Scientists know surprisingly little about the mental abilities of the dogs we live with and love. Partly, this is because it is so difficult to devise ways of testing mental abilities without provides cues about what is going on. The horse Clever Hans is a case in point. Clever Hans was thought to be able to add and subtract numbers, stomping out with his hoof the correct answer when given a problem to solve. Careful observation showed that Hans’ owner was providing subtle cues to the horse, by leaning slightly forward as Hans was stomping his hoof, and then relaxing slightly and leaning back when Hans came to the right number of stomps. Hans’ owner was totally unaware that he was doing this, and was extremely disappointed when he found out that Hans was not really a genius at doing arithmetic. Similarly, when dogs are asked to pick up objects they have never seen before as a test of their language comprehension, they can use very subtle cues such as the direction of the gaze of the person asking the dog to perform, and follow that gaze toward the object that they are asked to retrieve.
A recent article in the journal Animal Cognition has come up with a clever way of testing the mental capacities of dogs. The Austrian researchers (F. Range, U. Aust, M. Steurer, and L. Huber. 2007. Visual categorization of natural stimuli by domestic dogs. Animal Cognition DOI 10.1007/s10071-007-0123-2) trained four dogs to use a touch screen monitor, and touch with their noses a picture of a dog when they were given a choice of a dog picture or a picture of a landscape, getting a food reward when they made the right choice. The monitor was positioned so that none of the human observers could see the choice that the dog was making, and so could not cue the dog.
The pictures that the dogs had to choose were in color. For those who think that dogs do not have color vision, this is not the case. Dogs have dichromatic color vision, with peak sensitivity in the indigo-violet (429 nm) and yellow (555 nm) range. This compares with the trichromatic color vision of humans, who have peak sensitivities in the violet (420 nm), green (534 nm) and yellow (564 nm) range. Exactly what colors dogs see with their vision is still not entirely clear.
The dogs were shown a variety of dog pictures, including just dog faces, dogs lying down, and dogs standing, and they had little difficulty in distinguishing the dog pictures from the landscape ones. Also, when the dogs were shown pictures of a dog standing in a landscape picture that previously had no dog in it, the dogs had little difficulty in deciding that this was really a picture of a dog, rather than a picture of the landscape. This suggests that they had a mental concept of a dog.
One the one hand, we can look at these results and ask, what is the big deal here? My dog gets excited when he is in my car and sees another dog. This could be a St. Bernard or a Pomeranian. He does not get excited when we go past kids on tricycles, or past fenceposts, cows, horses, squirrels, or landscapes of stunning scenery. From this I conclude that he has a concept of a dog.
However, the big deal is that my conclusion is something that I can’t prove. Maybe he gets excited because I expect him to bark at a dog, and I make some kind of subtle movement that he picks up. Maybe I slow down just a tiny bit so that I can see the dog better, and he picks up on that. There are a whole host of maybes that could be true.
With this study, the maybes seem to have been eliminated, and we are left with the inescapable conclusion that dogs have mental concepts of other dogs.
--Con Slobodchikoff
Now that it has been proven that dogs have mental concepts of others dogs, it stands to reason that they also could have mental concepts of other objects as well, i.e., people, other species of animals, objects. At least, now there is a reliable way to test this hypothesis and see how far it goes. What about adding emotional states, like happy dog/sad dog or even happy human/sad human? And then there’s the elusive question of self-awareness. From what I’ve read, dogs don’t do well on the ‘mirror test’for self-recognition and this has led some investigators to conclude that they aren’t self-aware. (On the other hand, apes, dolphins, and elephants—all big-brained mammals—have done well on the mirror test and have been pronounced as having self-awareness.) However, I’m not satisfied that dogs lack self-awareness just because of their performance on one kind of test. If they have a mental concept of other dogs, then by logical extension, they should have a concept of themselves as individual dogs, distinct from other dogs, and that, to me, suggests a sense of self. There ought to be a way to scientifically test this notion. Perhaps the experimental design used by the Austrian researchers could be applied or adapted to determine whether--or to what extent-- dogs are self-aware.
Posted by: Randall Johnson | May 21, 2008 at 04:59 AM