TERRY GROSS, host:

What's it like to be a dog? My guest, John Bradshaw, says that some of the ways we interact with and train dogs are based on the false premise that - false premises about how dogs experience the world.

And this is what we're going to talk about today on FRESH AIR. Bradshaw studies the interactions of humans and their pets, as well as working dogs. His new book is called "Dog Sense: How the New Science of Dog Behavior Can Make You A Better Friend to Your Pet."

Bradshaw is the foundation director of the Anthrozoology Institute at the University of Bristol in England. His current research partners include the group's medical detection dogs, and the Royal Society for the Prevention of Cruelty to Animals.

John Bradshaw, welcome to FRESH AIR. Now, you say that a lot of dog training is based on the false premise that dogs are like wolves. What are some of the ways that wolf behavior has been projected onto dogs?

Mr. JOHN BRADSHAW (Author, "Dog Sense: How the New Science of Dog Behavior Can Make You A Better Friend to Your Pet"): Well, the main things that's been projected from wolves onto dogs is that wolves are essentially an intrinsically aggressive animal; that is, continuously trying to take over whatever group they find themselves in, and dominate it.

And the new wolf biology has really exposed that as an artifact. That particular view of wolves came from studies of wolves in zoos and in wildlife parks, where a bunch of unrelated wolves were put together and essentially, told to get on with it. And not surprisingly, they got on with it by being aggressive towards one another.

The new picture of wolf society is that wolves are very harmonious animals. They live in family groups. They get along really well together, and they're almost never aggressive to one another. The aggression comes out when two families meet. So they have very strong family ties.

GROSS: So give us an example of a training approach for dogs that's based on what you describe as the misguided premise about wolves - that wolves want to be dominant, that they're aggressive.

Mr. BRADSHAW: OK. Let's take a very simple piece of advice that some trainers hand out, which is that you should never allow a dog to go in front of you through a doorway because it will give the signal to the dog that you are submissive and that you are, therefore, allowing him or her - the dog - to become dominant.

Take another one: Many trainers advise against playing tug-of-war games because there is a risk that the dog will win. And the dog, by winning, will then think that you are being submissive and that he will, therefore, be able to control you in the future.

Now, we've done research into a number of these things, including the tug-of-war game, and have shown that the premise is just completely not true. If you do let a dog win over and over again at tug of war, it likes you. It wants to play with you more than it did to begin with because it's having fun.

If, on the other hand, you always win, the dog gets kind of slightly less attracted to you and doesn't want so much to play with you again. But there's absolutely no change in the dog's behavior, outside of that particular situation of play. The dog does not get into its head that you're some kind of a soft touch, that in the future, it will be able to control you and whatever you do.

GROSS: Well, what about things like you shouldn't let a dog, like, lie on your bed because your dog is not your equal?

Mr. BRADSHAW: The advice that we give out, and the wolf research points to, is that having - if you want to have your dog lying on your bed, I mean, that's your choice. It won't make any difference to the relationship, in terms of whether the dog will obey you or not. It's simply a matter of personal preference.

GROSS: What do you think about choke-chains?

Mr. BRADSHAW: Well, choke-chains, I don't think, really do any kind of good. I mean, there is a danger in some dogs, where there is some kind of weakness in the neck, that the choke-chain can actually harm the dog.

Simply producing some discomfort or even pain at particular times in the training schedule, the evidence says, confuses the dog rather than sharpening it up.

GROSS: How does it confuse the dog?

Mr. BRADSHAW: Well, it gets - you know, it generates its own discomfort by pulling on the leash. I mean, some owners will actually yank the leash. I mean, that is kind of the next stage and is even less advisable.

But the owner is not in control of what the dog is learning in that case. The dog is dictating, really, although the whole thing becomes completely arbitrary. The dog is deciding to pull on the leash. It then hurts a bit. It doesn't really know why it's hurting or what to do to avoid it, and you can see.

I mean, dogs with choke-chains, who's had choke-chains on for years, will still pull away at them. It hasn't really learned - it hasn't really taught them anything.

