Modeling decision processes in the use of lethal force : the role of racial bias in judging faces

To empirically address the question of whether and why police officers are more likely to shoot Black than White suspects, psychologists have developed the First-Person Shooting Task (FPST): a laboratory task in which participants must make shooting decisions based on rapid assessments of whether a Black or White target is holding a gun versus a harmless object. Typically, studies employing the FPST have found that participants' errors and reaction times show a bias toward shooting Black targets over White targets. Evidence for the mechanisms behind this bias is mixed, but several studies point to stereotypic associations between the category "Black" and some indication of threat (e.g. weapon possession). Collectively, this past work is suggestive that racial bias on the FPST is influenced by racial bias in threat perception. I investigated this hypothesis across three studies. Participants rated Black and White faces with regard to how "threatening" the faces appeared, then completed the FPST 3-15 days later. Behavioral and process-level (Drift Diffusion Model) methods were used to determine whether racial bias in a participant's threat ratings explained racial bias in the FPST. Across two stimulus sets, results indicated that although participants displayed process-level racial bias, this was not explained by biased threat perceptions. I consider implications such as the possibility that biased shooting decisions are produced by information-processing mechanisms rather than affective mechanisms.