Theory of attention


An article in Psychology Today – psychology for people who know nothing about psychology (like me), or more politely for the wider public – describes research that suggests that dogs have a rudimentary Theory of Mind, or one piece of a TOM.

Experimenters tested whether dogs beg more (or less) from people who (they can see) can’t see them or from people who can. It was the second. This of course causes not the faintest surprise in anyone who’s spent any time with dogs. Of course dogs cue in on human attention. I make eye contact with dogs in the street. I smile at them. They react. And these are stranger dogs. Of course dogs pay attention to human attention, including their line of sight.

And so the experiments confirmed.

Stanley Coren sums up:

Over all these results suggest that dogs do have a strongly entrenched, perhaps genetic, predisposition to try to “read the mind” of humans, at least at the level of understanding that humans must pay attention to them if they are going to be able to get them to respond. Wolves can learn this, at least at a low level for the most obvious of cues (when an individual has their back turned) but not for the more subtle cues, such as the familiar situation where a human is looking at a book and thus not attending to what is going on in front of them.

Or when she is looking at a computer screen and thus is not attending to the dog. Cooper starts awake when I put my reading glasses down on the desk. Yes, he knows from attention all right.

So, can dogs read your mind? Well they certainly seem to have a theory of mind and they are trying to figure out what you are thinking so that they can communicate with you and get a bit more of what they want out of life.

But do they have an elaborate enough theory of mind that they think about the feral cat outside who might be hungry and thus might want some of their food? Do they have an elaborate enough theory of mind that they then decide to leave some of their precious food behind for that possibly hungry cat outside who might come inside and eat it?

Uh, no.

Comments

  1. Wylann says

    Cats do the same. Our cat, Pixel, even looks at us in mirrors and easily takes advantage of that. Not sure if she recognizes herself in a mirror, but she seems to. That’s harder to tell, though. She certainly reacts to >other< cats when she sees them, but doesn't seem to care about her reflection.

  2. Shatterface says

    I’m pretty sure that dogs are capable of some degree of theory of mind because they can practice deception – and I think that requires ToM (apart from, say, camouflage, which is deceptive but not based on ToM).

    I used to have a spaniel and he wasn’t allowed to have brittle bones from lamb chops in case he choked.

    On one particular occasion he stood by the back door waiting to be let out apparently so he could ‘do his business’ – then as soon as the door was opened he ran back into the kitchen, stole a bone from the bin and ran outside with it.

    This suggests he was not only aware that he was not allowed to have the bone because we would not allow it but that he was also capable of tricking us into believing he had one intention while he had another in mind.

    That’s what, a fourth order of intentionality? He thinks that we think that he wants to go outside for a pooh when his intention is something else?

  3. Shatterface says

    One of the tests for Asperger’s is the ”Sally-Ann” or ”false belief” test:

    .n the test process, after introducing the dolls, the child is asked the control question of recalling their names (the Naming Question). A short skit is then enacted; Sally takes a marble and hides it in her basket. She then “leaves” the room and goes for a walk. While she is away, Anne takes the marble out of Sally’s basket and puts it in her own box. Sally is then reintroduced and the child is asked the key question, the Belief Question: “Where will Sally look for her marble?”

    For a participant to “pass” this test, they must answer the Belief Question correctly by indicating that Sally believes that the marble is in her own basket. This answer is continuous with Sally’s perspective, but not with the participant’s own. If the participant cannot take an alternative perspective, they will indicate that Sally has cause to believe, as the participant does, that the marble has moved. Passing the test is thus seen as the manifestation of a participant understanding that Sally has her own beliefs that may not correlate with reality; this is the core requirement of theory of mind.

    http://en.wikipedia.org/wiki/Sally–Anne_test

    I never had this test as I was diagnosed as an adult and in any case you can pass it applying logic rather than ToM.

    I think any test of whether an animal has ToM in anything like human terms would involve at least an understanding of ”false belief” – and very possibly the ability to produce ”false beliefs” by conscious deception.

