ABSTRACT
Traditional identification procedures—where witnesses either pick a lineup member or reject the lineup—suffer a number of limitations. First, non-memorial influences on witness’s decision criteria can contribute to identification error by increasing (or decreasing) the likelihood that a witness will pick someone, independent of the quality of the witness’s memory or the degree of match between individual lineup members and the witness’s memory for the culprit. Second, a categorical identification response is less informative than it might appear. Although a categorical identification presumably indicates that the identified person best matched the witness’s memory of the culprit, it provides no information on (a) the degree of match or (b) the extent to which that person was favored over the alternative lineup members. Further, a rejection provides no information about the extent to which the suspect matched the witness’s memory (other than indicating that the degree of match did not exceed the threshold for identification). In this chapter, we review the applied and theoretical motivation for a novel approach to collecting identification evidence: having witnesses indicate, for each lineup member, their confidence that this person was the culprit. We present empirical evidence from face recognition and eyewitness identification paradigms demonstrating that, compared to categorical responses, ratings provide a richer source of information when evaluating the quality of a witness’s memory and better discriminate a target among foils in a lineup. We also consider recent research investigating how mock-jurors respond to this evidence. Finally, we provide guidance for researchers on analyzing confidence rating data.
