Recent research has shown that dogs can understand where people are looking, but can dogs understand where other dogs are gazing? A recent article in the journal Animal Cognition suggests that dogs can do just that. The article is by Alexandra Horowitz (Horowitz, A. 2008. Attention to attention in domestic dog (Canis familiaris) dyadic play. Animal Cogntion, in press, DOI 10.1007/s10071-008-0175-y). Horowitz videotaped the interactions between 39 pairs of dogs engaged in play behavior in a park in southern California where dogs were allowed to run off leash. She then analyzed her videotapes to find out what behaviors the dogs were using during play.
Playing dogs used several different signals to indicate to each other that they were playing. These included a play-bow, an open-mouth play face, a head bow, a play slap with the forelegs slapping the ground, and a leap-on or jumping up and putting the front paws around the other dog’s head.
But sometimes one of the dogs looked away and seemingly lost interest. Then the other dog used attention-getting signals that most of the time caused the other dog to restart the play behavior. These signals included placing the body near the other dog’s face, making a backwards leap, turning around so that the dog’s rear end was next to the other dog’s face, bumping into the dog, touching the other dog with nose and closed mouth, biting the other dog with a soft bite, making biting movements in the air, pawing at the other dog, and barking.
When these attention-getting signals were used, play restarted 88 percent of the time. When the other dog who wanted to continue to play used a play signal or some behavior other than an attention-getting one, play restarted only 20 percent of the time.
Sometimes the dog who stopped playing was not looking at the dog who wanted to continue to play. When that happened, the dog who wanted the play to continue did not use visual attention-getting signals (e.g., placing the body near the dog’s face or making a backwards leap) but instead used tactile signals (e.g., bumping, touching the nose to the other dog’s body, biting with a soft bite).
The results suggest that dogs can understand that when another dog is not looking at them, it is better to use touch as a communication tool rather than using visual signals. This relates to what philosophers and psychologists call Theory of Mind. The question is, does the dog understand what is going on in the mind of the other dog? Horowitz takes a cautious approach and says that more research is needed in this area. Those of us who have been around dogs a lot are probably willing to be less cautious about that determination.
--Con Slobodchikoff
In her doctoral dissertation, Alexandra Horowitz developed a cogent, well-argued case in support of a ‘rudimentary’ theory of mind (rTOM), based on extensive analyses of social play behavior in dogs. Since then, she has been consistent in further describing and elaborating upon her theory. However, the sense of caution she sometimes shows may be due to the fact that not everyone understands the ‘theory of mind’ concept and it can be misunderstood or misrepresented. For instance, Dr. Horowoitz was a speaker at the first Canine Science Forum, held in Budapest, Hungary in early July of this year. Her presentation inspired a piece that was published in New Scientist, but the resulting article made it sound as though she had changed her mind and was now going in the opposite direction. I called attention to this seeming inconsistency and she told me, via e-mail, that the article had, indeed, misrepresented her theory and that “something got lost in the editing process”. She assured me that she has not changed her views and continues to support the rTOM concept as introduced in her dissertation.
Posted by: Randall Johnson | October 13, 2008 at 12:22 PM