The impact of anthropomorphic interfaces on influence understanding, and credibility

This paper describes a pilot study on decision influence, understanding, and subjective measures of credibility in human interaction with a computerized partner for a fictional decision-making task. Subjects (N=70) were randomly assigned to one of five different computer partners or to a human partner. Subjects completed the Desert Survival Task, and engaged in a dialogue with their partner. Pre- and post-interaction rankings were used to measure decision quality and influence. Results revealed that face-to-face interaction generated the most positive social judgments, and more positive social judgments were associated with greater understanding, but the computer conditions were more influential, especially when the computer was the least anthropomorphic.