Developing cognitive workload and performance evaluation models using functional brain network analysis

This study developed cognitive workload and performance evaluation models using Electroencephalogram (EEG) data from participants performing cognitive tasks on a robot simulator. Functional brain network and spectral analyses were used to extract features from EEG data recorded from 124 brain areas. Linear models were developed using these features, performance scores, and subjective workload assessments (SURG-TLX). Results showed that combined EEG features can evaluate cognitive workload and performance. A correlation was found between age and cognitive workload in the more challenging task, suggesting challenging computer games may help identify age-related cognitive workload changes. The findings offer potential for objective cognitive workload and performance evaluation and monitoring.

Saeed Shadpour, Ambreen Shafqat, Serkan Toy, Zhe Jing, Kristopher Attwood, Zahra Moussavi, Somayeh B. Shafiei