Close

Presentation

Combining Empirical and Computational Approaches to Model Trust Dynamics in Human-Autonomy Interaction
Event Type
Lecture
Virtuals
In-Person
Technical Groups
Human AI Robot Teaming (HART)
TimeThursday, October 7th11:06am - 11:24am EDT
LocationGrand Salon VII
DescriptionTrust has been identified as one central factor in effective human-autonomy interaction. However, most studies have used a “snapshot” view of trust and evaluated trust using questionnaires at the end of an experiment. This “snapshot” view does not fully acknowledge that trust is a dynamic variable that can strengthen and decay over time. With few exceptions, we have little understanding of the temporal dynamics of trust as a result of moment-to-moment interactions with autonomy. In this paper, we present the results of two studies examining trust dynamics in human-autonomy interaction. In study 1, we identify and define three properties of trust dynamics, namely continuity, negativity bias, and stabilization. The three properties characterize a human agent's trust formation and evolution process. In study 2, we propose a computational model of trust dynamics that adheres to the three properties and evaluate it against existing trust inference models. Our model provides superior performance.