GROSS: So a lot of people, even if they don't want to be punitive toward their dog, or they don't feel the need to be like, the alpha owner - if the dog is misbehaving, a lot of people revert to punishment for the dog or, you know, the choke-chain - that kind of command that you were talking about before, that the human has to walk out before the dog does to prove that the dog isn't the alpha member of the group; the human is.

So do you advise against doing this kind of training, even if your dog is misbehaving, even if the positive reinforcements haven't worked?

Mr. BRADSHAW: Well, I would advise that there is a particular kind of punishment, which is not only sensible to use but also almost unavoidable. It's not punishment in the physical sense. It is punishment in the mental or psychological sense, and it's basically just a withdrawal of attention.

Most dogs - there are a few exceptions, and they are very difficult to train -but most dogs, most pet dogs, require their owners' attention. They want their owners' attention. They want people's attention in general. And withdrawing that is a very powerful signal to the dog.

So if you have a dog that jumps up on the visitors that come into your home, then the best advice is not to slap the dog while it's doing it. It may actually perceive that as a reward, and it may in fact make that - because it's a form of attention. So the dog may actually do it more after you've slapped it, rather than less.

But if you get your visitors - and this does require a bit of - kind of advanced planning - if you get your visitors to ignore the dog, look away, fold their arms, not pat the dog even though it's jumping up at them, then you'll find that quite quickly, the dog begins to realize that this is not working.

You can then use a distraction technique to get the dog to do something else, so that as soon as the dog's attention is away, you get it to do something you've already trained it to do, like sit or lie down. And then it will get the idea that this is what it's supposed to do when visitors come, and not to jump up.

But simply punishing a dog, especially in front of visitors - most people won't want to punish the dog too severely, and the dog could easily mistake a mild punishment, actually, for attention and therefore, go on to repeat the action over and over again.

GROSS: So you're talking about training your visitors.

(Soundbite of laughter)

Mr. BRADSHAW: You have to train people to understand dogs, yeah. I mean, we have to take the responsibility. We're the - dogs are very smart, but it's the humans that are the smart one in the relationship, and we need to take responsibility for that.

GROSS: So your training approach is based on the premise that dogs want to please people, that dogs like people, especially the people they live with. They want to please those people. They want to play with those people. What have you learned scientifically about dogs' minds that might help explain why dogs want to please the people who love them?

Mr. BRADSHAW: Well, what we know is that domestication has changed the wolf's mind really substantially. What we think is that the wolf has a very sophisticated sense of social life, of family, of family connections and so on.

And if you look at studies of dogs which are allowed to live like wolves, you find that they don't really live like that. For example, virtually any adult female dog living wild in a village, for example, will breed, whereas in a wolf pack it's only one, the most senior female, who will breed.

What we've replaced that very sophisticated wolf behavior with is a very strong ability to learn about people right from the minute the puppy's eyes open - and particularly strongly, goes on until about six months, nine months of age.

But it does, indeed, persist throughout life. And you see it kind of coming back in a dog that has to be re-homed, that it will re-learn its relationships very easily.

So that's a very powerful attraction and surprisingly, it's even more powerful than the attraction to and preference for other dogs. Most dogs, given the choice - and there are a few exceptions here, like hounds that hunt in packs - but most dogs, given the choice, will actually prefer human company to other dog company.

That doesn't mean they don't enjoy being with other dogs, but humans are the social partner of choice.

GROSS: Are you suggesting that dogs, when they're puppies, actually like, study human behavior and try to learn it?

I think study implies a kind of conscious effort on their part, and it isn't -we don't think it's effortful at all. It's absolutely natural to them to want to do this.

Quite how that has been put into the dog's developing brain is still a mystery, but they do have a very - an exaggerated tendency to learn from anything that people do - say, right from the minute they're capable of doing it.

And they're particularly sensitive - or become particularly sensitive to human body language, to the way we - the direction we look in, what our whole body language is telling them, pointing gestures with the hands, and so on. They are much more sensitive to things like that, once they've learned them, than almost any other species on the planet.

