AuthorsG. A. Baugerud, M. Powell, M. Riegler, P. Halvorsen, M. E. Lamb, S. S. Sabet, S. Z. Hassan, P. Salehi, R. K. Røed and M. S. Johnson
TitleInvestigative Interviews using a Multimodal Virtual Avatar
AfilliationMachine Learning
Project(s)Department of Holistic Systems
StatusAccepted
Publication TypeProceedings, refereed
Year of Publication2022
Conference NameAmerican Psychology-Law Society Conference 2022
PublisherAmerican Psychology-Law Society
Place PublishedDenver USA,
Abstract

To meet best-practice standards, we are developing an interactive virtual avatar aiming as a training tool to raise interviewing skills of child-welfare and law-enforcement professionals. Therefore, we present the “Ilma” avatar that recognizes interviewers’ behavior during open-ended, closed and leading questions, and which can automatically respond to the conversation. We conducted a user study in which master students (N=3) and child protective workers (N=8)  interviewed “Ilma”  and rated their perception of the interaction. The results show that the participants valued the interaction and found the avatar useful. Thus, it has great potential to be an effective training tool.

Citation Key28231