Construction of a Quality Evaluation System for University Course Teaching Based on Multimodal Brain Data

Background: The objective evaluation of course teaching quality is not only a guarantee for high-quality university teaching but also a key basis for teaching reform in higher education institutions. Objective: This study explores a novel method to enhance the objectivity and generalizability of teaching quality evaluations and provide more accurate and comprehensive results for teachers and administrators. Unlike traditional post-quantitative evaluations based on student scores, peer reviews, or supervision feedback, and post-qualitative research relying on subjective data, this approach leverages the integration of “Brain Science and Education,” as well as “AI and Education.” Method: Using EEG and eye-tracking data from students, this study focuses on attention as a core variable. It begins with single-factor analysis (EEG or eye movement) and progresses to a multimodal data fusion system to construct an objective evaluation model. Results: After eliminating environmental interference, a strong correlation is found between EEG data, eye movement, and attention. While EEG reflects brain activity, attention judgment often requires self-reports, introducing bias. Eye-tracking offers real-time behavioral indicators but lacks physiological depth. Therefore, integrating EEG and eye movement data compensates for the temporal/ spatial limitations of single-source analysis. Integrating AI and computing technology maximizes the academic and practical value of brain data and supports more effective post-hoc evaluation integration. Conclusion: This approach enables student-centered, personalized course evaluations and targeted teaching reforms, promoting continuous improvement in higher education and the holistic development of students.