Presentation
Effects of Automation Error Bias on Dependence and Trust in Remotely Operated Air Vehicle Operations.
DescriptionAdvanced Air Mobility (AAM) operations will necessitate human operators teaming with automated systems with the goal of a single operator monitoring multiple aerial vehicles. Prior work has shown that interacting with unreliable automation affects trust and dependence behavior as a function of the type of error made. This study examined the effects of error bias of an automated decision aid on trust and dependence behavior within the context of AAM. Forty-eight participants performed a simulated collision detection task with the assistance of a collision avoidance aid that varied in error bias (false-alarm prone, miss-prone, even number of error types, or without errors). False alarm-prone systems reduced compliance measures to a similar extent that miss-prone systems reduced reliance measures. Saliency of miss errors can play an important role in trust development and dependence behavior in a situation that a consequence of misses outweighs that of false alarms.
Event Type
Lecture
TimeWednesday, September 11th10:05am - 10:25am MST
LocationFlagstaff
Human AI Robot Teaming (AI)