SleepCoacher: A Personalized Automated Self-Experimentation System for Sleep Recommendations
summary
SleepCoacher is a closed‑loop smartphone system that collects accelerometer and microphone data via a modified Sleep as Android app, computes sleep variables (onset latency, awakenings, self‑reported restfulness), selects from 114 recommendation templates authored with clinicians, and runs guided single‑case ABAB experiments (5+ nights per phase). Two studies including a three‑week trial with 17 undergraduates show improvements correlate with adherence: users who follow ≥65% of recommendations improve target metrics; system and code are available online.
AI-written summary · about
chapters (7)
- 0:00Introduction to SleepCoacher
- 2:35How SleepCoacher works
- 7:51Self-experiment design
- 12:06User study results
- 15:55Key findings and limitations
- 16:36Future 10-year sleep study
- 17:19Q&A session
related talks
44mYour Health Journey With Samsung Health
25mDesigning around people’s lives to unlock sustainable growth
07mDetecting Receptivity for mHealth Interventions in the Natural Environment
42mDesigning Compassionate Systems
19mKeeping Users Engaged: A Long-term Study of Wearable-based, Gamified Exercise Applications
20mQuantifying the Changeable Self: The Role of Self-Tracking in Coming to Terms With and Managing ...
12mScaling Context-Aware Task Assistants that Learn from Demonstration and Adapt through Mixed-Initi...
23mPersonal OS: Build Your Army of 100 AI Interns