Multimodal Neurophysiological Assessment of Cognitive Load in Children during Math Problem Solving

• In Event: Innovative Cognitive Approaches to Learning and Assessment

Abstract

While assessing learners’ cognitive states through neurophysiological signals holds promise for advancing personalized learning, this study investigated how cognitive load can be assessed based on multimodal neurophysiological signals in educational contexts. A two-phase within-subjects experimental design was employed, involving synchronous EEG and eye tracking recordings as 82 children performed math arithmetic tasks varying in complexity. Five supervised machine learning models were trained and evaluated. Results indicated that the models achieved reasonable predictive capabilities, with Logistic Regression and Support Vector Machine classifiers reaching accuracies of 77.21% and 75%, respectively. These findings highlight the potential of multimodal neurophysiological signals for real-time cognitive load assessment in ecologically valid learning environments, informing the development of adaptive educational technologies.

Authors

• Jialing Zeng, Peking University

• Renzhe Yu, Teachers College, Columbia University

• John B. Black, Teachers College, Columbia University