Effects of Mental Model Quality on Collaborative System Performance

Abstract

This ongoing study is exploring how varying levels of user mental model quality influence performance with an imperfect automated system (navigational aid). The number and types of errors produced by participants is of particular interest along with how mental models influence user trust and reliance with automation. The design attempts to manipulate the acquisition of mental models by varying exposure to the task environment between groups. A high acquisition group, low acquisition group, and no acquisition group navigated the confines of a simulated city with the help of an automated aid performing at two levels of reliability. Implications of this study could influence the design of automated systems and their associated user training programs. Additionally, this study could spur additional research into the effects of mental model quality on automated system performance.