Can Dogs Ease Pain? What Research on Social Closeness Shows

In brief

Social closeness can influence how pain is experienced. A study from Humboldt University of Berlin (Mauersberger et al., 2024) found in an experimental pain test that the presence of a dog was associated with lower pain perception than the presence of human companions. The more surprising part of the finding is rarely cited: in the first study, having a friend present did not help at all compared with being alone.

An important point for context: this is about influencing the experience of pain in the lab, not about healing and not about clinical pain. There is also a randomized study that found no effect at all in healthy participants. Dogs are no substitute for medical diagnosis, pain management, or necessary treatment.

Why does the same injury hurt more on one day than on another? Why do two people experience the same stimulus in completely different ways? Pain is far more than a signal from the body—and this is exactly where the dog’s role becomes interesting. Rather than the oversimplified question “Can dogs relieve pain?” it pays to ask a more precise one: How can a dog’s presence influence how we experience pain?

1. Pain is more than a physical signal

1.1 What lies between stimulus and experience

In modern pain research, pain is not regarded as a one-to-one translation of tissue damage into sensation. Between stimulus and experience lies processing in the brain, which brings together sensory, emotional, and cognitive components. Put simply:

Pain signal + attention + expectation + emotion + social environment → experienced pain

1.2 Why company doesn’t automatically help

Social closeness is one of these factors as well—but it is not always a relief. Krahé and colleagues (2013) systematically reviewed the evidence and reached a conclusion that fits well with the rest of this article: social support reduces the experience of pain when it acts as a reliable safety signal for the person concerned. If, on the other hand, the companion is perceived as a possible threat—for example, because you feel watched or judged—the effect can reverse (10.3389/fnhum.2013.00386).

This is exactly where the question about the dog comes in: many people don’t feel judged by a dog; it expects no explanation and demands no composure.

1.3 What this article does not claim

It does not claim that dogs relieve pain. It claims that a dog’s presence can influence how pain is experienced and expressed—and that this claim is better supported in the lab than in the clinic.

2. The Berlin study in detail

2.1 The design

The team led by Heidi Mauersberger (Humboldt University of Berlin) used the cold pressor test, an established method in pain research: participants immerse one hand in water at 2 to 3 °C and keep it there for as long as they can. The researchers recorded subjective pain, pain tolerance, facial muscle activity, skin conductance, and feelings of helplessness (10.1016/j.actpsy.2024.104418).

• Study 1 (74 women, all pet owners): The test was done alone, in the presence of a same-sex friend, or in the presence of the participant’s own dog. The companion sat about four meters away, silent, with no interaction.
• Study 2 (50 women): Each participant went through all three conditions—alone, with an unfamiliar person, and with an unfamiliar dog. This time with active support: participants were allowed to touch the dog’s head or the person’s hand. Both were introduced as caregivers.

2.2 The finding that is rarely cited

In Study 1, participants with their dog reported and showed less pain, held out longer, and felt less helpless than when they were alone. The truly remarkable part: there was no substantial difference between the friend condition and the alone condition. So the friend did not help—but the dog did.

One obvious explanation is ruled out. The authors had two blinded raters analyze videos of the friends and the dogs and found that the friends looked at the participant more often and communicated more than the dogs did. So it was not that the dogs had been more attentive.

2.3 What Study 2 adds—and where it sets limits

With unfamiliar, actively supportive people, the picture looked different: here the human helped as well. Overall the dog effect remained larger, but with an important caveat that usually gets lost when the finding is passed along.

For reported pain ratings, negative affect, and the increase in heart rate, the dog’s advantage depended entirely on how the participant felt about dogs. Those who liked dogs benefited. Those who were more reserved toward them did not—and for heart rate and negative affect the pattern even reversed: these participants showed more stress in the dog’s presence than in the presence of the unfamiliar person.

For expressed pain behavior, sense of coping, and positive affect, by contrast, the dog’s advantage held regardless of attitude.

It is also relevant that all participants were screened for fear of dogs beforehand. Anyone who felt uncomfortable around dogs was not included. The range of attitudes therefore ran from “likes dogs” to “is rather indifferent to dogs,” not to “is afraid of dogs.”

