The Unites States criminal justice system pays a substantial social cost for eyewitness misidentification. Eye-witness misidentifications represent some of the most devastating mistakes in the criminal justice system, and their likelihood of occurrence is greatest when identifications are made across racial lines. The cross-race recognition deficit, known more simply as the Cross-Race Effect (CRE), is a tendency to have better recognition accuracy for same-race faces than for cross-race faces. This effect has been shown to be surprisingly robust. It is one of the best-replicated phenomena in face perception, and has been of interest to psychologists for more than half a century. Although this phenomenon is interesting in its own right, the high degree of interest in the CRE is due, at least in part, to the potentially devastating legal consequences of misidentification. Given the reliance of the criminal justice system on eyewitness identification, and the substantial weight that eyewitnesses have on juridic decisions, understanding the mechanisms underlying the CRE and how to ameliorate these effects are clearly important goals for psychology. The current research will investigate the extent to which social categorization can account for the CRE. This research suggests that the CRE may in part be due to 1) the tendency to think more deeply about in-group members than out-group members (leading to better encoding of same-race face features), and 2) the tendency to pay attention to different information in the faces of out-group members, focusing not on what makes the face unique, but rather on what makes the face similar to other faces. Across 9 experiments, this research will test the hypothesis that the well-replicated CRE may in part be due to well-understood processes of social categorization. These studies employ social cognitive manipulations of 1) in-group vs. out-group membership, 2) motivation, and 3) perceived group distinctiveness to create, modulate, and eliminate the CRE. The most frequently used paradigm to study the CRE is a relatively straightforward two-phase encoding-recognition paradigm. At Time 1, participants complete an encoding phase, where they observe a series of Same Race (SR) and Cross Race (CR) faces, one at a time, each for a fixed period of time. At Time 2, after an intervening filler task designed to reduce the likelihood that participants were able to retain the previously-seen faces in working memory, participants engage in the recognition phase. In this recognition task, participants see a series of SR and CR faces, half of which were seen previously and half of which are new. Participants are asked to report which faces are 'old' faces (i.e., seen at Time 1) and which are 'new' faces (i.e., not seen at Time 1). All studies proposed here will use variants of this standard procedure. Findings from this research will help in understanding the cause of the CRE, and thereby reduce the problems associated with eyewitness misidentification.
