A Personal Experiment With Two Sleep Trackers
On a typical night, J. David Creswell goes to bed wearing a tracking watch on one wrist and a different type of tracker on the other. The next morning, while drinking coffee, he compares the sleep data each device produced. It's not just a personal quirk. Creswell directs the Center for Modern Digital Health at Carnegie Mellon University in Pittsburgh, where his team studies how wearables can be used to improve health and performance. His lab has relied on consumer devices for years—for science and for curiosity. And like many users, he wonders how much of what these gadgets report is actually true.
That question is becoming more important as sleep trackers grow more popular. Apple's newest watch, announced on Sept. 9, 2026, gives wearers a single "readiness" score each morning, partly based on how the device thinks they slept. Apple says the redesigned sensors offer the most accurate heart-rate tracking of any wearable, but independent labs have not yet confirmed that claim. With so much riding on these numbers, it's worth asking: can a sensor on your wrist or finger really tell you how your brain slept last night?
The Gold Standard: Polysomnography
To answer that, researchers compare consumer wearables against polysomnography, the clinical gold standard for sleep measurement. Polysomnography combines electroencephalography (EEG) with other sensors to record brain waves, eye movements, and muscle tone throughout the night. Because it directly measures brain activity, it can distinguish not only sleep from wakefulness but also different sleep stages, such as rapid eye movement (REM) and deep sleep.
Wearables, by contrast, do not record brain waves. They infer sleep from proxies like movement, heart rate, and heart-rate variability. That fundamental difference explains both the strengths and the limitations of consumer sleep tracking.
How Accurate Are Wearables at Detecting Sleep?
On the simplest question—asleep or awake—today's devices come surprisingly close to the lab standard. A 2025 meta-analysis of 24 studies involving nearly 800 participants found that wearables typically overestimate total sleep time by about 17 minutes compared with polysomnography. Over an eight-hour night, that error is under 4%, roughly the equivalent of hitting the snooze button twice.
For most people who sleep reasonably well, that margin is small. But the average masks a systematic problem that affects certain users much more than others.
The Blind Spot: Quiet Wakefulness
The gap between wearables and polysomnography comes from a shared blind spot: quiet wakefulness. When you lie still in the dark, awake but not moving, your tracker often mistakes that stillness for sleep. If you fall asleep quickly and stay asleep, you probably don't spend much time in that state, so the error stays small.
But if you have insomnia, the picture changes. People with insomnia often spend long stretches lying awake and motionless, hoping to drift off. Those still, wakeful periods add up, and your wearable likely counts them as sleep. As a result, the device may overestimate how much sleep you're actually getting—sometimes by a wide margin. That can be misleading, especially for someone trying to understand or treat a sleep problem.
What About Sleep Stages? The Brain Activity Gap
When it comes to sleep stages, the limitations are even more obvious. That wearable on your arm or finger does not measure your brain activity directly. Without EEG, it cannot truly know whether you are in REM sleep, light sleep, or deep sleep. Instead, it estimates stages by combining movement data with heart-rate patterns. Those estimates can be useful for spotting broad trends, but they are not a clinical measurement.
This distinction matters because sleep stages are linked to different restorative functions. Deep sleep helps with physical recovery and memory consolidation, while REM sleep plays a role in emotional processing. If a tracker mislabels stages, its reports about "sleep quality" may be less reliable than the raw sleep duration numbers.
Apple's Readiness Score and the Validation Gap
Apple's newest watch adds another layer: a single morning readiness score that blends sleep estimates with other physiological signals. The company says its redesigned sensors deliver the most accurate heart-rate tracking of any wearable, but independent labs have not yet verified that claim. Until peer-reviewed studies test the device against polysomnography, users should treat the readiness number as a rough guide rather than a medical-grade assessment.
This is not a knock on Apple specifically. The broader consumer wearable industry faces a validation gap. Many devices are tested internally or in small studies, but few undergo the rigorous, independent validation that clinical tools require. The 2025 meta-analysis helps fill that gap by aggregating data across many studies, but it also highlights how much variability exists between devices and across different populations.
Key Takeaways From the Research
- Wearables overestimate total sleep time by an average of 17 minutes per night compared with polysomnography.
- In an eight-hour night, that error is under 4%.
- The main blind spot is quiet wakefulness: lying still and awake is often recorded as sleep.
- People with insomnia are most affected because they spend more time in quiet wakefulness.
- Wearables do not measure brain waves, so sleep-stage estimates are inferred from movement and heart rate.
- Apple's new readiness score and heart-rate accuracy claims have not been independently verified.
What This Means for You
Sleep trackers can be valuable tools. They encourage awareness, help people spot patterns, and may motivate better sleep habits. For someone with generally healthy sleep, the 17-minute average overestimate is unlikely to cause harm. But for people with insomnia or other sleep disorders, the data can be misleading. If you regularly feel exhausted despite a tracker reporting seven or eight hours of sleep, that discrepancy is worth discussing with a doctor.
It's also wise to avoid letting a single number dictate your day. A readiness score or sleep duration estimate is just one data point, and it comes with known limitations. The most reliable way to evaluate your sleep remains how you feel and function—not what a wrist sensor says.
Creswell's dual-tracker experiment underscores a simple truth: consumer wearables are impressive engineering, but they are not brain scanners. They can approximate sleep, and they do it well enough for many purposes. Yet when the question is "How did my brain sleep last night?" the answer still requires electrodes, not a wristband.
This article is based on reporting by Medical Xpress. Read the original article.
Originally published on medicalxpress.com
