Development and Demonstration of an Integrated EEG, Eye-Tracking, and Behavioral Data Acquisition System to Assess Online Learning

Over the past several years, there has been a rise in online learning platforms and websites for skill acquisition. However, traditional learning analytics to evaluate the effectiveness of presented material are limited in that they do not acquire information regarding the user's biophysical state during this learning process. In this paper, we propose an inexpensive system for evaluating online learners' engagement and performance via electroencephalography (EEG), eye-tracking, and behavioral data methods. Such information is of interest to (i) neuroscientists and psychologists aiming to quantify computerized learning processes and developing models to represent learning states, as well as to (ii) individuals designing educational content to better understand how information is accessed and utilized by users. We first describe the selection, design, and integration of the bio-signal components. We then demonstrate the combined utility of our system through recording signals while participants (N=22) completed German language lessons on the free web-based platform, Duolingo. As low-cost hardware is utilized in our system, data acquisition can readily be scaled to multiple research sites or remote collection, allowing for access to more naturalistic datasets not typically studied using traditional laboratory research systems.