Real-Time Cognitive Load Measurement in Classroom Environment Using a Dry-Electrode EEG System

Cognitive load (CL) theory suggests that instructional materials need to be designed for reducing unnecessary CL and has been regarded as one of the most influential theories in science education. How to measure individual CL is still under investigation. In this study, we developed an eight-channel dry-electrode electroencephalogram (EEG) system and proposed an algorithm to real-time measure the depth of working memory of the N-back task in a classroom environment. The ocular artifact was removed by using the recursive least-square (RLS) method. Time-frequency analysis was applied to extract event-related theta- band activities in the artifact-suppressed EEG signals. Eight participants had the active duration for theta-band activities as 1.44±0.36 mv, 1.70±0.22 mv, and 1.97±0.04 mv for 0-back, 1-back, and 2-back tasks, respectively. In contrast to the previous research that has used spectral power of particular frequency bands as signal features, we found the detection of active duration provides better discrimination power in classifying different CL levels, compared to that of the classification using features of spectral power. The result in this study demonstrates the feasibility of theta-band EEG as an indicator to measure students’ cognitive load in a classroom environment.