Petting a dog did not speed recovery from stress, but it softened the next test, a 64-owner study finds

Petting a dog for five minutes after a stressful task did not help 64 dog owners recover faster, but it left their brains and hearts calmer during a second stress test. The University of Innsbruck study, published in Biological Psychology, was reported by Medical Xpress on October 22, 2026.

Two stress tests and a dog on a leash

The paper itself went online in August, according to Heilpraxis; the coverage that reached readers came in the second half of October. Medical Xpress opens its report with the old premise: coming home to a wagging tail can ease the stress of a day, and a large body of research acknowledges the well-being benefits of pet ownership. Pet ownership has been linked to better survival after heart problems, less loneliness and greater overall happiness. Yet the mechanisms behind those benefits, the same report says, remain murky, and touch is one candidate among several.

The team around Vanessa Kaufmann, Xiaoqin Cheng and Annett Schirmer at the University of Innsbruck in Austria recruited 64 owners with their own dogs. Fifty-one were women and 13 were men, with an average of 34.1 years. The dogs had to be at least two years old, healthy and at ease in a new room. We could not open the paper itself on ScienceDirect, so the design below comes from three accounts that agree on it.

Each owner sat through two blocks on the same day. In both, a researcher watched by video call and gave negative feedback. The first task was a general knowledge quiz with five seconds to answer each question, the second a countdown in fixed steps at the pace of a blinking red dot that never slowed down.

Then came five minutes with the dog. In the touch block the dog moved freely around the room and the owner could stroke, pet or hold it. In the no-touch block the dog stayed close on a short leash, where the owner could see it and talk to it but not reach it. According to the German health site Heilpraxis, the order of the two blocks was assigned at random.

Electrodes recorded brain activity (EEG) and the heart was followed with an electrocardiogram, which gave heart rate and heart rate variability, the small shifts in time between one beat and the next. Owners also rated how they felt at three points.

It is a strange evening to offer anyone: a stranger on a screen marking your arithmetic down while a dog waits at the end of a leash. The lab made it that way on purpose, so that the only thing that changed between blocks was whether a hand could reach fur.

Recovery was no quicker, the next round was quieter

The first expectation failed. Kaufmann's team had predicted that touch would help the body settle after the first task. Heart and brain measures returned toward resting levels at about the same speed whether owners petted the dog or only sat near it, according to Medical Xpress and the science site Knowridge.

The second round looked different. Owners who had touched their dog in the first break showed a weaker response when the stress came back. Their alpha waves, which the stress test pushes down, stayed higher, their heart rate was lower and their heart rate variability higher, according to Heilpraxis and Medical Xpress.

In the authors' own words: "These results indicate that touching one's dog dampens the bodily impact of an upcoming stressor and that pet owners may naturally seek such touch when interacting with their pet." They go on to describe touch as a possible "important mechanistic pathway through which upcoming threats are experienced as less threatening".

Five minutes of touch bought no faster calm.

The distinction matters for how a reader might use the finding. Comfort in the moment and protection for the next hour are different things, and the data support only the second, and only in this room.

The body noticed what the owners did not say

Owners' own ratings of stress did not always follow their heart and brain readings. ScienceAlert and Medical Xpress both note the mismatch, and ScienceAlert reads it as a sign that some effects of being with a pet may go unnoticed by the person who has them.

Heilpraxis adds a caution about the readings themselves. During the task the heart beat faster, heart rate variability fell and the brain signal shifted, and owners reported more negative feelings, but those feelings were only loosely tied to the size of the bodily changes. A single measure is also no clean sign of stress, because breathing, movement and mental effort can all move it. Read together, the heart and the EEG tell a steadier story than any one of them, and none of them tells what the owner felt.

Nobody in the sources claims this is new for people. Medical Xpress recalls earlier work in which holding hands or receiving a shoulder massage lowered heart rate and eased the body's stress response, and ScienceAlert adds that human touch has been linked to lower cortisol responses. What the Innsbruck experiment tried to isolate is touch itself. In earlier pet studies, as Medical Xpress describes them, participants could pet, look at and talk to an animal all at once, so the benefit could not be credited to any one of those.

That question about attachment is an old one, and the answer here is modest: a hand on a familiar animal, five minutes, a body slightly less braced. Our earlier brief on skin-to-skin contact after birth and maternal depression screens asks a related question about touch, between newborns and their mothers.

A buffer found in a side analysis

The buffering result carries the headline, yet Heilpraxis stresses that it comes from an exploratory analysis, one that looks for extra patterns beyond the main test. A study built to look for it would have to show that it repeats.

Heilpraxis also points to a quieter problem. The touch block differed from the no-touch block in more than touch: the dog was free to move in one and tied out of reach in the other. Not every change, the site argues, can be credited to the hand alone.

The effect appeared as a change inside the group that touched first. For the group that went the other way round, the matching changes were not statistically clear, and Heilpraxis notes that a difference that shows in one group and not in the other does not by itself prove a difference between them.

A weaker bodily response is not automatically a better performance either. And the volunteers were 64 people with their own, familiar dogs, in a short performance test in a lab. Knowridge says the findings do not establish that dog ownership prevents chronic stress or improves long-term health, and that unfamiliar dogs, other animals and people who are uneasy around pets would each need their own studies.

The sources give no minimum duration for petting, and Heilpraxis adds that the animal should be free to move away.

The cats, the oxytocin and the word synchrony

Earlier in October another team asked a neighbouring question about cats. Saho Takagi of Kyoto University and colleagues published in iScience a study of 19 cat owners and their 29 cats, with urine samples tested for oxytocin and for cortisol, the stress marker, according to Sci.News.

Owners with higher oxytocin tended to live with cats that also had higher oxytocin. Owners with higher cortisol tended to live with cats that had lower oxytocin. ScienceAlert says the hormone comparison covered 24 cats, and that 16 of the 19 owners were women.

Takagi told ScienceAlert she thinks the link may come from the petting itself: "oxytocin rises in both the owner and the cat through the petting-and-being-petted interaction, and that this interaction helps to form their bond". She then declines the stronger word. "To call a finding 'synchrony,' I think hormones would need to be measured at multiple time points", she said, and shown to change in a similar way over time. The team compared average levels, not moments. Its exploratory analyses were not preregistered.

Both studies handle touch with the same reserve: small groups, associations and a short lab window, not a proof of a bond. Oxytocin has also been pitched as a fix for marriages; our brief on a trial of the hormone in couples therapy shows how far a hormone can be asked to carry a story.

What the Innsbruck data keep is smaller and sturdier: a body that was touched, even by a dog, met the second hard question a little more quietly. Whether the same holds for other animals, other kinds of stress or people who are uneasy around pets, Knowridge notes, would need studies of its own.