Sustaining Vigilance by Activating a Secondary Task When Inattention is Detected

Abstract

Vigilance tasks, from driving a vehicle to surveillance to security monitoring, are both commonplace and high-stakes. Yet users have well known difficulties sustaining vigilance. We evaluate the ability of an augmented cognition closed-loop attention management (CLAM) system to sustain vigilance and task performance by monitoring operator's psychophysiology, detecting inattention, and activating a countermeasure when inattention crosses a threshold. Eighteen participants performed a vigilance task and were monitored for inattention via a combination of eye, head, and electroecephalographic (EEG) measures. A cognitively demanding secondary task was activated either when inattention was detected or randomly throughout a 40 minute session. While participants in both conditions demonstrated a vigilance decrement, as measured by an increase in misses over the course of the session, the CLAM condition produced 17% fewer misses overall than the random condition. This improvement was not due to the countermeasure, per se, but to the timing of the countermeasure to participant's detected inattention. The advantage for a tailored presentation of the secondary task is noteworthy because prior evaluations of continuous secondary tasks demonstrated degraded vigilance performance. The results inform our understanding of how human vigilance operates and the technology for its detection and manipulation.