Robyn Williams: One of humanity's first innovations was an alliance with dogs, 40,000 years ago at least. Just look at the article on wolves in this week's New Scientist magazine. One great advantage of wolves and dogs is they are smart, as Stanley Coren knows very well.

When I last talked to you it was about the intelligence of dogs and the idea that maybe the brightest ones come in at about 2.5 years of age in equivalence to children, human animals, and that was quite startling because 2.5 years, a bright kid knows quite a lot. How have you followed that up?

Stanley Coren: Well, you see, my major breakthrough…if I want anybody to remember me for anything, it's for the fact that I was the first one to suggest that one of the ways to study the intelligence of dogs was to take tests which were designed for human infants and modify them so you could study dogs. And furthermore, and this is a compulsion that psychologists have, you could turn the scores into mental age, which is how we found that, for example…this was in terms of language research, was when I found that by modifying a test of language comprehension we could show that the average dog was 2 to 2.5 years of age mental equivalence to humans. And the super dogs, the ones in the top 20%, were 2.5 to 3 years of age.

Right now there are a number of researchers who have taken up that particular line of thought in several different labs, and most recently some people have even started to look at the social reactions of dogs. So, for example, there is a test which is called the strange situation test. It's one in which a child is brought into a room and either they have to interact with a stranger or the caretaker leaves or there is something which is stressful. And you look at how the child reacts. And one of the things which we found is that young children start to use their caretakers as a safe base, safe haven, and finding, by the way, that dogs act very much like young children and do use their caretaker as a safe haven, which is quite wonderful.

Robyn Williams: By caretaker you mean the person looking after them.

Stanley Coren: That's right, it's the person we would call the owner or the master, whatever it is. You basically use the individual who is most bonded with the dog. And I've also been going back through and reanalysing some of the literature on the emotional responses of dogs and that kind of thing. And so, for example, emotions…there's a staircase, so different emotions develop at different times. So a young infant starts out and he only has this sort of excitement, you know, he's either excited or he's not. And then it becomes distress, it becomes the first of the negative emotions, and something which you would probably call joy or extreme pleasure is the first of the positive emotions. And then it goes up through all of the basic kinds of things. You can measure when disgust comes in and when surprise comes in, and you begin to move it up into anger and fear, which are all quite primitive.

But then there's a break which occurs maybe around 12, 14 months or something like that, and then you don't get much of an emotional differentiation until something after about in some cases 3 to 4 years of age, and that's when we start to show the social emotions, so guilt, shame, pride, those sorts of things.

And so when you go back and you look at that sort of thing, you can say, well, if our hypothesis is correct then dogs have all the basic emotions—fear, anger, pleasure—but they don't have all those social kinds of emotions, so dogs don't feel guilt. Of course I know a whole bunch of people who are listening to this are going to say, 'No, no, we had this really new white carpet and when we came home Lassie had redecorated it in earth and sunshine colours. And when we walked into the door she started slinking around because she was guilty. She knew that she had done something wrong.' Well, that's not true. What's really happening over here is that Lassie has learned that when you are visible and that stuff on the floor is visible, then bad things happen to puppies. And so she is not feeling guilt, she is feeling fear, she's afraid you're going to drop a piano on her head.

I was challenged at this at one point. I had a national TV show in Canada on dog behaviour, it ran for about 10 years. I think that some of the episodes were actually shown in Australia. Anyway, somebody challenged me on this sort of thing, and so I gave a demonstration in which I got the owner to go out of the house, and while the dog watched I went into the kitchen. So while Lassie watched I threw the trash out on the floor. And then we simply brought the owner back, and she went slinking off because it was not guilt, she knew, you know, 'That stuff is on the floor, I can see him, I'm going to die!'

Robyn Williams: Indeed, the reaction like that is quite well known, that the dog is sensitive to the fact that you might be cross. They don't know what it's about but they respond to your signals and go away and hide. So it's not guilt, as you say, it's something else. But recently there was a huge amount of global publicity to the fact that dogs, unlike many other animals, can perceive how we seem to be behaving, what our mood is, if you like.

Stanley Coren: They do read our emotions. In fact there's a wonderful research which has come out of Emory University which shows that in fact a dog's brain lights up in different places in these functional MRIs when they hear a positive note in a person's voice and when they hear a negative note. And those are exactly the same places which light up in the human being.

