JOINT HUMAN-AUTOMATION DECISION MAKING IN ROAD TRAFFIC MANAGEMENT

摘要

In this paper we explore automation bias in terms of joint decision makingrnbetween humans and automation. In an experiment, participants maderndecisions, and indicated the reason for this decision, in a road trafficrnmonitoring task with the aid of automation of varying reliability (i.e., 25% orrn81%). Reliability level had a clear impact on the user’s behavior: at lowrnreliability, participants ignored automation suggestion and rely on their ownrndecision making, whereas in the high reliability condition, participants tendedrnto accept the automation suggestion (even if this was incorrect). Overall,rnperformance is higher as a result of the human intervention that would bernexpected from automation alone, i.e., accuracy is in the region of 87-96% onrnall conditions. Performance is affected by the order in which the human andrnautomation give their answers and how much detail they are required tornprovide. We consider these results in terms of a theory of joint decisionrnmaking. [156 words]