Optimizing Human-Automation Team Workload through a Non-Invasive Detection System

We propose to investigate, in collaboration with MGH Voice Center and Altec, Inc., application of surface electromyography (sEMG) to assessing cognitive workload, strain, and overload. Specifically, sEMG sensors placed on the face and neck will detect emotional/motor responses to workload strain. The proposed effort will build on the substantial sEMG experience of our partner, MGH (including research on vocal/subvocal speech recognition) as well as Altec’s foundation of unobtrusive, wireless sEMG sensing/signal processing equipment and our state-of-the-art technology for real-time human state assessment—ultimately resulting in a robust, reliable, highly sensitive model for estimating workload strain and detecting and predicting cognitive overload in real time. The system, called Empathy, will employ a complement of cost-effective, unobtrusive psychophysiological sensors designed/chosen to minimize artificial constraints on operator behavior, as well as utilizing additional sources of evidence reflecting an operator’s internal state (e.g., attention dynamics, physical performance, status of the automation involved). Phase II will build on the analytical/theoretical progress of Phase I by empirically validating the utility of sEMG for this application. A complete, integrated prototype will be developed, tested in a laboratory setting under operationally relevant conditions, and delivered to DARPA/AFRL for further testing/evaluation by the end of the Phase II Option.