Dogs’ Responses to Affection from Familiar and Unfamiliar People

Dogs prefer to be petted on the side of their chest or under the chin, not on the top of the head, paws, or back legs--and it makes a difference who is doing the petting. 

Dogs respond differently to affection from familiar vs unfamiliar people
Photo: Ksenia Raykova/Shutterstock

When my Siberian husky wants affection, he will come and stand near me. If I don’t respond immediately, he will lick his lips and move closer, possibly leaning on me, until I respond. Sometimes when I start to pet him, he will lick his lips again, but if I take this as a sign that he’d like me to stop, he licks his lips even more and moves closer or paws at me to ask me to start petting again. At some point he will sit, and then lay down and ask for chest rubs. From his perspective, chest rubs should last at least half an hour, and if I stop at any point before this there will be more licking of lips and pawing at me.

I was interested to read a new paper by Franziska Kuhne and colleagues in Germany that investigates how dogs respond to petting from both familiar and unfamiliar people. Twenty-four pet dogs took part, of assorted ages, genders, breeds/cross-breeds and training levels. Since the experiment involved some actions that dogs might not like, they were pre-tested by a member of the research team to ensure they would not be aggressive; this person didn’t take any further part in the experiment.

Some of the dogs were tested by a person who was familiar to them as they had been seen whilst on walks with their owner. The other group of dogs were tested by the same person, but did not previously know her. The study took part in an office and dogs were given a treat at the beginning to encourage them to approach the experimenter.

In each case, the dog was tested with nine different human actions, each for a set time and with a gap in between. The behaviours were: petting on the shoulder, chest or neck, petting the dog whilst it was laying down and simultaneously holding it down, holding its front paw, petting on top of the head, scratching it close to the base of the tail, holding the dog’s collar, and holding one hand over the dog’s muzzle.

The sessions were videoed, and dogs’ responses to the different actions were analyzed. When the person was familiar to the dog, re-directed behaviours occurred more often than when the person was unfamiliar. They were especially likely to occur in response to holding the dog’s collar or covering its muzzle. Redirected behaviours include sniffing or licking the floor, digging, drinking, over-activity, visual scanning and playing with objects. Once the dogs began this kind of behaviour, it continued for longer in the familiar than the unfamiliar group.

There was also a significant difference in appeasement gestures, which include blinking, looking away, closing the eyes, licking the nose or lips, lifting a paw, turning the head or moving the body away, and laying down. Appeasement gestures were significantly more common in the familiar group. In addition, they lasted longer in response to some actions, namely petting on the head or shoulder, under the neck, having the collar held or laying down. 

Finally, panting (a sign of stress) was seen more often in the familiar group. There was no difference in terms of displacement activities, such as yawning, licking/scratching themselves, shaking, stretching and vocalizing.

The authors say these results show that dogs generally don’t like to be petted on the top of their head, on their paws, or on their hind legs, and that they prefer to be petted on the side of the chest or under the chin. They conclude that dogs may misunderstand some human behaviours since they mean something else during interactions between dogs. 

These dogs were pre-tested to ensure they weren’t aggressive, and so the results might not generalize to less friendly dogs. In particular, grabbing a dog by the collar or surrounding its muzzle with your hand may trigger a bite from some dogs.

This study raises a question that isn't addressed in the report. We could assume that actions dogs don’t like are more unpleasant coming from an unfamiliar, rather than familiar, person. In this case, we would expect to see some kinds of behaviours that happened more often in the unfamiliar group, and yet this isn’t the case with the body language measured. However, the authors do suggest the reason displacement activities were not observed could have been because of the mild restraint that was part of the petting.

It would be nice to have more detail of the body language that was observed, since the results are summarized as the category of behaviour (e.g. redirected) rather than individual behaviours. Presumably only some redirected behaviours were seen, since it is hard to imagine a dog digging in the office setting.  

The interaction was also unusual for dogs, since people don’t usually pet them for an exact period of time then stop, pause, and repeat with a new action. Future research could look at how dogs respond to affection in more natural situations.

The paper's main finding was that appeasing behaviours (such as licking the lips) and redirected behaviours (such as sniffing the ground) were more often observed with a familiar, rather than unfamiliar, person.

Next time you are petting your dog, observe their body language. How do they react? And if you stop, what happens then?

Kuhne, F., Hößler, J. C., & Struwe, R. (2012). Effects of human–dog familiarity on dogs’ behavioural responses to petting. Applied Animal Behaviour Science, 142(3-4), 176-181.

Follow me!