A new pilot study from the Swedish University of Agricultural Sciences and Uppsala University, published in PLoS One on April 22, 2026, ran an experiment that should make most dog owners feel a little better about themselves. The researchers strapped accelerometers on dog-and-owner pairs for two weeks, ran them through a joint outdoor exercise program (jogging plus circuit training), and then strapped accelerometers on them again.
The headline finding is what the researchers learned about the gap between how active dog owners think they are and how active they measurably are.
Dog owners systematically underestimated their own physical activity, especially the moderate-intensity kind. Self-reported activity was lower than what the accelerometer recorded. The authors interpret this in a way I found genuinely useful: dog walking is so woven into the daily routine of dog owners that they no longer count it as exercise.
If you have ever filled out a fitness app form that asked "how many minutes of moderate activity do you get per day," paused, decided your morning loop with the dog "doesn't really count," and entered zero, this paper is about you.
What the study actually did
The study, by Smedberg and colleagues, is a pilot, with all the caveats that label deserves. There were 15 dog-and-owner pairs. There was no control arm. The intervention was two weeks of structured outdoor exercise: jogging sessions and circuit training with the dogs. Body measurements were a secondary outcome. Sleep was tracked alongside activity.
The numerical results, before we get to the interpretation, were:
Vigorous-intensity physical activity in owners increased by 5 minutes per day (statistically significant, p = 0.04).
Sedentary time in owners decreased by 41 minutes per day (statistically significant, p = 0.01).
The dogs' overall activity patterns did not change much, although their median body condition score improved slightly but significantly.
Both dogs and owners spent the bulk of their daytime hours sedentary, even after the program.
Sleep was satisfactory in both species, both before and after.
That last bullet is worth dwelling on. The intervention pushed the humans toward more vigorous activity and less sitting, but it did not measurably change how much the dogs moved. Roger, my own dog, would not be surprised. He has views on jogging.
The accelerometer-versus-survey gap
Here is where the paper becomes interesting for anyone using a fitness tracker, a smart collar, or a dog-walking app.
When the researchers compared what the accelerometers actually recorded against what the dog owners said about themselves on standard physical-activity questionnaires, the questionnaires came up short. People underreported moderate-intensity activity in particular. The authors infer that this happens because dog walking gets reclassified, in the owner's head, from "exercise" to "errand."
The implication is interesting in two directions.
For pet parents, the accelerometer is doing you a favor. If your tracker tells you that you hit your activity goals while you "just took the dog out," it is not flattering you, and your self-assessment is probably the part that is wrong. Dog walking is real activity, even when it does not feel like it.
For the people building these products, the gap is a design problem. A consumer wearable that measures the same thing the user already discounts in their self-perception is going to lose some of the engagement battle to apps that frame movement as a choice the user made on purpose. There is a reason fitness culture sells you a class with a name and not a daily five-block walk to a fire hydrant.
What the dogs got out of it
The honest answer the paper gives is: not as much as the humans did. Dog activity patterns did not change in any major way. The slight reduction in body condition score (the canine equivalent of a body composition score, on a 1-to-9 scale) was real but modest.
There are several reasonable interpretations.
The first is sample size. Fifteen pairs is small enough that a real effect on dog activity could easily be missed. The authors flag this directly.
The second is the dogs were already pretty active. If a dog is being walked daily by an owner who is itself meeting global physical activity recommendations (which the owners in this study were, even at baseline), there is not as much room for the dog to gain.
The third is more sobering. The authors note that both dogs and their owners spent most of the daytime sedentary, even during the intervention. Two weeks of jogging plus circuit training, three or four times a week, is real exercise. It is not enough to overcome being on the couch the rest of the time.
That third finding is the most important pet-parent takeaway in the paper. The shape of the modern dog-and-human household is sedentary by default. The walks happen, then everyone goes back to sitting.
What this should change about how pet parents read their tracker data
A few practical reframes, none of them dramatic, that follow directly from the study:
If your activity tracker says you got more activity than you remember, believe the tracker. Especially on dog-walking days.
If your dog's activity tracker shows a flat line on most weekdays except for the morning and evening walks, that is the actual shape of pet life in 2026, and it is a reason to add play sessions in the middle of the day, not a reason to assume the tracker is broken.
If you are using a "joint" pet-and-owner app that pools the two together as one wellness graph, ask what each line is actually measuring. Owner activity tracked by accelerometer is not the same construct as dog activity tracked by collar.
I will not pretend that any of this is going to alter the structural sedentariness of how most of us live. The Smedberg paper does not claim that either. What it does claim, and what it shows pretty clearly, is that the gap between what dog owners say they do and what they actually do is in our favor for once.
