I had brought my dogs to be part of a "Doggy De-Stress Day" on the campus of a local university. These are becoming more common for many colleges in North America and usually take place during mid-term exam or final exam periods. The way it works is that dogs (often therapy dogs, but sometimes just well-behaved pets) are brought to campus and students get a chance to pet and interact with the dogs. The rationale here is that during exam periods stress levels run high in the student population, and there is ample evidence that shows that dogs can reduce stress levels. (Click here for more about that). So this seems like a simple method of making students feel a bit less hassled before and between their tests.
At one point during the event, a diminutive woman came over to my Nova Scotia Duck Tolling Retriever puppy and gave him a hug. At the time, he was about six months old, and, like most puppies, relatively tolerant of any form of interaction. Nonetheless, in response to the girl's hug he turned his head to break off eye contact, his ears slicked down and gave a small stress yawn. I leaned over and said to her: "You really shouldn't hug a dog. They don't like it and it raises their stress level."
The girl looked at me with an expression of disbelief and said, "I'm studying developmental psychology and there's lots of evidence which says that hugging is important and pleasant. When a mother hugs her child the child gets a surge of the hormone oxytocin and so does the mother, and that hormone is associated with loving and bonding. There is evidence that says that if parents don't hug and touch their child a lot, that child can grow up to be emotionally stunted. So how can you tell me that hugging isn't good for dogs, especially for a puppy?"
The real answer to her question is, of course, that dogs are not human children. Dogs are technically cursorial animals, which is a term that indicates that they are designed for swift running. That implies that in times of stress or threat the first line of defense that a dog uses is not his teeth, but rather his ability to run away. Behaviorists believe that depriving a dog of that course of action by immobilizing him with a hug can increase his stress level and, if the dog's anxiety becomes significantly intense, he may bite. For that reason, certain websites, which try to educate children and parents in order to reduce the incidence of dog bites (such as Doggone Safe), make a point about teaching children that they should not hug dogs. Furthermore, a few years back when a children's book entitled "Smooch Your Pooch" recommended that kids hug and kiss their dog anytime and anywhere, the American Veterinary Society of Animal Behavior (AVSAB) felt that it was necessary for them to release an official statement that strongly advised parents to avoid purchasing the book, since "this information can cause children to be bitten."
Given how widely accepted the idea is that hugging is not something that dogs like and that hugging a dog may be associated with increasing the likelihood of a dog bite, I was surprised that a search of the scientific literature produced very little experimental evidence to support that belief. I did find two articles that showed that getting bitten on the face was much more likely if you were hugging or kissing a dog. However, the authors of both studies seemed to suggest that the proximity of the person's face to the dog's mouth was the most important factor, rather than something like the hug itself. For that reason, I decided to collect some data on this issue.
The signs of stress and anxiety in dogs are well established, and are easily observable, at least by trained individuals. Obviously, at the high-end of stress, we have dogs who bare their teeth. But, there are subtler indicators. The most common sign of anxiety is when the dog turns his head away from whatever is bothering or worrying him, sometimes also closing his eyes, at least partially. Alternatively, dogs will often show what is commonly called a "half-moon eye" or "whale eye” which is where you can see the white portion of the eyes at the corner or the rim. One common visible sign of stress or anxiety is when the dog's ears are lowered or slicked against the side of his head. Lip licking or licking a person's face can also be signs of anxiety, as can yawning or raising one paw. These signs and other similar ones should be easy to detect in stressed dogs. All that I needed then to conduct the research was a source of photographic material showing people hugging their dogs.
Fortunately for me, the Internet abounds with photographs of people and their pets. If you put the search terms "hug dog" or "love dog" into something like Google Image Search, or Flickr, you will get a virtually infinite scroll of pictures of people and their children hugging their pet dogs. I decided to look at a random sample of 250 such pictures. I used a variety of criteria to try to keep the data as clean and precise as possible. I only used photos where the dog's face was clearly visible. I also eliminated situations where one might expect the dog's stress level to rise because of factors other than being hugged (such as when someone lifts a large dog off the ground while hugging them). Each picture received one of three possible scores:
- One could judge that the dog was showing one or more signs of stress or anxiety;
- One could judge that the dog appeared to be relaxed and at ease;
- One could decide that the dog's response was ambiguous or neutral.
Two examples of dogs that scored as being stressed while they were in the process of being hugged appear below.
I can summarize the data quite simply by saying that the results indicated that the Internet contains many pictures of happy people hugging what appears to be unhappy dogs. In all, 81.6% of the photographs researchers scored showed dogs who were giving off at least one sign of discomfort, stress, or anxiety. Only 7.6% of the photographs could rate as showing dogs that were comfortable with being hugged. The remaining 10.8% of the dogs either were showing neutral or ambiguous responses to this form of physical contact (detailed statistics can be found at the end of this article).
I suppose that one aspect of the data that struck me as interesting comes from the fact that the photographs that I used were obviously posts by individuals who wanted to show how much they cared for and shared a bond with their pet. This means that the people who were doing the Internet posting probably chose photos in which they felt that both the person and the dog looked happiest. Nonetheless, around 82% of the photographs show unhappy dogs receiving hugs from their owners or children. This seems consistent with other research which suggests that people, especially children, seem to have difficulty reading signs of stress and anxiety based upon their dogs' facial expressions. (Click here for more about that.) Much more relevant for the current question is the fact that this data clearly shows that while a few dogs may like being hugged, more than four out of five dogs find this human expression of affection to be unpleasant and/or anxiety arousing.
The clear recommendation to come out of this research is to save your hugs for your two-footed family members and lovers. It is clearly better from the dog's point of view if you express your fondness for your pet with a pat, a kind word, and maybe a treat.
Stanley Coren is the author of several books including Gods, Ghosts and Black Dogs.
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission
The data set consisted of the first 250 photos that came out of Google images and Flickr using the search terms "hug dog." To be used in the data set the image had to show a person hugging a dog, with enough of the dog's face visible to see if there were any stress signs. Photos were the position might cause stress independent of hugging (such as attempts to lift a large dog while hugging it) were not scored.
Data was divided into three groups,
- one or more stress signs;
- one or more signs of positive affect;
- neutral or ambiguous emotional signs (e.g. both positive and negative or otherwise unscorable).
Number of observations = 250
Number showing stress = 204
Proportion showing stress = 0.816
Null hypothesis = 0.50
[Here I can be extremely conservative since most people believe that the dogs like being hugged and therefore their expectation should be that there should be virtually no stress signals when they are hugging the dog, however, let us use the statistically conservative presumption which is that the appearance of stress signals as opposed to non-stress signals are distributed randomly which would give us the presumption that half of the photos should show stress signs which thus gives us the null hypothesis of 0.50]
95% confidence interval = 0.7611 — 0.860
The data were highly significant in the statistical sense.
Statistical significance of the deviation from chance [0.50] Z = 9.93 (P <0.0001).
The funding for this study came from my private corporation (SC Psychological Enterprises LTD), which sets aside a proportion of the royalties from my books and lectures to finance research on dog behavior and the human-canine bond, with the idea that the results from those research studies may be used in my various books and articles.