Reducing Social Demands in Child Eyewitness Memory Using Robot Interviewers
Ward, Alexander James
Buchanan, A. Zachary
MetadataShow full item record
A recent study showed that children's memory for a witnessed event was not impaired when misleading post-event information was presented by a robot interviewer. However, their memory was impaired when the misleading information was presented by a human interviewer, despite the implementation of the National Institute of Child Health and Human Development's Investigative Interview Protocol (NICHD). The NICHD is specifically designed to elicit veridical information from child eyewitnesses, in both human and robot conditions. The proposed explanation for these memory findings is that the social demands inherent in human interviewers were not present in robot interviewers. The current study sought to investigate why the implementation of the protocol failed to limit these social demands in human interviewers, and how the interview procedure differed between human and robot interviewers despite the implementation of the NICHD protocol. I conducted content analyses of the verbal and nonverbal cues given by human and robot interviewers; interviews were coded for word counts, question types, insertions of prior knowledge, and types of interviewer responsiveness. The human interviewer spoke significantly more to children during only the initial rapport building phase. Additionally, the human asked more questions (open- and closed-ended), provided more affirmation, and exhibited more nonverbal encouragement. The human also interjected more prior knowledge into the interview, violating a rule of the NICHD protocol. The robot, less able to interject spontaneous conversation and connect with the children through verbal and nonverbal responsiveness, did not elicit the social demands inherent with the human interviewer, therefore ameliorating the impact of the misleading information on children's memory.