Canine Behavior: Do Dogs Have a Theory of Mind? Part 2

Publishing Author – Charles Lee Kelley

Date Published – 15/07/16


What People Notice That Dogs Don’t, What Dogs Notice That People Don’t

Theory of mind is a scientific construct developed by David Premack and Guy Woodruff. There are essentially three levels. The first level is the capacity to know that other beings have the same sensory abilities that you do, which carries with it the capacity to realize that others may also see, hear, feel and experience things differently from the way you do.


One example might be pointing out to a friend that they have a smudge on their cheek.  If the smudge is on their right cheek you might touch the same spot on your left cheek—as if you were her mirror image—and your friend will usually rub or dab at that part of her cheek. Or let’s say you’re having brunch at an outdoor café, and you point behind your dining companion and say, “Is that Harrison Ford?” Your companion would probably turn around to see for herself. The reason you might point these things out to your friend is that you know that your field of vision, what you can see from your perspective or vantage point, is quite different from hers.


One example of perspective-taking commonly espoused by dognitive scientists is the idea that a dog will follow where a human points while chimps won’t. In dogs, however, this only happens under very artificial conditions, in controlled settings.



In real life, if you point, let’s say at something on the sidewalk behind a dog, 99 times out of 100 the dog will scan the area in front of him instead of looking where you’ve pointed. This suggests that dogs aren’t engaging in perspective-taking, but have learned a generalized pattern of behavior: point = find toy, etc.


The genesis of gestures—like pointing behind someone at brunch so they can get a glimpse of a Hollywood star—may be important because as Noam Chomsky says, “Humans have complicated gestural systems. This isn’t terribly well understood but it’s more or less assumed, probably realistically, that the gestural systems are continuous with the gestural systems of other primates. It’s Darwin’s assumption and probably plausible.”


Yes, gestures are communicative in nature, but do body language and gestures add up to a Theory-of-Mind in animals?


The Sense of Being Stared At: Eye Contact as a Physical Force

Another popular example of what some researchers believe is perspective-taking in dogs, is based on studies showing that a dog is more likely to act “sneaky” and misbehave by “stealing” food when the lights are off but won’t do so when the lights are on.


In a recent blog post on his website, author and dog trainer Kevin Behan writes about how human observation, just the mere act of looking at a dog and paying attention to his behavior, can act as a very real physical force, inhibiting certain unwanted behaviors.


“Even when a dog is not focusing directly on the [owner’s] eyes, he is nevertheless well aware of their presence because a person’s gaze is the precursor to where they will direct their force.”


This doesn’t mean the owner is using physical force to control the dog’s behavior but rather that the simple act of paying attention to a dog’s actions and activities—particularly during puppyhood—sensitizes the dog to any form of directed attention which then acts as a physical force on the dog’s body/mind during many daily interactions.


Behan: “A line-of-sight is like a search light, a beam that can sweep an area in a wide angle broadcast of peripheral vision, or can fine tune down to a laser-like pin point application. An animal is highly sensitized to areas of exposure created by the floodlight of human attention that might swivel about the surroundings or bear down in a tight focused beam.”


So, yes. Direct eye contact can act as a physical force.


Dogs first become keenly aware and highly sensitive to the potential force held by their owners’ attention during puppyhood, and change their behaviors accordingly.


For example, some clients of mine asked for help housebreaking their young female beagle. I made it very clear that the dog should be confined at times when they couldn’t “keep an eye” on her until she was old enough to reliably do her business outdoors, and that they shouldn’t under any circumstances try to stop her from piddling while she’s already in the act of doing so.


“Why not?”


“Because she’ll become wary and do it in places where she feels safe from being stopped or interrupted.”


A few weeks later they called to complain that the dog had started piddling under the coffee table. Did the dog choose that spot because she knew it was a place where her owners couldn’t see her, or because it was a place where her owners couldn’t grab her?


Field of Vision vs Line of Sight

I think dogs are actually more sensitive to following our line of sight than they are to where we point. This is based on the evolution of each species. When wolves hunt they need to pay attention to three things: the prey’s movement, the changing nature of the terrain, and what the other wolves are doing. If one wolf loses sight of the prey he may be able to pick up clues about it’s position from the gaze of a fellow pack member.


This ability to track the trajectory of moving objects, animate or inanimate, is clearly present in pet dogs as well. And it includes the ability to track objects that may momentarily go behind a bush, etc., by predicting the object’s angle, speed, etc., and, in some cases, by triangulating the object’s trajectory via the owner’s line of sight. This is a natural, automatic every day occurrence in most dogs.


But neither dogs nor wolves have fingers. So where in their evolutionary history would they have developed an ability to follow where humans point? And does this ability automatically mean that dogs have a theory of mind?


This brings us to one of the three studies Linda P. Case recently posted on her blog under the heading: Studies show that dogs can make decisions based on what they think humans know or see.


Case’s recap of this third study:


“A dog and a human were positioned on opposite sides of the barrier, and two identical toys were placed on the same side of the barrier as the dog. The dog was then asked to ‘Fetch!’ They [the researchers] found that the dogs preferred to retrieve the toy that both the dog and the person could see, over the toy that only the dog could see.”


I think this goes directly back to the evolutionary history of dogs and wolves, and the ability of both species to follow the line-of-sight of a hunting (or in this case play) partner. In other words, did the dogs choose the toy the person could see because the dog was capable of perspective-taking or because the dog was triangulating between a) his location, b) the location of his play partner, and c) a toy located in the dog’s and the owner’s line of sight?


Case: “An additional finding of this study was that the dogs were capable of this distinction only in the present, at the time that the owner’s view was blocked. When the researchers tested dogs’ ability to remember what the owner had been able to see in the past, such as a toy being placed in a certain location, the dogs failed at that task.”


They didn’t fail at anything. They showed the researchers something very important. They showed that for social animals—whether group predators or prey species—, developing an ability to follow the gaze of other members of your group, instantly, without having to think about what others are able to see or not see, serves an adaptive purpose, providing you with an evolutionary advantage. This carries with it another animalistic ability: the capacity to feel what other members of your group are feeling. If you’ve ever stared into the eyes of someone you loved or been stared at by someone with malicious intent, you know what I mean: eye contact and emotion go hand in hand.


Meanwhile developing a Theory of Mind—which is much less parsimonious—serves no such purpose and provides no such advantage.

Leave a Reply

Your email address will not be published. Required fields are marked *