LearnTune: Closed-Loop Music Neurofeedback
Student Researcher · Shing-Tung Yau High School Science Award project
A wearable system that reads a learner’s state and adapts music in real time. I designed and built a custom scalp-EEG headband, fused it with camera-derived heart rate and facial arousal, and ran a within-participant experiment during word learning.
- Designed the headband PCB (dual-channel EEG front end, nRF52840 Bluetooth MCU, 6-axis IMU) and the closed-loop control pipeline: individual calibration, quality-weighted fusion, Kalman filtering and negative-feedback music control.
- Ran controlled experiments comparing closed-loop and static music; coordinated participant recruitment and processed the multimodal data with Python and machine learning (29 participants).
- Under closed-loop feedback, variability of the composite arousal index fell by 12.8% — and I reported honestly that no learning benefit was shown, focusing the paper on system feasibility and measurement limits.
LearnTune: An Exploratory Pilot Study of Multimodal Closed-Loop Music Biofeedback During Word Learning
Co-authored with Ting Jiang, Faculty of Psychology, Beijing Normal University.