What the walk actually moved
Nearly a million hours of recorded movement, matched against 321,345 moments of people saying how they felt right now. The clearest finding was about energy. The most useful one was about what movement didn’t touch.
“Go for a walk.”
It’s the most-offered piece of wellbeing advice there is, and most of us have both given it and quietly resented receiving it. The resentment tends to come from the same place: a walk arrives sounding like an answer to something it can’t possibly answer. You’re not tired. You’re not restless. You’re sad, or worried, and a walk is not an argument against either.
A very large study suggests that resentment is, in a narrow and specific way, correct. And that the walk is still worth taking.
What they did
A team led by Johanna Rehder at the University of Salzburg, with Markus Reichert as senior author, pooled 67 separate datasets from 14 countries — around 55% of the eligible studies they could identify, gathered up to the end of 2023. Between them, those datasets covered 8,223 people, average age 36, who had worn accelerometers for anything from one day to a hundred days while their phones pinged them at intervals to ask how they were feeling at that moment.
The result is a mountain: nearly a million hours of objectively measured movement, matched against 321,345 momentary wellbeing ratings. It was published in Nature Human Behaviour.
The design matters as much as the size. Because the team had the raw participant data rather than other people’s published summaries, they could separate two questions that usually get tangled together. One: do people who move more tend to feel better than other people? Two: do you feel better at the moments when you’ve been moving more than your own usual?
Only the second one is any use to a person. Active people might be happier for a dozen reasons that have nothing to do with the activity — better health, more money, more company. Comparing you to your own average strips most of that out. They also ran it in both directions: does movement come before a shift in feeling, and does feeling come before a shift in movement?
What came back
Within people, movement tracked most strongly with energy — what the researchers call energetic arousal, the sense of being awake and alert rather than flat. Going from sitting to walking was associated with a 0.62-point rise on a four-point energy scale. That’s a noticeable difference, and it was comfortably the largest effect in the study.
Positive feelings shifted too, but less: 0.24 points for the same comparison. Overall pleasantness, 0.16.
And then the finding we haven’t been able to put down. Negative feelings showed no consistent relationship with movement. Not in either direction. Sadness, worry, low mood — they didn’t reliably lift after activity, and they didn’t reliably predict whether someone moved later on either.
Movement also went along with feeling less calm, which sounds like bad news until you notice it’s the energy finding wearing a different coat. Alert and settled are not the same state. Being switched on isn’t the same as being soothed.
Then the loop closes. When people felt better and more energetic than their own average, they tended to be more active afterwards. When they felt calmer, they tended to be less active afterwards. Feeling predicted movement about as reliably as movement predicted feeling.
Our take
The honest reading here is two dials, not one.
Movement, on this evidence, has a fairly dependable hand on the energy dial and a lighter hand on the positive-feeling dial. It does not appear to have a hand on the hard stuff at all. That isn’t a disappointing result. It’s a precise one — and precision is kinder than encouragement.
Because the promise smuggled inside “just go for a walk” is that movement is a lever on sadness. That sets people up to fail twice: once when the walk doesn’t work, and again when they conclude the failure was theirs. A million hours of data say the walk was never the right tool for that particular job. It was doing something else, quite reliably, the whole time.
And what it was doing isn’t nothing. Energy is what the next hour gets made out of. When you’re flat, the distance between flat and alert is often the distance between an evening that simply happens to you and one you have some say in. Moving doesn’t argue with the sadness. It sometimes hands you a little more to carry it with.
The loop is the other half of it. If feeling slightly better makes movement more likely, and movement makes energy more likely, then the picture isn’t a lever you pull. It’s a slow mutual nudge, running both ways, all day, mostly beneath notice. Which means the leverage was never in any single walk. It’s in knowing roughly where you are in the loop.
That’s the part sitting closest to why we built Nagi the way we did. The check-in asks about mood, sleep and energy separately — four taps, three dials — because they are separate, and a study this size is a decent argument that folding them into one number loses something real. Some days the energy is on the floor and the mood is fine. Some days it’s the other way round, and the thing that helps one won’t touch the other. Nagi doesn’t fix either. It gives you a quiet, private place to put them down often enough that a pattern can eventually show itself.
The caveats, plainly
This is observational. All 67 datasets watched people rather than assigning them to anything, so none of it establishes that movement causes the energy — only that the two travel together, closely and repeatedly. The studies used different devices, different questions and different definitions of wellbeing. And the overall correlations are small in statistical terms — roughly r = 0.04 to 0.08 — though the authors note they sit on a par with the effects of other ordinary daily activities. Nobody here has found a big lever. They’ve found a small one that is genuinely, measurably there.
The size of the effect also varied a great deal between individuals, with age, body size, sex and even whether it was a weekday explaining part of that variation. Which is another way of saying: the average is not you.
Where that leaves us
Two dials. Move the one that moves, and stop measuring the walk against the dial it was never going to reach.
If the heaviness is still there when you get home, that isn’t a failed walk. It’s a walk that did the thing walks do — and a heaviness that needs something else. Time, usually. Company, often. Sometimes someone qualified to sit with it, which is a perfectly ordinary thing to need.
If you want a private place to start noticing your own patterns, Nagi is on Google Play and the App Store — or in your browser at app.nagiandme.app. The first 15 messages are free. No name, no email. Just you, checking in.
— The Nagi team
Source: Rehder, J., Timm, I., Berretz, G., et al. “An individual participant data meta-analysis of how physical activity relates to affective well-being in daily life”, Nature Human Behaviour, 2026;10(7):1297–1315. DOI: 10.1038/s41562-026-02427-2. Reported by PsyPost, 17 September 2026.
Nagi is a mental health companion, not a medical device, and nothing here is medical advice. If you’re struggling, please reach out to a professional or someone you trust.