GROSS: If you're just joining us, my guest is John Bradshaw, author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You A Better Friend to Your Pet." And he's the director of the Anthrozoology Institute at the University of Bristol in England.

You write in your book about the problems of being a modern dog, problems dogs have today that they didn't traditionally have. What are some of those problems?

Mr. BRADSHAW: Well, the dog used to be a working animal, and it served mankind just exceptionally well in a huge variety of ways - and in a variety of ways that no other animal has ever done.

For, well, at least 10,000 years and probably longer, the roles obviously have changed, and they have not just changed but also broadened over those years. And they - the dog has done, you know, mankind a great service over that time.

But although dogs have always, I think, been companions - and I think that's part of the story of domestication, is that we just like having them around; they're not just tools. But although they've always been companions, that's usually been a secondary role, except for the very rich, perhaps 100, 150 years ago.

A dog had to work for its living, and so we derived a whole lot of different kinds of dog, and I don't mean breeds here because I'm talking about before the modern breeds were set up.

We derived a whole lot of kinds of dog to fulfill particular functions: dogs for guarding, dogs for herding, dogs for retrieving game, and so on and so on.

Now we have really replaced many of those. I mean, dogs still work. There are still plenty of working dogs in the world. But most dogs in the West are companions. They're there to provide friendship and companionship to humans and participate in activities which are, you know, largely human-focused.

So what I think hasn't happened - and there are a number of people around the world who are kind of getting together to try to get this onto the agenda, notably in Australia as well as in Europe - is to think about a dog as a companion, first and foremost. And I think it's...

GROSS: So would you like to see certain reforms in breeding so that breeders are breeding for behavior and not just for looks in the show ring?

Mr. BRADSHAW: Well, I think there are more problems in breeding than just breeding for behavior. Breeding for the show ring has generated very narrow specifications, and very great restriction in the dog's gene pool in each breed.

There's still plenty of genetic variability if you take the dog as a whole. But within a breed, the variation has diminished. And so you get all kind of inherited diseases coming up and being very difficult to eradicate at the moment, while the breed barriers are being maintained. So...

>GROSS: Is breeding, in some ways, like sleeping with your brother or sister? Do you know what I mean? Is it...

Mr. BRADSHAW: Well, yeah, I know exactly what you mean. I mean, the analogy is the - I'm not a geneticist, but the geneticists tell me that most dog breeds, except the very popular ones where there are several different lines - like the Labrador retriever, where you have a show type and a field type, and that sort of thing - but the less common, the more specialist breeds, the degree of relatedness within a breed is about the same as human first cousins, and can be less.

So you can imagine a human family where nobody - none of the husbands or wives were less closely related than first cousins. And that persisted, that inbreeding persisted over many generations.

You can see - you know, we know from human history that that is a bad thing to do, and we know from animal breeding history that's a bad thing to do. All kinds of mutations that previously would have been hidden emerge and seriously affect the breed - like the English bulldog, for example, where the pelvis is not wide enough for the head to get through, and so every single puppy has to be born by Caesarian.

I think we're getting - you know, that's taking it too far. You know, what is the point of generating an animal like that, other than to satisfy some kind of craving on the part of the owners and the breeders for something different? Do we have the moral right to go that far for an animal which is, you know, basically he's going to be a companion.

GROSS: Well, let me take a pause here and tell our listeners who you are. My guest is John Bradshaw. He's the author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You A Better Friend to Your Pet." He is the director of the Anthrozoology Institute at the University of Bristol in England, and he studies animal-human relationships and behavior together. Let's take a short break here. Then we'll talk some more. This is FRESH AIR.

(Soundbite of music)

GROSS: If you're just joining us, my guest is John Bradshaw. He's a biologist. He studies the interactions between pets and their owners. He's the author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You A Better Friend to Your Pet."

