ocial(2091-64 96 res from Amodio and Devine (2006)Studies 1-3 and Study 1 of the present projec Amodio and Devine (2006) Present work Study Study 2 Pre nary study Study and latency to neutral words were included as covariate redu it seem ment-re elir ment-rela o words dhis pres that alth General discussion weremuch than in The goal of this proiect was to explore whether Obama's histor bias.Ou that these le am ign were less ex posed to the campaign's media.th uced.Study 1 demonstrated th aor nderstand ng e to pos in tw n th ngh citi oss the nationi The signin ance of t ese findings is p ided a e ofe on night ger nced the u ciations at least some people carry dcoud hav mplication The unpre edented drop in implic bias observed in lab ute o the perpe n.Dunton.&William 1995.Fo example.McCo ugh the full i f this hi will pla with idant no erbal be ut over time,we are encouraged by the early returns. non bal behaviors may Acknowledgments 2 rgrou ome more pos ity g for 200he5 d to co Thiswork was by the National ut being ntainystigmatC University of Wisconsin Kellet Mid-career Award to thec ay be ough p ative and potentially v rtant,w ring set References t constru olicit bia ess. we believe the pa 6 The prejudice.Pap same tim drawin Lenton.A P.(20prime, and latency to neutral words were included as covariates. The analysis revealed that participants who responded more quickly to government-related words when primed with Black re￾sponded with less implicit prejudice, b = .59, t(47) = 2.87, p = .006. In addition, if the analyses were conducted with the la￾tency to the crime-related words instead of the government-re￾lated words, there was no impact of the crime-related latencies on the D-scores, indicating the effect was specific to the govern￾ment-related words. General discussion The goal of this project was to explore whether Obama’s historic presidential campaign and the resulting high levels of exposure to a positive, counter-stereotypic Black exemplar led to reduced racial prejudice and stereotyping. We found dramatically decreased lev￾els of implicit prejudice (Studies 1 and 2) and stereotyping (Study 1) compared with previously observed levels of bias at our own institutions and in the literature more generally. Providing some insight into why the bias was reduced, Study 1 demonstrated that the tendency to have positive Black exemplars come to mind or anticipate that other people had these positive exemplars come to mind when they thought of Black people was associated with low levels of racial prejudice. Study 2 found that participants who had an increased activation of qualities associated with Oba￾ma as a political figure when primed with ‘‘Black” had lower levels of implicit prejudice. The significance of these findings is poten￾tially extraordinary. Obama’s meteoric rise to fame and success provided a naturally occurring sequence of events whereby a coun￾ter-stereotypic Black individual’s rise to prominence seems to have influenced the underlying associations at least some people carry around in their minds about Black people. The reduction in implicit racial bias we observed could have a range of encouraging implications. Historically, these negative biases led to prejudiced responses on behaviors that are not easily controllable, behaviors that contribute to the perpetuation of pre￾judice (Devine, 1989; Dovidio, Kawakami, & Gaertner, 2002; Fazio, Jackson, Dunton, & Williams, 1995). For example, McConnell and Leibold (2001) found that anti-Black implicit prejudice was associ￾ated with avoidant non-verbal behavior (e.g., reduced eye contact, increased interpersonal distance). If implicit responses become more positive, avoidant non-verbal behaviors may be reduced and intergroup interactions may become more positive. Addition￾ally, intergroup interactions may become less draining for both majority group members, for whom the need to control bias may be obviated (Richeson & Shelton, 2003), and minority group mem￾bers, for whom concerns about being unfairly stigmatized may be reduced (Inzlicht, McKay, & Aronson, 2006). Though provocative and potentially very important, we note that our findings are correlational. Because the work capitalized on a naturally occurring set of events, we were unable to manipu￾late exposure to Obama, which, of course, would have provided stronger evidence that exposure to Obama caused the drop in im￾plicit bias. Nevertheless, we believe the pattern of findings across our studies provides compelling support for this possibility. At the same time, we encourage caution in drawing overly strong con￾clusions from these findings. We should note that the longevity of these effects is unclear. It may be that, as a major political figure, he will continue to be a salient and highly activated Black exem￾plar. Alternatively, over time, he could be subtyped and his general impact reduced (Rothbart & John, 1985). Furthermore, it seems plausible that the success of Obama’s presidency may have impli￾cations for his future role as an exemplar. If his presidency is highly successful, he would likely activate positive traits, thoughts, and feelings for most people. However, the result may be less positive should his presidency prove to be less successful. It is also important to note that although the levels of implicit prejudice and stereotyping were much lower than in previous studies, a sizable proportion of our participants responded with anti-Black bias. Our findings suggest that these people are less likely to have positive exemplars and words related to Obama’s campaign come to mind when they think of Black people. How￾ever, it is not clear why they responded in this way. Possibly, they were less exposed to the campaign’s media blitz. Alternatively, the strength and stability of their racial attitudes may have resulted in resistance to change. Exploring these possibilities will be impor￾tant for understanding resistance to prejudice reduction as well as the processes involved in the bias reduction we observed. Although we observed reductions in racial bias in two distinct regions of the county, all participants were non-Black college stu￾dents. It will be important to measure implicit racial bias in the more general voting public. And, although the impact of this elec￾tion is likely to reverberate through citizens across the nation, it will be essential to explore the implications of Obama and his pres￾idency for the attitudes and experiences of African Americans. In￾deed, the poignancy of election night was nowhere stronger than among African Americans for whom this election may signal that possibilities previously unavailable to them are now within reach. As we complete this project, we stand at the threshold of an￾other extraordinary moment in history. Barack Obama has been elected the first African American president of the United States of America. The unprecedented drop in implicit bias observed in our studies indicates that the impact of Obama’s historic campaign went beyond him winning the election. It appears to have pro￾duced a fundamental change in at least the minds of the American public. Although the full impact of this historic election will play out over time, we are encouraged by the early returns. Acknowledgments This work was partially supported by the National Science Foundation, Grant BCS-0544598 awarded to the first author and a University of Wisconsin Kellet Mid-career Award to the second author. References Amodio, D. M., & Devine, P. G. (2006). Stereotyping and evaluation in implicit race bias: Evidence for independent constructs and unique effects on behavior. Journal of Personality and Social Psychology, 91(4), 652–661. Banaji, M. R. (2005). Mind bugs: The psychology of ordinary prejudice. Paper presented at the Midwestern Psychological Association meeting, Chicago, May. Blair, I. V., Ma, J. E., & Lenton, A. P. (2001). Imagining stereotypes away: The moderations of implicit stereotypes through mental imagery. Journal of Personality and Social Psychology, 81(5), 828–841. Table 1 IAT D-scores from Amodio and Devine (2006) Studies 1–3 and Study 1 of the present project. Amodio and Devine (2006) Present work Study 1 Study 2 Study3 Preliminary study Study 1 Mean eval D-score (SD Eval) .51(.42) .32(.17) .38(.29) .03(.44) .03(.50) Mean ster D-score (SD Ster) .17(.43) .29(.23) .15(.18) .01(.39) .03(.38) E.A. Plant et al. / Journal of Experimental Social Psychology 45 (2009) 961–964 963