Developing a Memory and Decision Model for Eyewitness Identification

University Of California-Riverside, Riverside CA

Investigators

Abstract

Our understanding of eyewitness identification has increased considerably in the last 20 years. There is now a long list of empirical results that have been shown to be both reliable and illuminating with respect to the underlying psychological mechanisms. This research builds upon this foundation to develop a unified theory of the memory and decision processes that witnesses use when they are presented with a lineup and must make an identification decision, including the decision to not make an identification. This research will use a computer simulation program, Witness, where the model's fundamental assumptions are formalized mathematically. This model will be used to motivate and interpret the results of several experiments that are designed to address the fundamental question: How do witnesses make identification decisions? The experiments specifically test the decision rule presently incorporated in the model. The model will be fit to data from experiments that carefully covary suspect and foil similarity; examine the relationship between suspect-matched and description-matched lineups; compare showups to sequential and simultaneous lineups; and examine how more specific witness instructions affect identification.