Applying learning analytics approaches to detect and track students' cognitive states during virtual problem-solving activities

A virtual problem-based learning (PBL) environment can generate large amounts of textual or timeseries usage data, providing instructors with opportunities to track and facilitate students' problemsolving progress. However, instructors face the challenge of making sense of a large amount of data and translating it into interpretable information during PBL activities. This study proposes a learning analytics approach guided by flow theory to provide teachers with information about middle schoolers' real-time problem-solving cognitive states. The results indicate that the hidden Markov model (HMM) can identify students' specific cognitive states including flow, anxiety, and boredom state. Based on the findings, a teacher dashboard prototype was created. This study has demonstrated the promising potential of incorporating the HMM into learning analytics dashboards to translate a large amount of usage data into interpretable formats, thus, assisting teachers in tracking and facilitating PBL.