Empathic Tutoring System: Eye Gaze and Physiological Data for Online Learning

Empathic Tutoring System (ETS) is a prototype that records eye gaze, pupil dilation, heart rate, and skin conductance data from learners in real-time. The system, developed by Wang Hua at the University of Toronto, uses these physiological measures to infer learner attention and motivation, personalizing the behavior of on-screen character agents. Observations from usability studies suggest this approach may heighten learner concentration and lead to more effective learning.

Use Cases

• Training models to predict learner engagement based on physiological measures like eye gaze and pupil dilation.
• Developing adaptive tutoring systems that personalize instructional content based on inferred attentional status.
• Studying the correlation between physiological signals (heart rate, skin conductance) and learning outcomes in digital environments.
• Evaluating the effectiveness of character agents in directing learner attention and providing personalized advice.

Strengths

• Data includes multiple synchronized physiological streams: eye gaze, pupil dilation, heart rate, and skin conductance.
• System reacts to learner data in real-time, enabling study of dynamic interactions.
• Usability studies provide initial evidence of the system's potential impact on concentration and learning.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Row count and file formats are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.

Provenance

Source

University of Toronto (Wang Hua), via paperswithcode.

Collection Method

Data likely collected from a prototype e-learning system during controlled usability studies.