2.4 Two clarifications

• It does not mean “dogs work better than people.” What is correct: in this lab experiment, with this pain stimulus, and compared with these particular companions, the presence of a dog showed stronger effects. The paper’s title—“Pet dogs succeed where human companions fail”—is considerably more pointed than what was measured.
• In Study 1, the dog didn’t have to do anything. No petting, no reassuring words, no interaction—just being there.

3. Why might a dog influence pain?

3.1 Social safety without judgment

Human support is often ambivalent: when someone keeps you company while you are in pain, you may feel watched, feel you have to hold it together, or not want to appear weak. Many people, by contrast, don’t feel judged by a dog.

Contrary to what is often assumed, in the Berlin study this did not remain a mere hypothesis. In Study 2, perception was assessed directly, and the participants experienced the dogs as less judgmental, as closer, and as more empathic and supportive than the human companions—despite the humans’ active attention. What this does not settle is whether this perception actually causes the pain effect or merely accompanies it.

3.2 Attachment and social regulation

Certain forms of human–dog interaction can be accompanied by changes in the oxytocin system. Nagasawa and colleagues (2015) reported that longer mutual gazing between dog and owner was associated with higher urinary oxytocin concentrations in both (10.1126/science.1261022).

That does not mean “oxytocin reduces pain.” Oxytocin research provides a mechanism for social regulation, but it does not measure pain. The bridge from there to the experience of pain is a plausible inference, not evidence. For more, see Oxytocin in Dogs: Bonding, Trust, and Stress Regulation.

3.3 The stress–pain connection

Stress can influence how pain is processed and appraised, for example through attention and emotional appraisal. That a dog’s presence can reduce the experience of stress has also been observed outside pain studies: Polheber and Matchock (2014) found lower cortisol and heart rate values in the Trier Social Stress Test in the presence of a dog than in the presence of a friend (10.1007/s10865-013-9546-1). Allen and colleagues (2002) reported similar results for mental arithmetic tasks and a cold pressor test (10.1097/01.PSY.0000024236.11538.41).

The possible pathway, then, is: dog → less tension → altered pain processing. A pathway, not proof.

4. What animal-assisted interventions show

4.1 The overall evidence

Correale and colleagues (2022) screened 350 papers and included 21 studies on animal-assisted interventions for hospitalized children and adolescents; all of them involved dogs. Most dealt with stress, pain, and anxiety and used physiological parameters as well as behavioral and questionnaire measures. The conclusion: indications of less anxiety and less emotional distress, and in some settings also of lower pain burden (10.3389/fpsyg.2022.840107).

One of the methodologically stronger individual studies among them is a randomized study in children after surgery (Calcaterra et al., 2015), which found changes in pain ratings and physiological parameters (10.1371/journal.pone.0125813).

4.2 Why the conclusion is cautious

The studies are heterogeneous, vary in methodological strength, and some carry a considerable risk of bias. Above all, it is often impossible to cleanly separate how much of the effect is due to the dog and how much to the extra attention, the social contact, and the special setting. Blinding is practically impossible when a dog is in the room—everyone knows which condition they are in, and that alone can create expectations.

The defensible wording is therefore: animal-assisted interventions repeatedly show indications of reduced anxiety and distress in clinical settings. The dog’s specific contribution, as opposed to contextual factors, is still being researched.

5. The study that found no effect

If you only know the Berlin study, you get too neat a picture. Wagner and colleagues (2021) conducted a randomized controlled trial with healthy participants and found no effect of a dog’s presence on pain perception (10.3389/fpain.2021.714469).

The decisive difference probably lies in the kind of presence: in Wagner’s study, an unfamiliar dog was passively present; after a brief greeting, there was no further contact. In Study 1 of the Berlin paper, the dog was likewise in the room without contact, but it was the participant’s own dog. Mauersberger and colleagues cite this paper themselves and designed their second study with active support partly for that reason. That sharpens the message: it is not the mere presence of a dog in the room that works reliably, but a combination of relationship, accessibility, and attitude.

6. Where the evidence comes from

The core study is based on 74 and 50 participants, women only, young (mean age 27.9 and 26.6 years, respectively), healthy, and not on long-term medication; women with chronic pain were excluded. The data were collected in 2013 and published in 2024. The analysis was Bayesian, using Bayes factors instead of classical significance tests—methodologically sensible, because this approach can also provide evidence for the absence of an effect, but it makes the results harder to compare with other literature.