Now, let me quote something that you say in the book that I found really interesting. And you're talking here about some of the problems of being a dog in the 21st century. And you write: We expect them to be companionable when we need them, and unobtrusive when we don't. As for city dogs, we expect them to be better behaved than the average human child, and as self-reliant as adults. Would you expand on that thought for us, and on the problems these expectations are creating in dogs?

Mr. BRADSHAW: Well, I think the main problem that we have with the modern dog, and indeed - and I think it is a modern problem - is that we expect them to be quiet and peaceful when we leave them alone. And that doesn't work, in many cases.

The research that we've done, and others have done, has shown that many, many dogs - maybe as many as half the dogs in the West, you know, Western civilization - that are kept in homes have a real problem with being left alone, at some time during their lives.

And the problem isn't just once. It happens, may last for weeks or even months and sometimes, it goes on for years. Now this is - if you think about it just for a moment - is the flipside of breeding an animal like the dog, which is very -becomes very attached, and very easily attached to people.

They crave the company of people. They also have a mind which does not have a particularly good sense of time. And so when they get left alone, they can immediately begin to think: When's anyone coming back? Have I been abandoned forever? And they get very anxious as a result.

Or they may be OK for a while after they're left alone, but then something happens that scares them - like a gunshot going off in the distance, firecracker, something they really can't account for. And then they immediately look around for some kind of company to reassure them. And there's nobody there, no human there to reassure them, and so then they panic, even though before that they might have been sleeping peacefully.

So those two thing - those two varieties of that particular thing combined, we call separation disorders. And as I said, they're extremely common.

Now, they're also extremely preventable if you get to them early enough. And so that's one of the things that I've put in the book, one of the few bits of actual training advice that I put in the book because I think it's so important.

You train your dog to toilet outside. You train your dog to sit on command. You should also train your dog to cope with being left alone. And it's a very simple...

GROSS: How do you do that?

Mr. BRADSHAW: Well, it's a very simple thing to do, as I say, provided the dog is not yet anxious about being left alone. So, you know, as soon as you get a new dog, whether it's an adult or a puppy - the adults may take a bit longer because they learn a bit slower but - and they may, indeed, have been anxious when left alone in the past, so they may need extra reassurance.

But what you do, very simply, is you do all the things that you would normally do before you go out because those will become the dog's triggers. The dog will associate all those things - like picking up the car keys, putting on a coat, those sorts of things - with your absence a few seconds later.

So you do all those things. You go to the door. You come back from the door. You put the coat back on the rack. You put the car keys back down on the shelf. Then you do it again but this time, maybe you open the door, and then you close the door and put the things down.

Then the next time, you can go outside the door and come straight back in again. And the next time, you can go outside and stand on the step outside for 10 seconds, and then come back in again.

And what the dog learns through that, provided it doesn't panic at that stage -and if does start panicking, then you may need to call in expert help because this is a dog which has had some serious problem in the past; you do see this occasionally in rescue dogs.

But in most dogs - will learn very quickly the association between you going out and you coming home. And that is enough for most dogs, to reassure them. And so you can then start leaving them for longer and longer periods, and very quickly you find you can leave them for hours.

They've just learned that association between you going out - all the things you do when you go out - and you coming back and making a fuss of them. And that's a good thing. So the idea of you going out actually becomes pleasurable rather than something that causes them to panic.

GROSS: John Bradshaw will talk more about dog behavior in the second half of the show. His new book is called "Dog Sense." Bradshaw is foundation director of the Anthrozoology Institute at the University of Bristol in England. I'm Terry Gross, and this is FRESH AIR.

(Soundbite of music)

GROSS: This is FRESH AIR. Im Terry Gross, back with John Bradshaw, author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You a Better Friend to Your Pet." He studies the interactions of people and their pets. He also studies working dogs. Bradshaw is foundation director of the Anthrozoology Institute at the University of Bristol.

Working dogs today, some of them are so remarkable. And I'm thinking specifically here of some of the military working dogs. I don't know if you've been reading about the Navy Seal working dogs. And these dogs, they wear - like, some of them wear a $30,000 bulletproof vest that has a camera on it so the dog can walk ahead of the actual Navy Seals. And the camera will transmit back to the humans the images that the dog is seeing. And then the dog also has audio equipment on his vest so that the humans can speak messages to the dog.

