Excerpted From: Justin D. Levinson, Huajian Cai and Danielle Young, Guilty by Implicit Racial Bias: The Guilty/not Guilty Implicit Association Test, 8 Ohio State Journal of Criminal Law 187 (Fall, 2010) (101 Footnotes) (Full Document)
Legal scholarship on racial discrimination has turned to the science of implicit social cognition to explain how the human mind automatically manifests biases against disfavored social groups. Much of this discourse on implicit bias focuses on the potential for massive, but hard to detect discrimination in the employment context. Yet, other legal domains where implicit racial bias may lead to persistent racial inequalities remain underexplored, most notably in criminal law. Specifically, a crucial question still needs to be answered: do implicit biases affect jury guilty/not guilty verdicts in racially biased ways?
Despite the broad incorporation of social science knowledge into legal discourse, a critical chasm continues to deter legal scholarship from fully achieving the social cognition-informed perspective it craves. Namely, legal scholarship on implicit bias lacks law-focused science. Legal analysts have implicitly assumed that existing social cognition measures, many of which are carefully developed and rigorously tested (but not developed with the law in mind), are the only options for theory development in the legal context. These tests have been groundbreaking in social psychological scholarship and their introduction into legal scholarship has addressed the need for the law to possess an understanding of the human mind. Yet, the still emerging legal model of the human mind has failed to develop new empirical tests that measure how implicit cognitive processes function not just in society in general, but specifically in legally relevant contexts such as jury decision-making.
Here is one example: a frequently cited psychological measure of implicit bias, the Implicit Association Test (“IAT”), examines people's implicit associations by measuring response speed in a computerized test. In one of the most famous IATs, study participants are asked to pair together words representing attitudes (Good and Bad) and photos depicting target group members (Black and White) as fast as they can. The results of these studies show that, when measuring response times and error rates, the vast majority of people are faster to pair together Good with White and Bad with Black. These results are considered to be indicative of implicit bias, and are eye-opening when considered in the legal context. Yet might these studies do even more to examine implicit bias in the legal system? For example, why should legal scholars be satisfied to rely on psychological research relating to implicit racial attitudes of “good” and “bad,” (and then engage in heated debate about what it really means in the legal context when it is possible to specifically test implicit associations of well known legally meaningful constructs, such as “guilty” and “not guilty?”
To address the lack of legally-focused empirical studies exploring implicit bias, we developed a new IAT: the Black/White, Guilty/Not Guilty IAT (“Guilty/Not Guilty IAT”). We designed this IAT to examine whether people hold implicit associations between African Americans and criminal guilt, a finding that would call into question criminal law's presumption of innocence and evoke larger questions of racial justice. Although the debate over racial disparities in the criminal justice system has been raging for decades, scholars have rarely adapted social cognition methodology to examine specifically the role of race in criminal law decision-making. We therefore created and developed the Guilty/Not Guilty IAT, and predicted that people implicitly associate Black and Guilty compared to White and Guilty. Because it is important not just to test implicit associations themselves, but to investigate whether they predict meaningful behaviors, we also tested whether responses on the Guilty/Not Guilty IAT predict the way mock jurors evaluate ambiguous trial evidence. The results of our study confirmed our hypotheses: study participants held strong associations between Black and Guilty, relative to White and Guilty, and these implicit associations predicted the way mock jurors evaluated ambiguous evidence. Furthermore, we compared our measure to a frequently administered IAT that tests positive and negative attitudes towards race, the Pleasant/Unpleasant IAT, and found that the Guilty/Not Guilty IAT and the Pleasant/Unpleasant IAT functioned differently, a result that demonstrates the uniqueness of the Guilty/Not Guilty measure.
This Article introduces the Guilty/Not Guilty IAT, details the empirical study we conducted and argues for the need to increase collaborations to employ social cognition methods to test legal hypotheses.
Section II presents an overview of IAT research in the legal context and notes the limited number of empirical studies that have been employed.
Section III sets the stage for our empirical study, first by reviewing the science behind the IAT, and second, by contextualizing the meaning of the Guilty/Not Guilty IAT within the doctrine of the presumption of innocence.
Section IV details the empirical study we conducted. The study tested implicit associations within an important legal domain and examined whether these implicit associations matter in legal decision-making. Results of the study showed that participants held implicit associations between Black and Guilty compared to White and Guilty, and that these implicit associations predicted mock-juror evaluations of ambiguous evidence.
Section V briefly discusses the implications of the study, and calls for increased empirical collaborations. Section VI concludes.
. . .
We consider this study just one part of a broader effort to investigate whether implicit biases facilitate societal and legal inequality. It is our hope that future empirical studies of implicit bias in the law will continue this effort, specifically by investigating a broad range of legal domains where implicit bias may affect legal decision-making. There are so many areas of the law that have yet to be considered as possible hideouts for implicit bias. A broad range of legal areas, including but not limited to immigration law, contract law, and property law may unknowingly be functioning with the covert and dangerous help of implicit bias. We hope that future studies, particularly those conducted by interdisciplinary research teams, will pursue these areas, while also continuing to investigate the Guilty/Not Guilty IAT.
Associate Professor of Law, University of Hawai'i..
Professor, Key Laboratory of Mental Health, Institute of Psychology, Chinese Academy of Sciences.
Department of Psychology, University of Hawai'i.