Crime seriousness and its effects on eyewitness accuracy, confidence, and arousal

Kevin M. Brooks, The College of Wooster


This study tested the effect of crime seriousness on eyewitness recognition, confidence, and selfreported arousal, using a single factor design with 3 levels; no-crime, moderately-serious crime, and serious crime. Participants (n=44) were asked to role-play as a security guard and observe a surveillance video looking for a suspect. The audio portion of the surveillance video was manipulated and the message varied in reported crime seriousness, along with subsequent voice urgency, and intensity. The participants then completed a self-report arousal instrument, the Activation-Deactivation Adjective Check List (Thayer, 1989), and a suspect characteristics recognition test. The key finding in this study included a significant difference using a planned comparison at a one-tailed level, for eyewitness accuracy, where performance in the no-crime condition was better than in the crime-present conditions. The trends for confidence ratings declined as the seriousness increased, although not significantly. The crime manipulations had no impact on self-reported arousal ratings. Thus, from this study it can be concluded that: crime presence had an effect on eyewitness recognition accuracy, and confidence and arousal were not affected by crime seriousness.