Wells, G.L. & Bradfield, A.L. (1998). "Good, you identified the suspect": Feedback to eyewitnesses distorts their reports of the witnessing experience. Journal of Applied Psychology, 83, 360-376.

BACKGROUND: False identification is the primary cause of wrongful imprisonment. While a witness may be hesitant and unsure while initially identifying a suspect, by the time it comes to trial, there is no doubt in the witness’s mind. Wells and Bradfield theorize that this may be due to feedback received at the time of the initial identification.

EXPERIMENT 1:

PURPOSE: To see whether the manipulation produces broad effects on a variety of measures that contribute to the credibility of an eyewitness

PROCEDURE: Subjects viewed a video and were told to pay particular attention to the suspect. They were then shown mug shots. Unbeknownst to them, the suspect was not pictured in the mug shots. All subjects made a false identification. Some subjects were then given feedback (confirming or disconfirming) about their selection. Others were not given any feedback. Subjects then filled out a questionnaire.

RESULTS: Based on the questionnaire, subjects given confirming feedback were significantly more confident of their identification of the suspect. Likewise, they felt they had a better view of the suspect and paid more attention to his face. They also judged it took them less time to make the identification, and were more willing to testify at trial.

EXPERIMENT 2:

PURPOSE: To see if eyewitnesses report accurately on the influence of the feedback, as well as whether or not the feedback effect can be prevented.

PROCEDURE: All subjects received either confirming or disconfirming feedback, but it was withheld until after some subjects assessed their confidence or the view they felt they had.

RESULTS: Again, all subjects made a false identification. Results of Experiment 1 were replicated. Additionally, asking subjects about their certainty prior to feedback tended to null the feedback effect, while asking them about their view prior to feedback did not eliminate the feedback effect. Fifty-eight percent of the subjects claimed the feedback did not influence them, although the study indicated that subjects who claimed not to have been influenced were no less influenced than other subjects.