Over this past weekend I was at a dog show and noticed typical canine social interaction. Two dog handlers were having an animated discussion. One had a large, darkly marked, German Shepherd dog at the end of her leash and the other had a small, black and white, Cocker Spaniel leashed next to her. The German Shepherd took two stiff steps in the direction of the Cocker Spaniel, stared at it in a dominant manner and gave a low growl. The spaniel immediately dropped down so that its belly was against the ground. He then turned his head away from the threatening dog and his tongue came out as he seemed to be licking the air. The large dog seemed to be appeased by the spaniel's reaction and he stopped staring and turned to watch some of the other activities in the hall. The moment that the aggressive threat had disappeared the spaniel got up and skittered around to place his owner between himself and the German Shepherd.
My interpretation of this scenario (and I believe the interpretation that most people with knowledge of dog behavior would offer) is that the German Shepherd showed behaviors which are associated with aggressive intentions and the spaniel responded with a set of calming signals showing that he was no threat. In this case it looked like the spaniel's communication was successful and the aggression level in the other dog was reduced.
The label "calming signals" was introduced by the Norwegian dog trainer Turid Rugaas in a short (78 page) book "On Talking Terms with Dogs: Calming Signals" first published in 1996 and then reissued 10 years later. In it she describes around 30 forms of behavior or body language which dogs use in social interactions with other dogs, particularly in interactions where the dog is uncomfortable or feels threatened. It is her belief that the use of calming signals tends to lower the probability that another dog will act aggressively toward the dog doing the signaling. Any list of the most common calming signals should include:
- Looking away [turning the head to the side away from the other dog]
- Turning away [here the whole body is turned so that the side or back faces the other dog]
- Making a "soft face" [including ears back, eyelids lowered, forehead smoothed, mouth closed]
- Freezing [stopping all movement regardless of whether the dog is standing, sitting, or lying down]
- Lip licking [or nose licking, or just flicking the tongue in front of the face]
- Sniffing the ground [as if interested or exploring something there]
- Lying down [belly against the ground, paws out in front]
- Raising one paw
- Scratching [as if the dog is simply responding to an itch]
- Shake off [this can be a slight shake or involve the entire body as if it was wet]
- Play bow
- Slow movements
- Slow tail wagging with the tail held low
- Moving in a curve [when approaching or leaving the dog shows its side and does not approach the front of the other dog directly]
- Lip smacking
Rugaas is not a scientist and did not offer experimental evidence that these kinds of signals actually worked when it came to lowering the probability of aggression between dogs. Nevertheless her observations were insightful, and the examples which she gave were very convincing. However there is a problem which often plagues our ability to understand behaviors and events correctly, and that is something called the confirmation bias in which we tend to only remember examples which confirm our beliefs and expectations. That is why so many people feel that they have a "system" that they can use in gambling to beat the house in craps, or roulette or other games of chance. They believe that they have a winning system because they only remember the times that they win and ignore the times that they lose. Perhaps those of us who work with dogs only remember the times when aggression de-escalated after a calming signal and forget the times when it did not.
The reason that we should be concerned with such a bias here is because Rugaas is suggesting that these calming signals are a form of communication. A psychologist would suggest that these are appeasement gestures, which are a specific form of communication that is meant to prevent any hostile behaviors that might be contemplated by another individual. However, it is also possible that these signals are simply instinctive signs of stress and are not chosen by the threatened dog as a means of communicating. For example yawning is known to be a sign of stress in dogs (click here to see more about that) and may not be intended as a communication signal at all. In addition some of the calming signals may simply be what psychologists call displacement behaviors which occur when an animal is conflicted between two drives, such as wanting to approach something but being unsure at the same time. This kind of psychological conflict will sometimes produce a behavior that has nothing to do with the situation at all, such as, in the case of dogs, scratching, sniffing the ground, and so forth.
Given the absence of scientific data showing that calming signals actually calm, I was pleased to see the publication of a new study by a team of researchers headed by Chiara Mariti of the Department of Veterinary Sciences at the University of Pisa in Italy. The study involved 24 dogs who were videotaped when interacting two at a time. The dogs were placed in the test area with either a familiar or unfamiliar dog. Data was collected on calming signals and their effects. A total of 2130 of such signals were observed during the course of the study.
The most common calming signals were head turning, lip licking, freezing and turning away. If these signals were really meant to be communication then they ought to appear more often when the target dog was interacting with another dog rather than when the dogs were just in the test area and not interacting. This appears to be the case. In fact most of the calming signals occurred when the dogs were fairly close, within 1.5 body lengths. Overall more calming signals were exhibited when a dog was interacting with an unfamiliar dog, rather than one which it had encountered and had social contact with previously. This makes sense since all interactions with unknown individuals contain a higher possibility of threat, and perhaps sending out a calming signal might just be a polite canine custom to avoid hostility.
Do the calming signals actually make a difference? During this study there were 109 instances of aggressive behavior. None of these aggressive acts followed a calming signal, which suggests that had a calming signal been issued the aggression might have been avoided. Something which seems to support this idea is the fact that in 67% of the instances where aggression had just occurred at least one calming signal occurred immediately after. The effectiveness of that calming signal is supported by the fact that in over 79% of the instances when it occurred after the initial aggressive act, the hostility de-escalated. This kind of data certainly supports the idea that calming signals actually do have a calming effect.
Unfortunately the effect was not universal. These researchers report that there was one dog in their sample who was unresponsive to the calming signals of the other dogs and continued the aggression as if no calming signal had been given. I am tempted to draw a parallel to human criminal psychopaths in this instance. These psychopaths have been shown to be unresponsive to the emotional content of communication from other people and seem to show no empathy. Given all that we have learned about the similarity between the psychology of dogs and humans, there is no reason to expect that an analog to human psychopathy might not exist in dogs as well.
If we ignore the data of this possible canine psychopath, the results of this study seem to be that calming signals given by a dog may actually change the behavior of dogs that received these signals and reduce the level of aggression. However the researchers describe their experiment as a pilot study, and it does still leave some unanswered questions. For example we still don't know if these signals are truly displayed by the dogs in a conscious attempt to calm down a potentially threatening situation. We do know that dogs have empathy for other dogs in distress and seem to try to act in a supportive way (click here for an example). Perhaps the calming signals are not communication in the standard sense of the word, but rather they provide information to the potentially aggressive dog. Dogs do read the indications of the emotional state of other dogs, and perhaps seeing that the dog that they are confronting is stressed and acting submissively, the potential aggressor understands that this other dog is probably no danger. Furthermore it is possible that the potentially hostile dog responds to the emotional state of his counterpart, feels empathy for the frightened dog, and in so doing the aggressive impulses drain away.
Turid Rugaas (2006). On Talking Terms With Dogs: Calming Signals, 2nd ed. Dogwise publishing, Wenatchee, WA.
Chiara Mariti, Caterina Falaschi, Marcella Zilocchi, Jaume Fatjó, Claudio Sighieri, Asahi Ogi & Angelo Gazzano (2017). Analysis of the intraspecific visual communication in the domestic dog (Canis familiaris): A pilot study on the case of calming signals. Journal of Veterinary Behavior 18, 49-55
Stanley Coren, Ph.D., FRSC., is a professor of psychology at the University of British Columbia.
Posted Jun 07, 2017