"I'm going to take a walk in the next hour": what a smartwatch saw next

Most of us know the gap between meaning to move and actually moving. A study published in Frontiers in Digital Health on November 2 measured that gap hour by hour, for a year.

The team, led by Max Samuelson and Genevieve Dunton at the University of Southern California, used data from the TIME study. It followed 228 adults aged 18 to 29 from across the United States, with an average age of 23.6. All of them wanted to be active, and all used an Android phone. The work was funded by the National Institutes of Health.

One question, asked again and again

For four days every two weeks, over twelve months, participants got short phone surveys about once an hour while awake. Some prompts asked them to answer yes or no to one statement: "I am going to take a walk or exercise in the next hour."

At the same time, they wore a Fossil smartwatch whose accelerometer recorded movement around the clock. The team then compared each answer with how much the person moved in the 60 minutes after it. In total, that gave 20,491 paired moments.

Answers like "trying to decide" or "already exercising" were set aside. Those made up about 47% of the responses to that question.

Plans did predict the next hour

When people said yes, they moved more in the following hour. Without any adjustment, average movement was about 14.3 units per minute after a yes and 9.5 after a no. The link held in 13 of the team's 14 statistical models.

That sounds obvious, but it is worth having in numbers. A momentary plan, measured on a phone, lined up with what a watch recorded an hour later.

The surprise: plans mattered most when people had less to lean on

The team expected that people with stronger self-regulation would turn plans into action more reliably. They tested 14 possible influences, from personality traits to habits like tracking one's own activity, to how focused someone felt in the moment.

Seven of them changed how tightly plans and movement were linked. All seven worked the other way from the team's expectation. The link was strongest when those resources were lower.

For example, at moments when people felt more focused than usual, their plans predicted the next hour less well. People who reported less social support for being active showed a tighter link between a plan and what followed. The authors suggest that a clear intention may act as a substitute route to action when other supports are thin.

What it can't tell us

The authors list several limits. Everyone in the sample already wanted to be active, so the results may not apply to people who don't. The plan was a simple yes or no, with no sense of how strongly it was meant. Mood and focus were each asked with a single question. The many excluded answers could skew the picture. No correction was made for running many tests, and the authors call the work exploratory.

Where Soma fits

Soma is a general wellness product, not a medical device. It doesn't diagnose anything. The Soma app, which isn't available yet, reads what your phone and watch already record through Apple Health, and it doesn't quiz you about your plans every hour. What we like about this study is its design: it puts what people say next to what an instrument records, many times over. Soma Axon, the pod we are still building, uses the same light, cameras and stance every week, and shows each value with its margin of error, so a change over time is easier to trust. We wrote more about that in The same instrument, every week.

Sources, as of November 2, 2026: Samuelson, Wang, Li and colleagues, Frontiers in Digital Health, published under CC BY 4.0. This post describes research and isn't medical advice.