Monitoring attention span dynamics in online classes using electroencephalography

Abstract

This study investigates the utilization of electroencephalography (EEG) for monitoring attention dynamics in online classrooms. With the widespread integration of virtual learning due to the impact of COVID-19 pandemic, managing student engagement has become increasingly crucial. Through EEG technology, educators can gain access to real-time data on fluctuations in student attention levels, facili- tating timely adjustments in teaching methodologies. By gathering EEG data from students participating in various online learning activities including live lectures, interactive discussions, and multimedia presentations, and analyzing these signals, we can track fluctuation in attention. A multimodal classification method is used and different classifiers such as Bi-LSTM, Bi-GRU, Multi-Head Attention Mecha- nism, Vision Transformer (ViT), Contrastive Learning etc are used to detect the attention levels efficiently. This allows us to identify periods of sustained focus as well as moments of distraction. Additionally, we investigate how factors such as lec- ture style, instructor engagement, and external distractions impact attention span dynamics. Consequently, our research provides valuable insights into the cognitive mechanisms governing attention in online learning, offering a promising approach for real-time monitoring and intervention to create a more engaging online educational experience.