In Study 1, assignment to conditions was only pseudo-randomized: not all participants had their dog available in Berlin, and about a quarter of the dogs lived with the participants’ parents in the countryside. Those who were able to have their dog with them may also differ from the others in other respects.

The second pillar, intervention research, consists mostly of small, heterogeneous, rarely blinded studies. The third, oxytocin research, measures a mechanism, not pain.

And the cold pressor test itself has limitations, which the authors explicitly acknowledge: it stimulates not only pain fibers but also produces numbness and general stress, and some people tolerate it without any notable physical stress response.

7. What the research does (not yet) show

• Not that dogs influence chronic or clinical pain. What was measured was briefly induced lab pain in healthy people.
• Not that the effect applies to men. Both studies included women only.
• Not that the mere presence of any dog is enough. A randomized study with an unfamiliar, passively present dog found no effect.
• Not that dogs help everyone. When participants were reserved toward dogs, the advantage disappeared and, on some measures, reversed.
• Not that being free of judgment is the cause. That dogs are experienced as less judgmental has been measured; that this causes the pain effect is an interpretation.
• Not how much of the clinical effects is due to the dog and how much to attention, social contact, and setting.

8. But dogs are not painkillers

A dog does not replace medical diagnosis, pain management, physical therapy, or necessary treatment. It can be a supportive social factor—no more, but also no less.

This applies in both directions: if a person is in pain, it should be checked out. And if a dog is in pain, the same holds—how to recognize it is covered in Recognizing Pain in Dogs and, with a focus on behavior changes, in Chronic Pain and Behavior: The Neurobiological Link Between Osteoarthritis and Aggression in Dogs.

9. What does this mean for everyday life?

For everyday life, and for work with visiting and therapy dogs, a sober but useful line can be drawn from this:

• The calm presence of a familiar dog can ease stressful moments—no program required.
• Don’t force closeness. The effect depends on how much a person likes dogs; with people who are reserved, a dog that is pushed on them can make things worse. That is not a side note; it is in the study.
• Familiarity and accessibility seem to matter. With the participant’s own dog or when contact was possible, effects appeared; with an unfamiliar dog and no contact, in one study, they did not.
• The dog itself needs to be relaxed. A stressed dog is not a therapy dog. The dog’s well-being is a prerequisite, not an afterthought—how strain shows in a dog is covered in Chronic Stress in Dogs: Neurobiology, Cortisol and Long-Term Behavioral Impact.
• Not every dog is suited for supportive work. That is a question of temperament, training, and how the dog is doing on a given day. The three dogs in Study 2 had been specially trained and habituated to the lab.

10. The side missing from the discussion: the dog

10.1 Who is under strain in these studies

Almost all of the literature on this topic asks what the dog does for the human. What the work means for the dog is rarely measured. In the Berlin study, the dogs sat in a lab with measuring equipment, cables, and a person holding a hand in ice water and visibly showing pain. The authors report that two participants were excluded because the dog tried to walk away, turned its head away, or barked or whined.

A word of caution about interpretation: a dog that gets up and walks away is not necessarily stressed—it may simply be exploring. Combined with whining and barking, however, discomfort becomes plausible. What is most notable is how this was handled: these observations were treated as a disturbance in the analysis, not as a finding in their own right. That is methodologically understandable, and at the same time it shows where the research question ends.

10.2 What this means in practice

A dog working with people under strain operates in an environment full of noises, smells, unfamiliar people, and clearly visible stress in the person in front of it. That it copes with this cannot be taken for granted, and it is not a matter of goodwill but of selection, gradual preparation, and ongoing observation.

The three dogs in the second Berlin study were specially trained and gradually habituated to the room, the equipment, and the procedures before the first session took place. That is exactly how it should be done—and it is exactly what determines whether this kind of work is justifiable for both sides.

10.3 Resilience is only half the answer

Discussions about whether a dog is suitable usually focus on resilience: What can it cope with? That is the wrong half of the question. A dog suited for social work needs not only resilience but also choices—the option to keep its distance, withdraw, and take breaks without anyone calling it back.