And I think it's just like, so remarkable. And these dogs can parachute out of helicopters. I'm not sure whether the - I assume that a Navy Seal person is holding the dog as it parachutes. But to think of a dog flying through space like that and not totally freaking out, it's such remarkable training that must go into this. Have you been reading about this in amazement? Of course, there was the bin Laden dog. We don't know exactly what the dog did, but I'm sure it was amazing.

Mr. BRADSHAW: Yes, I've been reading about it with great interest. In my case, not with amazement because I've been involved in advising the military on training of dogs in those kinds of things...

GROSS: Really?

Mr. BRADSHAW: ...for a decade or more since really, since before 9/11. So I think everybody but me has been surprised by the dog that went in to help to find Osama bin Laden. Yeah, I mean, they're very valuable dogs. And I must say, if I was in an environment like that, I would actually much rather have a dog ahead of me than another human being because, you know, it's another set of senses - and particularly, the olfactory sense. These dogs are trained to detect and then indicate all manner of things. In that particular instance, it would presumably be explosives and ammunition and guns, and so on. But you know, in many other, there are many other applications for this particular olfactory ability that dogs have and, you know, and we don't.

One of the ones that I have been particularly intrigued by is that conservationists are now using dogs to monitor the populations of very rare animals because the dog, with its nose, is able to tell the difference between the feces of one species and another. And so they literally just let the dogs run around in the habitat where these animals are thought to be, but nobody's seen them for months. And the dog will say yup, there's one of them went by here yesterday; you can see it by the poop.

(Soundbite of laughter)

Mr. BRADSHAW: So - and you know, the number of times and the number of locations they do that, they're able to map animals, which unless they had a radio collar on, nobody would be aware it was even there. So dogs are, you know, their noses are so valuable to us. And I think it's just a - almost the limits of human ingenuity and imagination that we haven't probably even yet tapped into all those possibilities. But theres certainly a huge range of them out there already.

GROSS: Well, I'm very interested in the fact that you helped advise the military on their working dogs. So without betraying any military secrets, can you share with us some of the insights that you shared with the military in training dogs for combat or for sniffing explosive devices?

Mr. BRADSHAW: Id imagined, as I think probably many people would, that the military would use the same kinds of methods to train dogs as they do to train soldiers, which is to put them through hell. And quite the opposite, they don't. I mean, most of the military dogs are - virtually every dog that I've ever come - military dog I've ever come across or similar dog in the public service, whether it's, you know, looking for narcotics in prisons or checking out whether a house fire has been caused by arson or not, dogs are used for all those things. And the vast majority - in fact, every dog I've ever seen - has been trained with positive reward. And that's what kind of really woke me up to the idea that, you know, if the military, who are, you know, the hardest, toughest, most macho guys around, can train dogs exclusively to do these tasks with reward-based systems then surely, everybody else can.

And they use different kinds of reward. Some of them use food rewards. But most of them use the bond with the handler as their reward. The reward they get is a tennis ball thrown by the handler, or some kind of game with the handler. And the bond between dog and handler is a joy to behold because they genuinely get on.

I was just looking at some photographs that appeared on the Internet recently, around this whole Navy Seal business. I saw there was a couple of soldiers, American soldiers, resting in the shadow of a wall somewhere in Afghanistan with their sniffer dogs, their Labradors, lying on top of them. You know, and this whole thing about, you know, you shouldn't take your dog to bed with you or it'll come to dominate you, if those Navy Seals can literally let their dogs lie on their legs and go to sleep while they rest themselves, surely we should be able to allow our dogs to do that at home.

GROSS: What other insights have you gotten, or what other insights have you shared with the military?

