BY CHRISTINA LINCOLN
n the Journal of Applied Psychology, I read an article about eyewitness identification accuracy. Subjects watched a 5-minute film that was prepared for the experiment. The film showed a robbery of a cashier's box from a refreshment stand in a park. Subjects were then asked to identify the criminal in a series of photographs and to rate their level of confidence in their identification choice both before and after their choice. The photographs were presented one at a time (sequential) or all together (simultaneous).
The experiment investigated the accuracy of a witness's choice in sequential lineups versus simultaneous presentation of the criminal in a line. The study hypothesized that the witnesses choosing in the sequential lineup condition would have faster response times for the correct identification of the criminal (a hit) than the time taken to incorrectly choose one of the innocent people in the lineup (a foil). It was further hypothesized that the more confident witnesses would respond faster than the less confident witnesses (a negative correlation). The researchers also investigated if among the witnesses choosing somebody in the lineup (yes response), accurate witnesses (correct identifications) would respond faster than inaccurate witnesses (false identification).
In the traditional, simultaneous lineup situation, witnesses must compare the relative similarity of the people in the lineup with their memory of the thief (relative-judgement processing). When the photos are presented sequentially the witness must match a specific lineup member with their representation or image of the thief in memory (absolute-judgement processing). If a specific member of the lineup exceeds the witnesses' internal threshold used as a basis for their identification decision they will choose that specific member of the lineup. If the person being viewed does not exceed the internal threshold of the witnesses' identification decision, he will be rejected.
The results found that there were significantly fewer false alarms in the sequential lineup condition (18.3%) than in the simultaneous lineup condition (35%). Absolute- judgement processing (sequential identification) seems to lead to fewer false alarms without losing the power to detect if the criminal is present. The results for the confidence levels were correlated and it was found that predecision confidence was related neither to postdecision confidence, r(70)= .19, .05
REFERENCE
Sporer, S. (1993). Eyewitness identification accuracy, confidence, and decision times in simultaneous and sequential lineups. Journal of Applied Psychology, 78, 22-33.