Does visual attention to diagnostic facial features predict eyewitness identification accuracy? Implications for diagnostic feature detection theory.

A police lineup includes a suspect and known-innocent fillers. The purpose of fillers is to protect an innocent suspect as witnesses must rely on their memory to make a lineup decision (i.e., select or reject). Diagnostic Feature Detection theory proposes that witnesses compare facial features across lineup members, weighting shared and non-shared features differently during the decision process (Wixted & Mickes, 2014). According to this view, only non-shared features are diagnostic because they provide distinguishing information, compared to features that are shared amongst all lineup members. Using eyetracking, we examined eyewitnesses' viewing behavior and decision making in lineups when features amongst lineup members were shared. After encoding twelve target faces (six Black, six White), participants (n = 100) made identification decisions on target-present lineups (one lineup per target), with lineup members presented either simultaneously or sequentially, providing confidence after each lineup decision. In the shared-feature lineups, we digitally edited one of the target's features (nose, eyes, or mouth) onto all lineup members so that they shared that feature (e.g., in one lineup, all members had the target's eyes). In the control lineups, we edited each lineup member's own feature (nose, eyes, or mouth) to control for effects of photoshopping. Results indicate there were no differences in allocation of overt visual attention (via eye movements) to diagnostic features between lineup types or by accuracy. These findings do not support the proposition that eyewitnesses overtly evaluate shared features in lineups, or account for their diagnosticity via allocation of overt visual attention.