Mr. BRADSHAW: Well, we've done some studies on how to get the best dogs. You can imagine that since 9/11 in particular, the demand for sniffer dogs of all kinds has gone up many time many fold. And so there has been a real supply problem over that time, or there could have been a supply problem. And so we put in place a monitoring program to look at the best way to raise these dogs, but also how to evaluate them. What's the best time to evaluate a dog for training in the military? Theyre very, very expensive dogs once they come to be fully trained. The whole training program can cost several tens of thousands of dollars. They are, therefore, very well-protected when theyre in a combat zone. And clearly, if you're going to put that amount of investment into an animal, then you need to be absolutely sure - as sure as you can be -that that investment is going to pay off. You're going to generate a dog that is really going to protect its handler and all the people around it.

So we had a long look at puppies from the age of eight weeks old right up to the age of a year or more, where the dog is ready for - the earliest they're ready for training, and monitored how we could - how soon it was we could detect whether they were going to become a good training dog or not.

GROSS: Now, you talked earlier about some of the problems of inbreeding in dogs. Of, you know, purebred, pedigree dogs. From what I've read, it sounds like the military likes to use Labrador retrievers and German shepherds. Do you think those dogs are purebred? Or would purebred dogs create problems because of the inbreeding?

Mr. BRADSHAW: Those are purebred dogs. And the other breed that they use a lot of is the Belgian Malinois, which isn't a very common dog, I believe, in the United States. And I think they get all of those dogs from Europe. They're all breeds which have still got a very healthy gene pool. They're very popular breeds in, obviously in their respective countries. And they're so - they don't suffer - to the same extent, they don't suffer from inbreeding that maybe some other breeds do. That's not to say that German shepherds don't have their problems. You have to be careful about the lines they have. Some of the show-bred German shepherds have - have had hip problems for a number of years. But the German shepherd breeding club - in the United Kingdom, anyway - has taken very strenuous steps to try and eliminate that, and to change the breed standards so that the bad hips are bred out. So I think there occasionally are problems with those particular breeds, but not too often. The Labrador retrievers are taken from the field-trial lines of Labradors, not the show lines.

GROSS: Mm-hmm.

Mr. BRADSHAW: And those animals, you know, have been bred for generations to work with people shooting - and are very healthy.

GROSS: Why are Labrador retrievers, German shepherds the best dogs for the military?

Mr. BRADSHAW: Well, I think that actually, its just a question of the size of the dog. They don't want a dog that's too small or too large. Military people like working with German shepherds and similar dogs like the Malinois because they can be dual-purpose. They are quite scary looking dogs, and many of those dogs are trained to detain suspects as well as to sniff out contraband -explosives and so on. So that makes a very good dual-purpose dog. And the Labrador - and in this country, we use a lot of spaniels - are better dogs for working environments where you don't want to frighten people - like airports, for example. You know, people just don't mind too much having a cocker spaniel sniff around their luggage or - and their person, come to that, whereas they might object if it was a big German shepherd, with big teeth. So I think those are really the reasons.

I'm involved with a charity over here called Medical Detection Dogs, which places dogs with people who have particular issues like epilepsy, like diabetes or brittle diabetes - where they find it very difficult to predict the next time they're going to have an attack. The dogs are exceptionally good at this. We don't even quite know how they do it. They probably do it due to a change in the - slight change in the body odor of the person. But recently, we trained a little toy dog called an Affenpinscher - which is a little, tiny handbag dog with a squashed face. We thought, you know, this dog probably can't smell anything. I mean, his nose has been bred to be tiny. But it's brilliant at its job. I mean, there really doesn't seem to be - the dog sense of smell is so good that even in a dog where you'd think it wasn't that great, it's good enough to vastly outperform any human being.

GROSS: So these dogs can smell biological changes?

Mr. BRADSHAW: Well, that's what we think. I mean, the research is still going on but - and some people think it could be, also, a change in - just a very subtle changes in body language...

GROSS: Mm-hmm.

