For a long time, checking brain health meant sitting through tests, answering long questionnaires, or going to clinics. These methods take a lot of time, can feel stressful, and don’t happen often enough to notice small, everyday changes.
But a new study from the University of Geneva suggests that the future of monitoring cognition and emotional well-being may be as simple as wearing a smartwatch or carrying a smartphone.
Researchers working on the Providemus alz project followed 82 healthy adults for ten months, collecting streams of passive data from consumer-grade wearables and mobile devices. These included everyday measures such as heart rate, sleep patterns, physical activity, and environmental factors, such as air quality and weather. Every three months, participants also completed traditional “active” assessments, questionnaires, and cognitive tests, to provide benchmarks.
The twist came when artificial intelligence was applied to the passive data. Could it predict the ups and downs of participants’ emotional and cognitive states? The answer was a resounding yes.
Wearable devices are less likely to be used by patients with cardiovascular disease
Igor Matias, a doctoral assistant at the Geneva School of Economics and Management, explained: “Once data collection was complete, the passive data were analysed using artificial intelligence developed as part of the project. The aim was to determine whether AI could predict fluctuations in participants’ cognitive and emotional health based on these data.”
The results were striking. On average, predictions had an error rate of just 12.5%. Emotional states were predicted with surprising precision (error rates of 5% to 10%), whereas cognitive states were more challenging, with error rates of 10% to 20%.
Put another way, AI was better at predicting participants’ responses to emotional questionnaires than cognitive performance tests.
The study also found strong passive metrics: air pollution, weather, daily heart rate, and sleep variability were important predictors of cognition. For emotional states, weather and sleep variability, as well as heart rate during sleep, had the most effect.
This is intuitively plausible: sleep deprivation or stressful environmental circumstances can, in subtle ways, influence how we think and feel, even when this is not on our radar as we go about our lives. Connected devices can monitor these influencers continuously and provide a window into brain health that traditional methods often fail to capture.
Smartwatches could help people give up smoking
The implications are enormous. Easy, everyday tracking could help doctors and researchers notice early signs of memory problems or emotional struggles before they get serious. For healthy people, it’s also a way to understand how daily habits and surroundings, like sleep, stress, or weather, affect mood and mental sharpness.
Rather than depend on episodic check-ups, brain health could be monitored in real time, and in life. And because the study employed consumer-grade devices, the approach is scalable and can extend to millions of people without specialized hardware.
This study doesn’t aim to replace standard brain health check-ups, such as tests or questionnaires; it’s meant to work alongside them. The exciting part is that it could make early detection much easier.
As Igor Matias explained, the results “open up new possibilities for the use of connected devices in the early detection of abnormalities or changes in brain health.”
Journal Reference:
Matias, I., Haas, M., Daza, E.J. et al. Digital biomarkers for brain health: passive and continuous assessment from wearable sensors. npj Digit. Med. 9, 197 (2026). DOI: 10.1038/s41746-026-02340-y