The Cross-Race Effect in Lineups Versus Showups

Abstract

A witness's ability to discriminate innocent from guilty suspects is reliably higher for same-race than for cross-race faces. We investigated the relative magnitude of this well-established effect in lineups versus showups. Based on diagnostic feature-detection theory, we hypothesized that the cross-race effect (CRE) for showups would be larger than that for lineups. However, contrary to our predictions, a similarly strong CRE was observed for showups and lineups, and high-confidence accuracy trended lower in the cross-race condition regardless of presentation format (showup vs. lineup). These results may indicate that witnesses did not initially encode certain diagnostic features of cross-race faces that might otherwise be put to good use in lineups. Moreover, the apparent failure of participants to adjust for the lower discriminability in the cross-race situation when expressing confidence may indicate that they may not fully appreciate how much more difficult it is to identify faces of a different race.