Towards an index of cognitive efficacy EEG-based estimation of cognitive load among individuals experiencing cancer-related cognitive decline

This paper describes an effort to estimate variations in cognitive effort among cancer survivors experiencing treatment related cognitive decline. EEG-based cognitive state sensing algorithms were validated in the context of an experiment with 5 brain cancer and 5 breast cancer survivors. Workload was manipulated by varying text complexity and time pressure. Analysis indicates that EEG-based cognitive state sensing algorithms were able to distinguish between high and low cognitive workload with an average classification accuracy of 0.84. Results suggest that 5 to 10 channels of EEG can provide enough information to achieve classification accuracies exceeding 0.80. The highest density of informative sites were over the left temporal and mid to inferior frontal regions in the left hemisphere - regions that play a major role in language.