INTERPERSONAL RELATIONS AND GROUP PROCESSES Individual Differences in the Activation and Control of Affective Race Bias as Assessed by Startle Eyeblink Response and Self-Report David M. Amodio, Eddie Harmon-Jones, and Patricia G. Devine University of Wisconsin—Madison The activation and control of affective race bias were measured using startle eyeblink responses (Study 1) and self-reports (Study 2) as White American participants viewed White and Black faces. Individual differences in levels of bias were predicted using E. A. Plant and P. G. Devine’s (1998) Internal and External Motivation to Respond Without Prejudice scales (IMS/EMS). Among high-IMS participants, those low in EMS exhibited less affective race bias in their blink responses than other participants. In contrast, both groups of high-IMS participants exhibited less affective race bias in self-reported responses compared with low-IMS participants. Results demonstrate individual differences in implicit affective race bias and suggest that controlled, belief-based processes are more effectively implemented in deliberative responses (e.g., self-reports). Recent years have seen an increasing interest in the physiolog- ical correlates of social and affective processes. The incorporation of physiological measures into the study of racial prejudice has been especially appealing because these measures have the ability to circumvent response biases that often affect self-report instru- ments (e.g., Vanman, Paul, Ito, & Miller, 1997; Vrana & Rollock, 1998; for a review, see Guglielmi, 1999). The use of physiological measures as affective indices of racial attitudes has a relatively long history in social psychology, dating back to the mid 1950s. It was during this time in the United States that social norms pro- hibiting the public expression of racial prejudice began to emerge. Being wary that participants’ concerns over these norms might threaten the veracity of their self-reported racial attitudes, some prejudice researchers turned to physiological indices known to be resistant to overt control efforts. This early work demonstrated that White American participants had larger autonomic responses, for example, greater skin conductance, in response to Blacks com- pared with Whites across a number of different experimental paradigms (e.g., Rankin & Campbell, 1955; Vidulich & Krevan- ick, 1966). Although this work suggested an affective race bias, the skin-conductance measure was poorly suited to differentiate the valence of White participants’ affective responses to Black people. More recent work has begun to elucidate the neural substrates of race-biased affect in order to further probe the affective qualities of intergroup bias. A large body of research has identified the amyg- dala, a small collection of nuclei located bilaterally in the anterior region of the temporal lobes, as playing an integral role in affective processes (Adolphs, Tranel, & Damasio, 1998; Davidson & Irwin, 1999; Whalen et al., 1998). Much research has associated amygdala activity with negative affect and, more specifically, with the detection of threat (LeDoux, 1996). Hart et al. (2000) used functional magnetic resonance imaging to measure amygdala activity in White and Black participants while they viewed pictures of White and Black faces. Greater amygdala activity was observed in participants as they viewed pictures of out-group members compared with in-group members, indicating a race bias in basic-level affective physiology. 1 1 We deliberately avoid referring to race bias identified at implicit or basic affective and physiological levels as a “prejudiced” response. Al- though the presence of bias at these levels of processing may contribute to discriminatory behavior, it is important to distinguish such biases from explicitly held racist attitudes and beliefs (Devine, 1989; Devine, Monteith, Zuwerink, & Elliot, 1991). David M. Amodio, Eddie Harmon-Jones, and Patricia G. Devine, De- partment of Psychology, University of Wisconsin—Madison. Portions of this research were presented at the 2000 meeting of the Society for Psychophysiological Research in San Diego, California; at the 2001 meeting of the Society for Personality and Social Psychology in San Antonio, Texas; and at the 2001 meeting of the Society for Psychophysi- ological Research in Montreal, Quebec, Canada. This research was sup- ported in part by a grant from the Fetzer Institute, and by National Institute of Mental Health Emotion Training Grant T32-MH18931 for support of David M. Amodio. We thank Nikki Graf for her assistance in multiple stages of this research; Alison Covert, Donna Erhardt, Amber Peterson, and Anne Schoeneborn for their assistance in data collection and processing; Amanda Brodish and Suzanne Klonis for comments on earlier versions; and Jeremy Biesanz, Rodrigo Espana, and Alex Shackman for their helpful advice. We gratefully acknowledge the technical and mechanical support provided by Dirk Wilker, Ziggy Bialzik, Andy Mulder, and Greg Kant and the computer programming assistance provided by Craig Rypstat. Correspondence concerning this article should be addressed to David M. Amodio, Department of Psychology, University of Wisconsin—Madison, 1202 West Johnson Street, Madison, Wisconsin 53706. E-mail: dmamodio@ wisc.edu Journal of Personality and Social Psychology, 2003, Vol. 84, No. 4, 738 –753 Copyright 2003 by the American Psychological Association, Inc. 0022-3514/03/$12.00 DOI: 10.1037/0022-3514.84.4.738 738