Characterization of the Operator Cognitive State Using Response Times During Semiautonomous Weapon Task Assignment

Abstract

The increase in autonomy of unmanned systems is projected to continue in the foreseeable future. As a result, a single operator may be expected to monitor and control several unmanned systems. The US Air Force’s Warfighter Interface Division conducted a simulation and testing with human in the loop. Their objective was to examine target acquisition performance for unaided human operators with that of an automated cooperative controller in accomplishing a complex task involving the prosecution of ground-based targets with wide area search munitions (WASMs). Their experiments provided empirical data on a humans ability to manage multiple tasks with varying mental task difficulty. The concept of the response time (RT) is often used by psychologists and neuroscientists to better understand cognitive and corresponding neural processes. Recently, the RTs have been shown to be correlated with the cognitive task difficulty. Based on these findings, a new approach for assigning task difficulty during multitasking experiments is introduced. The approach constructs a monotonously increasing mapping of the operator’s RT data into the corresponding task difficulty levels with a continuous range of values.