Performance Comparison of Automated EEG Enhancement Algorithms for Mental Workload Assessment of Ambulant Users

Abstract

Mental workload (MW) assessment is important for numerous mentally-demanding applications, including first responders, air traffic control, amongst others, as it quantifies the cognitive capabilities of the operator. Recently, there has been a push for wearables based MW monitoring for real-time feedback and human performance augmentation. Most previous studies have focused on immobile subjects. Realistic applications, however, rely on ambulant users under varying types and levels of physical activity. Movement artifacts are known to hamper the quality of signals measured by wearable devices, thus the impact on MW assessment in situ is still unknown. In this study, we compare the performance of several automated artifact removal algorithms for electroencephalograms (EEG), as well as the robustness of two classical feature sets, for MW assessment under varying physical activity levels.