology Bulletin, 26, 891–906.

Cairns, E., Kenworthy, J. B., Campbell, A., & Hewstone, M. (2006). The role of in-group identification, reli- gious group membership, and intergroup conflict in moderating in-group and out-group affect. British Journal of Social Psychology, 45, 701–716.

Correll, J., Park, B., Judd, C., & Wittenbrink, B. (2002). The police officer’s dilemma: Using ethnicity to disambiguate potentially threatening individu- als. Journal of Personality and Social Psychology, 83, 1314–1329.

Correll, J., Park, B., Judd, C., & Wittenbrink, B. (2007). The influence of stereotypes on decisions to shoot. European Journal of Social Psychology, 37, 102–1117.

Correll, J., Park, P., Judd, C., Wittenbrink, B., Sadler, M., & Keesee, T. (2007). Across the thin blue line: Police officers and racial bias in the decision to shoot. Journal of Personality and Social Psychology, 92, 1006–1023.

Correll, J., Urland, G., & Ito, T. (2006). Event-related potentials and the decision to shoot: The role of threat perception and cognitive control. Journal of Experimental Social Psychology, 42, 120–128.

Coull, A., Yzerbyt, V., Castano, E., Paladino, M., & Leemans, V. (2001). Protecting the ingroup: Motivated allocation of cognitive resources in the presence of threatening ingroup members. Group Processes & Intergroup Relations, 4, 327–339.

Crisp, R. J., Stone, C. H., & Hall, N. R. (2006). Recategorization and subgroup identification: Predicting and preventing threats from common ingroups. Personality and Social Psychology Bulletin, 32, 230–243.

Doosje, B., Ellemers, N., & Spears, R. (1995). Perceived intragroup variability as a function of group status and identification. Journal of Experimental Social Psychology, 31, 410–436.

Dunton, B. C., & Fazio, R. H. (1997). An individual difference measure of motivation to control prej- udiced reactions. Personality and Social Psychology Bulletin, 23, 316–326.

Ellemers, N., Spears, R., & Doosje, B. (1997). Sticking together or falling apart: In-group identification as a psychological determinant of group commitment versus individual mobility. Journal of Personality and Social Psychology, 72, 617–626.

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from

318 Group Processes & Intergroup Relations 14(3)

Fritsch, J. (2000). The Diallo verdict: The overview; 4 officers in Diallo shooting are acquitted of all charges. The New York Times Feb. 6. Retrieved Jan 1, 2009, from http://query.nytimes.com/gst/fullpage. html?res=9C01E7D61639F935A15751C0A9669 C8B63

Govorun, O., & Payne, K. (2006). Ego depletion and prejudice: separating automatic and controlled components. Social Cognition, 24, 111–136.

Hutchison, P., & Abrams, D. (2003). Ingroup identi- fication moderates stereotype change in reaction to ingroup deviance. European Journal of Social Psychology, 33, 497–506.

Jetten, J., Spears, R., & Manstead, A. (1997). Distinctiveness threat and prototypicality: Combined effects on intergroup discrimination and collective self-esteem. European Journal of Social Psychology, 27, 635–657.

Leonardelli, G., & Brewer, M. (2001). Minority and majority discrimination: When and why. Journal of Experimental Social Psychology, 37, 468–485.

Lindeman, M. (1997). Ingroup bias, self-enhancement and group identification. European Journal of Social Psychology, 27, 337–355.

Payne, B. (2001). Prejudice and perception: the role of automatic and controlled processes in misper- ceiving a weapon. Journal of Personality and Social Psychology, 81, 181–192.

Payne, K. (2006). Weapon bias: split-second decisions and unintended stereotyping. Current Directions in Psychological Science, 15, 287–291.

Peruche, B. M., & Plant, E. A. (2006). The correlates of law enforcement officers’ automatic and controlled race-based responses to criminal suspects. Basic and Applied Social Psychology, 28, 193–199.

Pickett, C., Bonner, B., & Coleman, J. (2002). Motivated self-stereotyping: Heightened assimilation and dif- ferentiation needs result in increased levels of positive and negative self-stereotyping. Journal of Personality and Social Psychology, 82, 543–562.

Pratto, F., Sidanius, J., Stallworth, L., & Malle, B. (1994). Social dominance orientation: A person- ality variable predicting social and political atti- tudes. Journal of Personality and Social Psychology, 67, 741–763.

Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behaviour. In S. Worchel & W. G. Austin (Eds.), Psychology of intergroup relations (pp. 7–24). Chicago: Nelson.

Tausch, N., Tam, T., Hewstone, M., Kenworthy, J. B., & Cairns, E. (2007). Individual-level and group-level mediators of contact effects in Northern Ireland: The moderating role of social identification. British Journal of Social Psychology, 46, 541–556.

Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D., & Wetherell, M. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Blackwell.

Wann, D., & Branscombe, N. (1995). Influence of level of identification with a group and physiological arousal on perceived intergroup complexity. British Journal of Social Psychology, 34, 223–235.

Biographical notes jared b. kenworthy is an Assistant Professor of Psychology at the University of Texas at Arlington. His research focuses on the role of identification with social groups in several areas of intragroup and intergroup dynamics.

melisa a. barden is an Assistant Professor of Psychology at Walsh University. She received her PhD from the University of Texas at Arlington. Her research concerns group processes within religious group memberships with a focus on ingroup identification.

steven diamond is a graduate student in the Psychology Department at the University of Texas at Arlington. He is currently researching the effect of cognitive load on automatic and controlled processing.

alejandro del carmen is a Professor and Chair of the Department of Criminology and Criminal Justice at the University of Texas at Arlington. He earned a Ph.D. in criminology and criminal justice from Florida State University in 1997. His research interests include law enforcement, racial profiling, crime prevention, and corrections.

at UNIV OF TEXAS ARLINGTON on September 3, 2012gpi.sagepub.comDownloaded from