Bias and regulation of bias in intergroup interactions: Implicit attitudes toward Muslims and interaction quality

Previous research suggests that automatically activated bias manifests itself in behavior that can jeopardize the quality of intergroup interactions. However, regulation of automatic associations has the potential to attenuate their influence on intergroup interaction. To test this possibility, 46 non-Muslim White participants interacted with a Muslim confederate and completed an implicit measure of attitudes toward Muslims. The Quadruple Process model [Sherman, J. W., Gawronski, B., Gonsalkorale, K., Hugenberg, K., Allen, T. J., & Groom, C. J. (2008). The self-regulation of automatic associations and behavioral impulses. Psychological Review, 115, 314–335] was applied to the implicit measure to estimate participants’ strength of negative associations with Muslims and their ability to overcome those negative associations. The confederate’s ratings of howmuch he liked the participants were predicted by an interaction between automatic negative associations and the ability to overcome them. Specifically, when the strength of participants’ negative associations with Muslims was low, participants’ level of overcoming bias was unrelated to the confederate’s ratings. In contrast, the ability to regulate automatic negative associations predicted greater liking when those associations were strong. 2008 Elsevier Inc. All rights reserved. The biases that people hold, even those that they may prefer not to have, can lead them to behave negatively toward outgroup members. Evidence for this link has emerged with research employing implicit attitude measures, which seek to capture automatically activated biases that people may not personally endorse or even be aware of having (see Fazio & Olson, 2003). Compared to those who do not show bias on implicit measures, White individuals with high levels of implicit race bias are more likely to display nonverbal discomfort during cross-race interactions (Dovidio, Kawakami, & Gaertner, 2002; Dovidio, Kawakami, Johnson, Johnson, & Howard, 1997; McConnell & Leibold, 2001), be rated as unfriendly by Black experimenters and confederates (Fazio, Jackson, Dunton, & Williams, 1995; McConnell & Leibold, 2001; Sekaquaptewa, Espinoza, Thompson, Vargas, & von Hippel, 2003), and have interracial roommate relationships that are short-lived (TowlesSchwen & Fazio, 2006). Similar patterns have been found whether bias was assessed using the Implicit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998; e.g., McConnell & Leibold, 2001; for a review, see Greenwald, Poehlman, Uhlmann, & Banaji, in press), evaluative priming tasks (e.g., Dovidio et al., 1997), or implicit stereotyping measures (e.g., Sekaquaptewa et al., 2003). These findings indicate that implicit attitudes and beliefs are important predictors of how people behave in face-to-face interactions. Although the link between implicit biases and intergroup behavior has been established empirically, less is known about how and when this relationship emerges. One possibility is that automatic biases ‘‘leak” through behavior in attitude-relevant situations (e.g., Dovidio et al., 2002). According to this account, if an individual associates an outgroup with negativity, this association will precipitate prejudice-consistent nonverbal behavior in the presence of an outgroup member, even when the actor might prefer to act differently. However, when people have sufficient motivation and cognitive resources to regulate the effects of automatically activated associations, they may be able to prevent their biases from influencing behavior (Fazio & Towles-Schwen, 1999). Research indicates that individuals engage in self-control when interacting with or perceiving outgroup members (Richeson & Shelton, 2003; Richeson et al., 2003), who in turn appear to appreciate these self-control efforts. For example, White participants who were instructed to ‘‘try not be prejudiced” during an 0022-1031/$ see front matter 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jesp.2008.07.022 q This research was supported by an Australian Postgraduate Award to the first author and a grant from the Australian Research Council. We thank Brad Hill for running the experimental sessions and the confederates who took part in the study. * Corresponding author. Fax: +61 2 9036 5223. E-mail address: [email protected] (K. Gonsalkorale). Journal of Experimental Social Psychology 45 (2009) 161–166