On increasing the usefulness of eyewitness research.

Abstract

A large amount of psychological research on eyewitness testimony has been conducted in the last fifteen years (see Wells & Loftus, 1984, and Yarmey, 1979, for comprehensive reviews). The thrust of this research is that witnesses often are confident and convincing despite serious errors in their descriptions and identifications, and that various conditions and procedures can either increase or decrease the chances of these errors being committed and detected. The potential value of this research to police investigators is manifold. Cases may be forwarded prematurely for prosecution if eyewitness testimony is given more credence than it deserves. Discrepancies within or between witnesses' recollections could lead to undue suspicion of motives and to damaging levels of ill will if the inherent weaknesses of human memory are not fully appreciated. Procedures could be modified so as to avoid the introduction of bias or to stimulate more complete and accurate testimony. To what extent will police investigators use psychological research on eyewitnesses as an aid in their work? I recently conducted interviews that shed considerable light on this issue. A representative sample of police throughout the state of New York were asked to discuss their experiences with eyewitnesses, and these unstructured conversations were recorded and content-analyzed. Interested readers may write to me for a detailed description of methods and results, but I would like to share the highlights of those interviews in this Note. The most striking and consistent theme to emerge is the virtual absence of any references, however oblique, to psychological research, even when it is