Accuracy of eyewitness identification in a field setting

64 White and 9 Black 17-60 yr old clerks working alone in convenience stores were asked by "law interns" to identify from photograph lineups (prepared by the local police department) 2 male customers, one Black and one White, who had been in their store 2 hrs earlier. Ss were able to make correct identifications about one-third of the time. Even when no-guesses were omitted, identifications were correct less than half (46.8%) of the time. There was a substantial relationship between accuracy and Ss' confidence that they were correct. Only slight evidence of an own-race bias in accuracy was found among the Whites. White Ss' ability to identify the Black customer was significantly related to the amount of self-reported cross-racial experiences. The attractiveness and distinctiveness of the customers was related to the frequency of correct identifications, as was the effective size and functional size of the lineups used. Black Ss showed better overall recognition accuracy than did White Ss. (29 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved).