Accelerating Sleep Onset with Wireless EEG Biofeedback

Poster Session D - Monday, March 31, 2025, 8:00 – 10:00 am EDT, Back Bay Ballroom/Republic Ballroom

James Glazer¹, Ken A. Paller¹; ¹Northwestern University

Sleep is essential for human brain function. Yet, many people have occasional difficulty falling asleep and 10-15% meet diagnostic criteria for insomnia disorder. Although sleep is well-studied, many unanswered questions remain about why people sometimes struggle to fall asleep. Electroencephalographic (EEG) studies have linked delayed sleep onset with prolonged high-frequency activity (e.g., alpha and beta) and faster sleep onset with rising low-frequency activity (e.g., theta and delta). Given advances in wearable biosensors, we attempted to improve sleep using brain and body biofeedback in the home over multiple nights. Auditory feedback changed based on real-time data from a consumer sleep-monitoring device (Muse-S); slower brainwaves and decreased movement produced reductions in sound intensity. Participants (N=19) slept as usual on one night, followed by 2 days with 20-min biofeedback sessions each day while awake and again at bedtime while falling asleep, and they were generally able to reduce feedback intensity. Furthermore, participants successfully reduced alpha-theta and beta-theta ratios on biofeedback nights relative to the first-night baseline, with lowest ratios on the final night. Both ratio reductions correlated with reduced subjective sleep latency. These findings need to be confirmed using additional control conditions, given the possibility of order effects between the first night and subsequent biofeedback nights. We are also planning more extensive studies. These preliminary results nevertheless suggest that individuals can learn to modulate brain rhythms and accelerate sleep onset after a few biofeedback sessions. At-home biofeedback could provide an easily scalable approach to supplement other strategies for treating mild sleep difficulties.

Topic Area: METHODS: Electrophysiology