This is not a concession to the dog; it is part of the job. An animal that cannot leave a situation is in exactly the position this article describes as stressful for people: it has no control over it. In practice, that means a retreat spot in the room that the dog is actually allowed to use, limited working times, breaks between contacts—and a person who watches the dog and ends the session before the dog has to do it itself.

11. Conclusion

The more convincing story is not “dogs relieve pain,” but this: through social closeness, stress regulation, and emotional processes, dogs can influence how people experience and express pain. The Berlin study provides an unusually cleanly measured experimental building block for this, and its truly interesting finding is not that the dog helped, but that the friend sitting quietly nearby did not.

What remains is a nuanced picture: the connection between human and dog can influence the subjective experience of pain and certain stress-related responses—not as medical treatment but as a social and emotional factor, and not equally for everyone. One more good reason to nurture this relationship deliberately, as described in Building Trust with Your Dog.

Key takeaways

• In the presence of their own dog, 74 participants reported and showed less pain than when alone; the presence of a friend did not differ from the alone condition (Mauersberger et al., 2024).
• This is not explained by the friends being less attentive—they looked at the participant more often and communicated more than the dogs.
• With unfamiliar dogs, the advantage for pain reports and heart rate depended entirely on attitudes toward dogs; among participants who were reserved toward dogs, it partly reversed.
• Dogs were experienced as less judgmental, closer, more empathic, and more supportive than the human companions.
• A randomized study with an unfamiliar, passively present dog found no effect on pain perception (Wagner et al., 2021).
• A review of 21 hospital studies with dogs found indications of less anxiety and distress, with heterogeneous methods and a dog-specific contribution that is hard to separate out (Correale et al., 2022).
• All findings come from acute lab pain or short-term clinical settings, not from the treatment of chronic pain.

References

Allen, K., Blascovich, J., & Mendes, W. B. (2002). Cardiovascular reactivity and the presence of pets, friends, and spouses: the truth about cats and dogs. Psychosomatic Medicine, 64(5), 727–739. https://doi.org/10.1097/01.PSY.0000024236.11538.41

Calcaterra, V., Veggiotti, P., Palestrini, C., De Giorgis, V., Raschetti, R., Tumminelli, M., Mencherini, S., Papotti, F., Klersy, C., Albertini, R., Ostuni, S., & Pelizzo, G. (2015). Post-operative benefits of animal-assisted therapy in pediatric surgery: a randomised study. PLOS ONE, 10(6), e0125813. https://doi.org/10.1371/journal.pone.0125813

Correale, C., Borgi, M., Collacchi, B., Falamesca, C., Gentile, S., Vigevano, F., Cappelletti, S., & Cirulli, F. (2022). Improving the emotional distress and the experience of hospitalization in children and adolescent patients through animal assisted interventions: a systematic review. Frontiers in Psychology, 13, 840107. https://doi.org/10.3389/fpsyg.2022.840107

Krahé, C., Springer, A., Weinman, J. A., & Fotopoulou, A. (2013). The social modulation of pain: others as predictive signals of salience – a systematic review. Frontiers in Human Neuroscience, 7, 386. https://doi.org/10.3389/fnhum.2013.00386

Mauersberger, H., Springer, A., Fotopoulou, A., Blaison, C., & Hess, U. (2024). Pet dogs succeed where human companions fail: the presence of pet dogs reduces pain. Acta Psychologica, 249, 104418. https://doi.org/10.1016/j.actpsy.2024.104418

Nagasawa, M., Mitsui, S., En, S., Ohtani, N., Ohta, M., Sakuma, Y., Onaka, T., Mogi, K., & Kikusui, T. (2015). Oxytocin-gaze positive loop and the coevolution of human-dog bonds. Science, 348(6232), 333–336. https://doi.org/10.1126/science.1261022

Polheber, J. P., & Matchock, R. L. (2014). The presence of a dog attenuates cortisol and heart rate in the Trier Social Stress Test compared to human friends. Journal of Behavioral Medicine, 37(5), 860–867. https://doi.org/10.1007/s10865-013-9546-1

Wagner, C., Gaab, J., Locher, C., & Hediger, K. (2021). Lack of effects of the presence of a dog on pain perception in healthy participants – a randomized controlled trial. Frontiers in Pain Research, 2, 714469. https://doi.org/10.3389/fpain.2021.714469

Frequently asked questions about dogs and pain