Thursday, 7 February 2013

From Cognition to Dognition: does your dog possess Theory of Mind?

I came across an article on ‘Dognition’ yesterday on Scientific American, under the category ‘Mind & Brain’, and I was ready to dismiss it until I looked up the Duke University Canine Cognition Center (yes, it’s a real thing!), where I found a collection of fascinating papers on dog psychology and cognition (Tomasello even co-authored a few of them!). I will summarize a few of their findings, which I personally found pretty mind blowing.

In this paper by Brian Hare, it was suggested that dogs might be regarded in some respects as more intelligent than chimpanzees. Indeed, in a series of experiments conducted to test dogs and chimps’ ability to solve problems by relying on humans’ social cues, it emerged that dogs might be closer than primates in possessing a theory of mind. Hare defines T.o.M as the ability to think about the thoughts of others; it can be seen as providing the developmental basis for much of what is considered unique to human cognition.

The main experiment involved an ‘Object choice task’ in which the experimenter indicates to an animal subject the location of hidden food by using a social cue such as gaze direction or pointing. It was found that while dogs excelled at spontaneously using humans’ cooperative-communicative social cues (most dogs were successful from the very first trial and were able to adapt to arbitrary and novel cues throughout the experiment), chimps showed little spontaneous skill at using human social cues and even after a dozen trials were unable to generalize this skill when it was presented with a slightly modified social cue.

In Hare’s words, dogs should be viewed not only as humans’ best friends but also as ‘human tool users’. Can we conclude from this that dogs are using us as part of their extended mind?

Joking aside, the researchers suggest that dog’s unusual social cognition could be a result of domestication. This hypothesis is supported by several domestication studies. In one of them, a group of Russian researchers have been breeding two groups of foxes over the last 45 years. The ‘experimental’ foxes were selected for particular traits such as low levels of fear and aggression toward humans while the ‘control’ group was bread regardless of their behavior towards humans. They found that the experimental fox kits were as skilled at using human social cues as the dog puppies and performed significantly better than the control kits. This finding suggests that “domestic dogs, like Belyaev’s foxes, use cooperative-communicative social cues as a result of selection on levels of emotional reactivity during domestication (Hare et al., 2005). That is, dogs’ specialized social-problem-solving skills may have first appeared after systems mediating fear and aggression were altered.” (Hare, 2007).

These studies could have implications for our understanding of the processes responsible for human cognitive evolution. Indeed, research suggests that these specialized skills that dogs have evolved could represent a case of convergent evolution in dogs and humans. If so, could humans’ social and communicative abilities have evolved in a similar fashion as that observed in dogs, namely through the selection of certain traits and temperaments?

For more information about these studies, I refer you to Tomasello and Hare’s excellent paper, which can be found here.


  1. Fascinating! It's far too commonplace for philosophers of mind to dismiss animal intelligence, or even consciousness, out of hand without any particular basis. It seems to me a much less solipsistic starting point to assume the above of any entity that exhibits any form of skillful behaviour and let the evidence complete the picture.

  2. Those taking the Masters in Cognitive Science may be interested to know of Donal Cahill, who did our masters, and talked his way into (now discredited) Marc Hauser's lab at the same time. He is now studying dog psychology in Harvard.

    Fluff article partly about Donal:

    He is involved in the Moral Sense Test, about which you can read here:

  3. This is cool. I think your suggestion about dogs using humans as part of their extended mind is quite thought provoking, as it carries interesting implications considering the nature of dogs as "pack animals." Of course, though, this seems to undermine the theory that dognition developed evolutionarily beside domestication and converged with human social behaviors. On the other hand, perhaps it suggests that for the domestic dog, humans constitute "the pack." This would seem to imply that wild dogs, who are less attuned to human social cues (ala Belyaev's foxes), would be more attuned to the social cues of other dogs. Would this mean domestic dogs are smarter than wild dogs by virtue of the fact that the extended minds of the former use humans, while those of the latter use of other dogs? If not, would that not make the mind of a dog equivalent to the mind of a human?

    1. I think you're spot on with the idea that for our domestic dogs we humans are the pack but I think bestowing them with a theory of mind superior to that of wild dogs is a bit of a stretch! My dog serves no purpose whatsoever while her wild counterpart at least has some role in a larger ecosystem. If anything, the continued survival of her wild cousins within their pack shows an evolutionary superiority over my largely ornamental mongrel.

  4. Hi Rebecca, great topic! I came across this article last week on the topic of canine intelligence; which links to this piece about multiple intelligences by Maria Popova; which in turn borrows from Howard Gardner’s 1983 "Theory of Multiple Intelligences". According to Popova this theory "revolutionized psychology and education by offering a more dimensional conception of intelligence than the narrow measures traditional standardized tests had long applied"

  5. Ah, dogs, the only parasites we love.

    I experienced an unusual piece of dognition the other day. I was playing ball with my dog on the beach - the usual thing, she was putting it down in front of me (covered in sand and drool), I was reluctantly throwing it. Then the ball got lost in the water. She plonked about for a bit but couldn't retrieve it, and I wasn't about to get wet.

    Anyway, I put her on the lead and we left the beach. Here comes the weird dognition. Even after we were a full kilometre away from where the ball had been lost, she was still looking for it obsessively under every bush and behind every lampost. She seemed to believe either that the ball was capable of independent locomotion (perhaps, due to her overactive ToM, she even furnished it with a mind, the way I'm now doing for her) or else that something lost in one location can be found in another (maybe it's a quantum thing). Or perhaps she was just stuck in a behavioural loop.

  6. Keeping with the area of animal cognition, some of you may have heard that last week while speaking at the American Academy for the Advancement of Science in Boston, Professor Tetsuro Matsuzawa of the Primate Research Institute at Kyoto University has stated that chimps have much faster working memories than humans!! Here is a link with more information and a video showing a chimp named 'Ayumu' carrying out some WM tasks.

    Do chimpanzees really possess some memory capacity that human beings, for whatever reason, have lost? For more on Tetsuro Matsuzawa and his research click this link below: