The tractability of eyewitness confidence and its implications for triers of fact.

Abstract

A theft staged for 80 unsuspecting eyewitnesses was followed by a picture lineup that did or did not contain the thief. In an attempt to see if eyewitness confidence is tractable after the identification, half of the eyewitnesses who identified the thief (accurate witnesses) and half who identified an innocent suspect (inaccurate witnesses) were briefed by a prosecutor who suggested they rehearse answers to potential questions that would be asked under cross-examination. Cross-examinations of 10 accurate briefed witnesses, 10 accurate nonbriefed witnesses, 9 inaccurate briefed witnesses, and 9 inaccurate nonbriefed witnesses were viewed by 152 subject-jurors in groups of 4. Briefed eyewitnesses rated themselves as more confident that they had identified the thief than did nonbriefed witnesses. This increase was primarily due to inaccurate eyewitnesses increasing their confidence, and the briefing manipulation served to eliminate the confidence-accuracy relationship. Subject-jurors were significantly more likely to vote guilty in conditions in which the eyewitness had been briefed than in the nonbriefed conditions. It is argued that briefing eyewitnesses, although legal, simply serves to increase the eyewitnesses' confidence in their memory, not the accuracy of memory. It is also argued that an accurate eyewitness may have memories that are already associated with high confidence, and therefore, briefing may primarily inflate the confidence of inaccurate eyewitnesses.