Ten cases of eyewitness identification: Logical and procedural problems

Abstract

Procedural aspects of identification tests constitute a useful area for expert testimony by experimental psychologists because experimental psychologists have the relevant expertise and because police investigators make numerous potentially fatal mistakes. This article outlines the dual purpose of identity testing: one test is supposed to prove simultaneously the guilt of a suspect and the reliability of the witnesses. This dual objective can be achieved only if the underlying logic is not disturbed by procedural flaws. Twelve of these flaws are discussed. Procedural flaws can be avoided through the adoption of a number of rules, as proposed by various bodies and authors. However, in practice such proposals are rarely followed, probably because the seriousness of procedural mistakes is not realized. Ten cases of disputed identity in which the authors acted as expert witnesses are presented. The cases are shown to have been severely undermined by procedural flaws that disrupted the evidentiary strength of the identifications, errors that could have been avoided, had stricter rules been adopted. It is argued that procedural rigidity will increase the discriminatory power of identity tests so that more innocent suspects are acquitted and more guilty suspects convicted. Until this objective has been achieved, it will remain useful for experimental psychologists to testify about identification procedures.