Electroencephalography (EEG) classification of cognitive tasks based on task engagement index

Abstract

The application of autonomous systems is on an increase, and there is the need to optimize the fit between humans and these systems. While operators must be aware of the autonomous systems dynamic behaviors, the autonomous systems must in turn base their operations, among other things, on an ongoing knowledge of operators' cognitive state, and its application domain. Psychophysiology allows for the use of physiological measurements to understand an operators behavior by noninvasively recording peripheral and central physiological changes while the operator behaves under controlled conditions. Electroencephalography (EEG) is a psychophysiological technique for studying brain activation. In the present study, EEG task engagement index, defined as the ratio of beta to (alpha + theta), are used as inputs to an artificial neural network (ANN) to allow identification and classification of mental engagement. Six separate feedforward ANN with single hidden layer trained by backpropagation were designed to classify five mental tasks for each of six participants. The average classification accuracy across the six participants was 88.67 %. The results show that differences in cognitive task demand do elicit different degrees of mental engagement, which can be measured through the use of the task engagement index.