The reliability of eyewitness identifications by the elderly: An evidence-based review

The elderly make up an increasingly larger portion of the population and hence will be more likely to become victims or witnesses of crime. Eyewitness performance of elderly witnesses in person identification tasks, which are primarily based on face recognition, may not follow the same pattern as in recall tasks. We review the empirical evidence of lineup and face recognition studies, focusing on the results of two recent meta-analyses of our own, which both show deficits in the elderly's performance compared to younger people. In particular, fewer hits, more false identifications and false alarms as well as a more lenient decision criterion by elderly witnesses have been observed. Although there seems to some evidence for an own-age bias as postulated by Sporer's (2001a) in-group/out-group model, evidence regarding this effect is weak. The evidence-based importance of different moderator variables as well as special measures to improve identification accuracy by elderly witnesses are reviewed. Research deficits, a future agenda, and practical implications for evaluating identification evidence by elderly witnesses are pointed out. Evaluation of elderly witnesses is embedded in the context of stereotypical beliefs about the elderly within the framework of an integrative model of eyewitness testimony.