Should we call these canine behaviors calming signals?

“Calming signals” is a term coined by Norwegian dog trainer Turid Rugaas to group a large number of behavior patterns that she says dogs use to avoid conflict, to prevent aggression, to calm other dogs down and to communicate information to other dogs and to people. Since the publication of Rugaas’ 2006 book On Talking Terms With Dogs: Calming Signals, it has been a popular idea that actions such as lip-licking, sniffing the ground, yawning, scratching, looking away, play bowing, sitting down, lying down, softening the eyes, blinking and even sneezing (along with many others) are social signals that help calm down those around them.

Rugaas’ observations are compelling, and many dog trainers and behaviorists, including me, have learned a lot from her work. However, the term “calming signals” entered the lexicon without much analysis, which is problematic. Using a term that ascribes functionality to behavior patterns prior to scientifically testing whether or not that’s true creates challenges, and is a big no-no in ethology. One problem is that claiming that certain behaviors are “calming signals” creates a bias such that people tend to accept that this is, in fact, what they do. The idea that these signals are functioning in this way is an intriguing hypothesis. However, in the years since Rugaas shared her ideas with the dog community, there have yet to be adequate tests of their function, or substantial efforts to determine if the various behaviors have different functions. Rather, the idea that they were calming signals was broadly accepted without being subject to rigorous scientific study.

There is, however, a recent pilot study investigating the function of the behavior patterns that have all been placed into the category of calming signals. The purpose of the study “Analysis of the intraspecific visual communication in the domestic dog (Canis familiaris): A pilot study on the case of calming signals” was to assess if the behaviors that have been called calming signals are used to communicate, and if they de-escalate potentially aggressive situations between dogs. In the study, 24 dogs were observed interacting two at a time. Dogs interacted with familiar and unfamiliar dogs of both sexes.

Throughout the course of the study, 2130 calming signals were observed, with the most common being head turning, nose licking, freezing and turning away. Dogs were more likely to display calming signals when they were interacting with the other dog compared with when they were not interacting, which does suggest a communicative role. It does not prove it, though, as it is possible that these behaviors indicate stress and that they are performed during social interactions more often than they are performed outside of that context because such interactions are stressful. In fact, most of the signals that Rugaas has called “calming signals” are also considered indicators of stress.

More calming signals were displayed when dogs were interacting closely (within 1.5 body lengths of the dog displaying) than when interacting at a greater distance. Overall, more calming signals were exhibited during interactions with unfamiliar dogs than with familiar dogs, but licking the other dog’s mouth was more frequently observed when the other dog was familiar.

During the interactions in the study, there were 109 instances of aggressive behavior. A calming signal never came right before the aggressive behavior, but 67% of the time, at least one calming signal followed the aggressive behavior. In over 79% of the instances in which a calming signal followed the start of the aggression, there was a de-escalation in the aggressive behavior. These data are consistent with the idea that these behaviors function to calm other dogs down and lessen their aggression, but the work is too preliminary to conclude this for certain. More research is needed to explore other possibilities, such as the role of stress in these behaviors and their effects, and the potentially different functions of each of the dozens of behaviors that have been lumped under the term “calming signals”.

This is a pilot (or preliminary) study, and though the results are intriguing, they are in no way a definitive test of the function of “calming signals” in dogs, which the authors correctly point out in their paper. Though this research makes an attempt to test the often-accepted hypothesis that many behavior patterns function as calming signals that de-escalate aggression, its biggest flaw is that it lacks a very important control. De-escalation of aggression is quite common, and in this study, the authors report the frequency of de-escalation after a calming signal, but do not report on the rate of de-escalation in the absence of a calming signal. Part of the problem is that with so many possible calming signals, it is quite likely that one will be exhibited as a response to aggression. (Dogs are unlikely to have no reaction to such behavior.)

To evaluate the function of the behaviors, it is necessary to know the frequency with which the aggression de-escalates in the absence of any calming signals. We know that there was often de-escalation in the absence of calming signals because the authors report that in quite a few cases, the dog on the receiving end of the aggression walked or ran away, increasing the distance between the two dogs, which was often associated with a de-escalation in aggression. Fleeing is not considered a calming signal, and yet when the distance increased between the two dogs, there was also usually a de-escalation of the aggression. Future research should explore the differences in behavior in cases in which there was de-escalation and in which there was not.