This article examined the feasibility of a computer-based program that alleviates the human resource challenge associated with blind photoarrays (photoarrays in which the investigator is blind to the suspect’s identity). Students watched videotaped crimes and attempted to identify the perpetrators from photoarays conducted by a “virtual officer” who responds to simple voice commands or by research assistants playing the role of investigators. The student investigators and virtual officer produced comparable identification performance and student reactions to the photoarray procedures. Results of this evaluation study are encouraging, and the authors recommend further laboratory and field testing of the virtual officer technology for conducting blind lineups.

One solution to mistaken identification by a crime's victims and eyewitnesses is to use a virtual officer to conduct identification procedures. Results from a study comparing a virtual officer with a live human investigator indicate that the virtual officer performs comparably to the human in terms of identification accuracy, emotional affect, and ease of use.

An analysis of court cases has revealed that the mistaken identification of the wrong person by victims and witnesses of a crime is the single most common error leading to the arrest and conviction of innocent people [Wells et al. 2006]. Recognizing the role of mistaken identification in erroneous conviction, a growing number of states and police departments have reformed their eyewitness identification procedures. In this paper, we investigate a new procedural reform: the use of a virtual officer who does not know the identity of the suspect in the lineup and therefore cannot bias the witness toward false identification.