Mr. BRADSHAW: ...that, you know, even other humans can't pick up in the household, but the dog can. But some of them must be relying largely on smell because we now have dogs that will wake people who are about to go into a diabetic coma in their sleep - which is, obviously, the most dangerous time because the person has no warning of it. They go from sleep into coma without ever waking. The dog can detect the changes. Now, that person is not moving. And so we think it must, or must inevitably be the smell of the person that does it. We train them on smell of diabetics so - or epileptics. So the primary method, I think, of identification must be smell.

GROSS: Dogs are so amazing. They have this incredible sense of smell that humans don't have. They have hearing that humans don't have. They see differently than we do. So they're perceiving the world in a pretty different way than we are, don't you think?

Mr. BRADSHAW: Yes, they are. I mean, it overlaps with ours a lot, of course. Otherwise, we kind of couldn't share their world very well, or they couldn't share ours. So their vision overlaps, pretty much, with ours. They're colorblind to a certain extent. But I mean, colorblind humans are not that badly handicapped. Their hearing is a little bit more sensitive than ours in the high-pitched region. And it's their sense of smell that really distinguishes them from us.

And I think we don't really take up too much recognizance of that. I mean I think dogs have a right - if you believe dogs have rights at all, and I do - to sniff things whenever it's, you know, whenever it doesn't cause a problem to us. When I meet a dog, I hold my hand out. I don't stick my fingers right out, just in case, but I just make a loose fist and hold my hand out to the dog. And I'll get - squat down if it's a small dog. And that dog will want to come and sniff my hand and lick it, if necessary. I mean, I could always wash my hands off afterwards if I've got a problem with it. But that's a greeting, and I think if we don't do that, it's as upsetting to the dog as if we were talking to somebody that we never met before and covered our faces at that point in time, as if we were trying to disguise who we were.

And I think the dogs don't suffer, necessarily, by not being able to sniff everything, but I think that is their world. That's the world they live in; sight is secondary. We should acknowledge that in the same way that we make our own lives - base our own lives around things we can see.

GROSS: So if you extend your fist and the dog not only sniffs it but licks it, is the licking a sign of affection, or is it just further investigation?

Mr. BRADSHAW: Well, I think it's a mixture of both. I mean, dogs wolves lick one another. The young wolves will lick over their parents as a sign of affection so I think there's that in it, too. But also, dogs have a secondary sense of smell, which we think they use their tongues to sort of get it working. They have another nose, which is between the roof of the mouth and the nose itself. And it's got two little ducts that open out just by the front teeth. And what they seem to do, they will lick something...

GROSS: Wait. This is like, inside their mouth?

Mr. BRADSHAW: This is just above the mouth.

GROSS: They have a nose inside their mouth?

Mr. BRADSHAW: Just above the mouth. Between the palates - between the hard palate and the nose, there is a space. We have - we used to have that organ, but it got lost somewhere in primate evolution. Most mammals do have it - cats have it, horses have it, cows have it. But humans and monkeys don't. And dogs and many other animals will lick something, and then theyll flick the saliva, which has got some odor in it, up into this little pouch above the roof of their mouths. And that seems to store information about - or analyze information about what that person smells like in a way that complements what's coming in through the conventional route, the nose - the one we understand.

GROSS: If youre just joining us, my guest is John Bradshaw, the author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You a Better Friend to Your Pet." And he's a biologist who studies the interactions between pets and their owners.

Let's take a short break here, then we'll talk some more.

This is FRESH AIR.

(Soundbite of music)

GROSS: If youre just joining us, my guest is John Bradshaw, author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You a Better Friend to Your Pet."

How many dogs, and how many cats, have you had over your life?

Mr. BRADSHAW: I've had - well, I've had many dogs; four dogs that I kind of owned, and many other dogs that I've had a lot to do with. And I've had about seven or eight cats. I find them kind of equally fascinating. But you know, when you're a scientist studying these animals, you get to meet an awful lot more animals than you would if you're an average dog owner. But nevertheless, I mean, over the course of 20-odd more years that I've been studying dogs, I never cease to be surprised by things that I come across. A dog is such a diverse and interesting animal, and comes in so many different shapes and sizes and temperments.

