Although automated diagnostic aids are becoming increasingly common in complex systems such as aviation and health care, operators may not always trust them since these aids are unlikely to be 100% reliable. This study examines if automation errors on tasks easily performed by humans undermine trust in automation, and the manner in which the easiness and type of imperfect automation errors affect trust and dependence. Participants performed a target detection task using an automated aid. In the first study, the aid missed targets either on easy trials or on difficult trials. In the second study, both easiness and type of error (miss vs. false alarm) were manipulated. The aid erred on either difficult trials alone or on difficult and easy trials. In both experiments, easy errors led to participants mistrusting and disagreeing more with the aid on difficult trials, as compared with those using aids that generated only difficult errors. This resulted in a downward shift in decision criterion for the former, leading to poorer overall performance. Misses and false alarms led to similar effects. The authors suggest that providing human operators with specific information regarding the functional limitations of automation should give operators a better opportunity to appropriately discriminate between the relative accuracy of automation and their own abilities.
Abstract