Adaptive automation refers to a system capability that enables task sharing between a human operator and a system. The purpose for this type of collaborative sharing is to maintain a moderate level of task load, particularly in a multi-tasking environment. However, some costs might accrue from switching automation on and off, as is shown from task switching literature. Additionally, it is possible that congruency between task demand and the level of automation affects performance. Thus, before system-controlled adaptive automation is implemented into an operational environment, the goal for the present experiment is to examine the costs associated with turning automation on and off and to investigate the effects of demand/automation congruence. Analysis of the congruence effects revealed performance to benefit from higher levels of automation, regardless of task load. Task switching caused by adaptive automation was found to be detrimental to performance during periods of high task demand, but was beneficial during periods of low demand. (Author/publisher)
Abstract