GROSS: Would you describe the difference in the relationship you have with the dogs who've been in your life, the dogs you've owned - I don't mean the research dogs - and the cats who you've lived with?

Mr. BRADSHAW: Well, they're very different animals. I think the primary difference for me as a biologist is to say, well, a dog has a territory that moves around with its owner, and a cat has a territory and it's not really too bothered whether its owner is in there or not. Cats are very location focused. They're territorial animals and not nearly as domesticated as dogs are, and just by the fact that they're pretty well adapted to living with us. So they do all kinds of things which - and that presents them with problems too - that they do all kinds of things which are very different to a dog.

The main thing that arises with cats is because they're so territorial - that they get into disputes with neighboring cats, and they don't seem to be able to resolve them very easily. And so the main problem with a cat, as I see it at the moment, is with people keeping cats in cities, especially - is that they're too crowded. They end up being constantly stressed because they're having to be so vigilant to other cats that are coming over the fence into the backyard.

GROSS: A lot of cats aren't even allowed out of the apartment or the house.

Mr. BRADSHAW: I think if you've got an indoor cat, that's fine. The same with a dog. I mean, I don't think they necessarily miss what they've never had.

GROSS: Mm-hmm.

Mr. BRADSHAW: A cat that's been allowed to roam around and is then, for some reason, then kept in an apartment, I think will suffer, at least initially, as a dog would. But if you raise a kitten in an apartment and you give it plenty of stimulation, I don't think they're particularly worried about how much space they have. I think it's the quality of that space that matters.

GROSS: So have you studied cats as well as dogs?

Mr. BRADSHAW: Oh, absolutely. I mean, yeah, I think they're equally interesting animals. I'm kind of fundamentally a biologist who - one of the few biologists who thinks the animals we have around us are equally fascinating as the animals that live, you know, in the Serengeti in Africa. And, of course, the added level of complexity and interest is that here we have a relationship between two very different species - the human and the dog or the cat. And how that relationship works and how, indeed, it doesn't work are things that I have been studying for 20 years - and I'll go on studying.

GROSS: Now, one of the things you point out about a dog's brain, I think is probably also true of a cat's brain. And that is that they live in the moment, and it's hard for them to make connections between something they did two hours ago and a punishment you're going to administer when you get home later. They don't think about the past; they don't project into the future - which maybe makes training a little more challenging. But I think that's one of the things we love about our animals, is that they are so in the present, they are so in the moment.

Mr. BRADSHAW: Yes. I mean, the sense of time is not, you know, it's not as sophisticated as ours. They don't seem to think into the future. They will do things that seem to fit them for the future, but they're probably just kind of pre-programmed. Whether they really ever think about the past, I think, is something that we don't know enough about yet. I mean, do they actually have imagination in the way that we do? We know they dream because you can measure the brain waves and the movements of the animal, and so on, which are very similar to the sorts of things that go on in humans when we dream.

GROSS: Do cats dream, too?

Mr. BRADSHAW: And cats dream as well. Yeah.

GROSS: Really?

Mr. BRADSHAW: Yeah. Yeah. So what we don't know is whether they remember their dreams when they wake up, and we tend to remember the last dream we had in the night, when we wake up. We don't know whether a dog or a cat will do that or whether it is purely a, you know, a mechanical refreshing of the brain that goes on, and the animal is never aware of it in the way that we are aware of some of the things we dream about. But it does point to the possibility that dogs and cats are capable of some sort of limited thoughts about the past. They're not simply little robots programmed by training to do particular things, that they have minds of their own. I mean, they can count, for example. They can count small numbers. Both species can count. So if they can do that, you know, they have a level of cognitive complexity which is not totally robotic at all.

GROSS: Well, John Bradshaw, thank you so much. It's been great to talk with you.

Mr. BRADSHAW: It's been a pleasure. Thank you.

GROSS: John Bradshaw is the author of the new book "Dog Sense: How the New Science of Dog Behavior Can Make You a Better Friend to Your Pet." You can read an excerpt on our website, freshair.npr.eu.

Copyright © 2011 NPR.