Increasing the level of detail in adult witnesses’ event accounts: Can an example of required detail help?

Abstract

Because eyewitness accounts are a critical source of evidence in many police investigations, it is important for this evidence to be both detailed and accurate. Eliciting evidence that fulfills both of these criteria, however, poses a considerable challenge. We know that witnesses’ responses to broad, free recall questions (e.g., “Tell me everything that happened”) tend to be highly accurate, but insufficiently detailed for investigative purposes. When interviewers ask more specific questions (e.g., “Was there a weapon?”), they elicit more detail, but accuracy decreases. To increase free recall detail, police investigators have recently begun to describe an object (e.g., a pen) to witnesses, to illustrate the level of detail desired. However, the validity of this technique is yet to be empirically tested. Here, we investigated whether providing a detailed description of a nearby object (e.g., a pen) would increase the amount of detail reported by adult witnesses, and—if so—whether that increase would come at the expense of accuracy. Our participants watched a short film clip of a simulated aggravated robbery. After completing a distractor task, participants were given the Cognitive Interview instructions used by the New Zealand Police. For half of the participants, the interviewer described a pen to illustrate the level of detail required during the interview; remaining participants were not exposed to this description. Participants were then asked to provide a free recall report of the robbery before being asked a series of open-ended and yes/no questions. The quantity (i.e., the number of details) and quality (i.e., accuracy) of participants’ responses to these questions were measured. The pen description led to a small increase in the number of details that participants provided during free recall, without compromising accuracy. The description did not influence participants’ responses to the open-ended or yes/no questions. Although our study provides some preliminary evidence for the validity of the pen description technique, the size of the effect must be weighed up against the potential drawbacks of such a technique—particularly in cases involving sexual violence.