  4. badgersdaughter says

    I don’t think he’s theorizing about your state of mind and what you think his intention is, Shatterface, I think he’s thinking something more like “get the human to stand in the door corner so I can steal the bone”. He knows what actions to perform to get you to stand over there, so that is what he did.

    My cat Ink does the same sort of thing, and I know she has less theory of mind than a dog does. She has a characteristic meow for when I am at the computer and she wants a drink out of the bathroom faucet. In past weeks she has started giving this meow, heading for the bathroom, cutting an abrupt turn right in front of the bathroom door, and heading to the spot where she gets kitty treats. This is not mind theory; it’s “I can get her out of her chair for something I want by doing this, so I’ll try it when I need to get her out of her chair for something else I want, too”. I do not give in, so eventually she will stop doing it for treats and will probably retain the behavior for the water faucet. Every now and then a cat will regress because it seems to associate the failure of the behavior to get the new result with the behavior being generally unwanted.

  5. Shatterface says

    Incidentally, I prefer Daniel Dennett’s term ”the intentional stance” to ”theory of mind” because, well, ToM is usually pre-theoretical. It makes a folk scientific principle sound scientific when it really isn’t so schematic. They’re rules of thumb rather than scientific rules.

    Also ”intentionality” can be extended into orders: first order, second order, third…

    http://en.wikipedia.org/wiki/Intentional_stance

  6. Shatterface says

    I have read a study on ”laughter” in rats but it turns out its more a somatic response to tickling rather than the schadenfreude of seeing a cat who’s trapped his head in a tin can because someone cut it’s whiskers off.

  7. John Morales says

    This of course causes not the faintest surprise in anyone who’s spent any time with dogs.

    Indeed.

    It is my experience that dogs can (and do) prevaricate — and even lie.

  8. John Morales says

    badgersdaughter,

    My cat Ink does the same sort of thing, and I know she has less theory of mind than a dog does.

    This is equivalent to stating that a dog has at least some theory of mind; which itself is tantamount to claiming a dog has theory of mind.

    </pedant>

  9. Shatterface says

    One reason that I’m wary of attributing ToM to cats is that their behaviour becomes pretty sadistic if they actually understand the suffering they inflict when ”toying” with their food.

  10. Shatterface says

    That they enjoy inflicting pain? Would you want a sadist in your house?

    Anyway, I’m a dog person rather than a cat person.

  11. John Morales says

    [meta + OT]

    Shatterface @11, I note you’ve evaded the import of the question with an appeal to emotion.

    (To assuage you, I note that I believe the infliction of pain per se is probably incidental to cats’ enjoyment; mine is happy to play with the corpses of the juvenile mice she occasionally catches on subsequent days, even if perhaps not quite so much so as when they react)

  12. Shatterface says

    Incidentally, children’s ability to pass the false belief tests seems to emerge around the same time they learn to use pronouns correctly, i.e. to distinguish between ”you” and ”me” – something autistic kids often have problems with. Pronoun reversal was one of Leon Kenner’s original diagnostic indicators of autism.

    Might be a coincidence but interesting nonetheless.

  13. Shatterface says

    Shatterface @11, I note you’ve evaded the import of the question with an appeal to emotion.

    I was trying to avoid implying moral judgements about people who own cats, rather than being scientific.

  14. John Morales says

    [OT + meta]

    FWIW, Shatterface, I want to make it clear that I pretty much endorse (and have enjoyed) your comments in this thread.

  15. iknklast says

    Cooper starts awake when I put my reading glasses down on the desk

    Our dog can tell the closing credits of a movie; he knows if we’re watching a movie, he gets his walk when it’s over. He can tell the difference between the closing credits and the other music of the movie.

    He also expects his walk anytime someone hangs up the phone. My husband often has a call from a friend in the evening, and the dog’s walk often follows.

  16. John Morales says

    iknklast:

    Our dog can tell the closing credits of a movie; he knows if we’re watching a movie, he gets his walk when it’s over. He can tell the difference between the closing credits and the other music of the movie.

    Hm. Beware the confounding factor of the clever Hans situation.

