Sensor Measures of Affective Leaning.

Area of Science:

• Educational Psychology
• Cognitive Science
• Human-Computer Interaction

Background:

• Self-regulated learning is crucial for academic success.
• Accurate, real-time measurement of learning states is challenging.
• Wearable sensor technology offers potential for unobtrusive data collection.

Purpose of the Study:

• To predict self-report data for self-regulated learning using sensor data.
• To evaluate the efficacy of consumer-grade electroencephalography (EEG) and electrodermal activity (EDA) sensors.
• To explore the relationship between psychophysiological data and learning outcomes.

Main Methods:

• Longitudinal study with 100 university students.
• Collected self-report data (questionnaires, experience sampling) and sensor data (EEG, EDA).
• Applied machine learning models to predict learning states and outcomes.

Main Results:

• Consumer-grade EEG and EDA sensors failed to predict embedded experience sampling.
• EDA data was ineffective in predicting outcome measures.
• EEG activation at T7 predicted lower intrinsic motivation, aligning with PSI theory.

Conclusions:

• Consumer-grade wearables face technical challenges for accurate learning state prediction.
• Advanced sensor technology may enable unobtrusive tracking of affective learning.
• Potential for supporting adaptive digital learning environments exists.