The fallibility of memory : investigating its reliability in eyewitness testimonies

Abstract

The purpose of this dissertation was to explore factors affecting eyewitness memory. An experiment with a two-by-two between group design was conducted in which recall memory accuracy was measured by an ad hoc recall test on a video of a simulated crime. The independent variables were time between viewing video and recall test (immediately, a week later) and testing condition (individual, group). Confidence in recall memory accuracy was measured for those tested individually, and conformity (Mehrabian & Stefl, 1995) was measured for those tested in a group. Data was primarily analysed quantitatively. Time between viewing of video and testing did not have an effect on recall accuracy in either of the two testing conditions, however, it did have an effect on confidence. Participants tested individually outperformed those tested in a group when testing was done immediately but not when testing was done a week later. Confidence was not found to correlate with memory recall accuracy during individual testing and conformity was also not found to correlate with the number of changes to a reply during group discussions. From these findings one could gain an understanding on the relationship between time and confidence, as well as how cowitness discussion affects the outcome of eyewitness testimonies.