But we know even more. We know, for example, that dogs would not only read our emotions but attach our emotions to things. So if you are angry at someone, then the dog picks up the fact that you are not angry at them but you are angry at the person who you are directing your attention to. It's always very interesting to notice that some dogs become very shy of strangers and that sort of thing. And it's usually around people, usually women, who are also shy, feel uncomfortable around men. And you'll get the story from them, 'I don't understand, she just doesn't like men.' And it's not, it's this process that the dog reads your emotions, assigns that emotional response and emotional valence we would say technically, to whatever the objects are. And we can use this if we are training dogs, for example, to solve problems.

So, for example, what do you do if your dog is, let's say, chewing your shoe? The obvious thing is put your shoes in the closet so the dog can't chew them. But the problem here is that the event occurs when you are not at home. And so how do you control this sort of thing? Well, you don't beat the dog for it. What you do is you come home, you look at the shoes, while the dog watches, you grab a newspaper or something like that, and you beat the bejesus out of the shoes. And at that point basically a dog reads what's happening, and so the next time they encounter the shoes, even if you are not there, it's going to make them uncomfortable and they're going to go elsewhere.

So we could apply these kinds of things. I talked a moment before about this strange situation test. Well, you can use that too. So, for example, in one study dogs were brought into a room and in the corner there was a fan, one of these oscillating fans. And they put little streamers of green plastic ribbon on it. So these things were flopping around like that, going back and forth, and if a dog had never seen that sort of thing, it would look pretty frightening.

And then they programed how the dog's owner responded. Either the dog's owner responded with something like, 'Oh, that's interesting,' came over and looked at it, or, 'Oh, what is that?' And in fact the dogs then subsequently responded in very much the same way. If the owner gave positive things they actually came over to explore. If they get negative things, in fact they stayed as far away from it as they could.

Robyn Williams: Interesting, isn't it. And my final question, here are dogs responding in a way that practically no animal I can think of can do. And the big question is whether dogs have always been able to do that or whether somehow the decades, centuries of training, of adaptation to being with us has changed them from the wolf mind to the dog mind, and our selection has made them respond to us in the way you describe.

Stanley Coren: Well, that's easy, dogs are not wolves. You have to understand that for at least 14,000 years, and if we believe that maybe 17,000 years, since that time we have been systematically applying a seat-of-the-pants behaviour genetics to dogs. And it really works very simply. If a dog has certain characteristics which we like, then that dog is going to be treated better, more likely to be bred and more likely to be bred with other dogs which have similar characteristics. And one of the things which we prize is the ability to communicate and the ability to be empathic and that sort of thing. For example, if you point at something, that's one of the most primitive kind of communication gestures, the dog looks in the direction that you point.

Robyn Williams: Chimps won't.

Stanley Coren: Chimps won't. They will learn eventually to do it. And, by the way, even a puppy, an eight-week-old puppy will look in the direction that you point. But suppose you have a wolf who has been raised from a puppy in a human house, so he's got every opportunity to learn, and you point to something, the wolf looks at your hand, not at the direction. So the brain of a dog is different than the brain of a wolf, and I think that we have selectively created dogs.

And I think the reason why a lot of people consider dogs to be absolutely the finest companions that we can have in the animal world is because we built them that way. It's like over the years we built cars to be more and more people-friendly, well, we've built dogs to be more and more people-friendly, and that shows up. If you look at the older breeds of dogs, they are less empathic, they are much more likely to go off on their own business or seem to miss things.

So some of the hounds, for example, the reason that hounds are not particularly trainable is because we have changed them less than we've changed any of the other dogs. And they were originally supposed to be more or less solitary hunters, you know, they went out over there and all the human being had to do was to catch up to them before they ate whatever the game was, and there was no cooperation involved. That doesn't make them less wonderful as dogs.

I just recently lost a beagle who was an absolute love, but I have put many advanced obedience degrees on all of my dogs, and he did well for a beagle, but for a real dog…! Whereas some of the later inventions like the retrievers, they are wonderful. You put the dog training manual in their dog bed and by the time you wake them up in the morning they've already learned how to sit and go down and all that sort of thing and they are wondering why your signals aren't as precise as in the photos.

Robyn Williams: Stanley Coren is emeritus professor of psychology at UBC.