Sensor-based cognitive state assessment in a mobile environment

Inferring cognitive state from non invasive neurophysiological sensors is a challenging task even in pristine laboratory environments. Artifacts ranging from eye blinks, to muscle artifacts and electrical line noise can mask electrical signals associated with cognitive functions. These concerns are particularly pronounced in the context of the Honeywell team’s ongoing efforts to realize neurophysiologically driven adaptive automation for the dismounted ambulatory soldier. Besides the typical sources of signal contamination, the Honeywell team has to deal with the effects of artifacts induced by shock, rubbing cables and gross muscle movement. This paper presents the Honeywell team’s efforts to make reliable sensor based cognitive state assessments given the constraints just cited. Cognitive state classification results suggest that it is feasible to classify cognitive state in ambulatory, military-relevant task contexts