Millions of people use smartwatches and other wearable devices every day to manage stress levels. Researchers say nearly all of these methods are wrong because there appears to be no correlation between physical sensations and digitally recorded data.
A recently published study found "essentially no" correlation between data from wearable sensors and users' self-reported stress levels. However, smartwatches and wearables are not completely useless if users fully understand the limitations of the devices.
The Dutch study tracked stress, fatigue and sleep data on 800 young people for three months using Garmin Vivosmart 4 activity trackers. Researchers asked participants to report their levels of stress, fatigue or sleepiness four times a day.
When the researchers cross-referenced the data, they found little correlation between the data and the participants' personal experiences. None of the 800 subjects' "stress scores" on the tracker matched their perceived stress levels. Additionally, a quarter of the participants felt stressed or relaxed, while their smartwatches measured the opposite.
Co-author Eco Fried noted that the study's results were not surprising. The Garmin watch they used measured heart rate, and that specific metric has nothing to do with a person's actual mood. For example, heart rate during sexual arousal is often as high as when angry.
"These findings raise an important question about what wearable data can or cannot tell us about mental state," Fried warned, "and you should be careful not to rely on smartwatches — these are consumer devices, not medical devices."
The study also examined the reliability of Garmin's "body charge" feature, which is designed to measure body fatigue. The correlation between this data and actual personal experience is stronger than the correlation with mental stress, but it is still too weak to correlate with actual experience. Garmin doesn't explain how the body charge score works, but researchers suspect it's a combination of pulse measurements and physical activity levels.
While smartwatches are of little use as stress gauges, they can do a much better job of measuring sleep. The devices tested helped measure sleep duration, although they gave little indication of how rested a person was after waking up.
The researchers believe this type of sleep-related data could help developers create an "early warning system" for states of depression. When the watch predicts an impending onset of depression, it alerts users so they can take early action or get the most appropriate treatment.
Margarita Panayiotou, a researcher at the University of Manchester, said: "Wearable data can provide valuable insights into people's emotions and experiences. But it is crucial to understand its potential and limitations."