A Learning Attention Monitoring System via Photoplethysmogram Using Wearable Wrist Devices

Abstract

Many studies have proved that sustained attention strongly affects student learning performance. However, it is challenging for teachers to perceive the attention level of every student in time, especially in online learning environments and large-scale physical classes. In this study, we propose a method for assessing students' attention levels that uses a photoplethysmogram (PPG) signal measured by a wearable device, with an average accuracy of 0.69. A novel attention monitoring system was developed for both teacher and student users in a massive open online course (MOOC) environment that provides a convenient interface for visualizing the real-time attention levels of both individual students and groups of students. In order to verify the reliability of the system, an empirical study was conducted in the MOOC environment for a preliminary exploration of the relationship between attention states and learning content. The results of this study provide a valuable resource for selecting and designing video lectures for online learning. The system can also be extended to provide appropriate adjustments in both classroom and ubiquitous learning environments.