    (Might it be you that unconsciously signals that it’s nearly walkie-time?)

  17. Shatterface says

    FWIW, Shatterface, I want to make it clear that I pretty much endorse (and have enjoyed) your comments in this thread

    Thanks. I’m writing an essay on Philip K Dick’s Do Androids Dream of Electric Sheep and the film Blade Runner and the Voight-Kampff test used to identify the androids/replicants in both is based on measuring empathy – which the book identifies with ”role-taking” – so the issue is one I’ve been thinking a lot about recently.

  18. says

    I don’t know where this falls as a “theory of mind”, but…

    My cat is a cuddle cat. My cat is very clingy. He likes to hop up on laps uninvited and get pets. He loves having his belly rubbed. My wife and I love giving him attention, too, and we play with him more or less the same.

    My wife loves cats who vocalize in loud, yowly ways. Me, I don’t mind the noise unless they’re right there in my lap making it. The funny thing is that my cat will happily take the exact same play, including rough play, from me, and simply purr the whole way through. And for my wife he will happily yowl away (as well as purr — it’s a hoot to hear as long as I’m on the other side of the room). Everything else is the same. Yes, I get a bit of teeth and claw from him. it’s his belly I’m playing with…what should I expect? He never stops purring and no matter how rough he or I get, I guarantee he’ll want more the second I put him down. But he only ever yowls for me when I’m doing something he genuinely doesn’t like (like wrapping him in a sheet — he doesn’t like having his head covered). But for my wife he happily yowls in play.

    All I can think is happening is that he’s figured out that for me, yowls means I’m putting him down, and for my wife, yowls means more attention.

  19. Amy Clare says

    The problem with trying to imagine a theory of mind for the dog in this example (the feral cat story) is that we’re relying on our human ways of experiencing the world. Dogs experience the world differently, for example they rely on smell a great deal, and have pack instincts which come from a history of complex wolf societies. It’s possible that this particular dog could smell that the cat was pregnant (due to hormones and pheromones being secreted) and this triggered a social instinct to care for it, providing it with food. Perhaps it didn’t realise that they were different species and may even have been acting as though the kittens were its own family. There are quite a lot of examples of cross-species friendships and parenting behaviour.

    It’s just a theory, though – do dogs behave differently around pregnant humans? We tend to mask our natural scents a lot more so it would be difficult to tell I think.

    FWIW I think animals have a lot more going on in their brains and minds than we think they do, and ultimately it will be impossible to tell just how much as we can’t adequately communicate with them. But perhaps this example just needs us to try and think in a more dog-orientated way…

  20. says

    But do they have an elaborate enough theory of mind that they think about the feral cat outside who might be hungry and thus might want some of their food?

    I wouldn’t be surprised if they did — but this research doesn’t prove it.

  21. karmacat says

    What I have read in the past about cats and dogs is that dogs are pack animals while cats are “loners.” So a dog is likely to treat its human as one of the pack and probably the alpha dog. Cats relate to humans as if the human is its mother. Cats “tormenting” their prey is probably related to mother cats teaching their kittens how to hunt by giving them half dead prey. It is probably instinctual because cats who haven’t been taught to hunt will still play with a half-dead prey. Sadism is really about getting pleasure out of someone else’s pain. I don’t think cats see “prey” as having feelings, etc. By the way, some cats will purr with any feeling, i.e. when they are hungry, happy, even irritated. You can tell a cat is ready to attack when the ears go back and the tail starts swishing

  22. opposablethumbs says

    It would make evolutionary “sense” for any social animal (which cannot or does not normally live or reproduce without a group) to have something like a TOM or at any rate to develop tendencies to “look after” other members of the group (which could have the splash effect of inclining us/other animals to look after even members of other species).

    No need to add a layer of god-complexity (or other woo), of course, when the capacity for altruism is among the many things we and many other animals have hard-wired.

  23. Wylann says

    You can tell a cat is ready to attack when the ears go back and the tail starts swishing.

    A lot of people miss animal cues like that, that seem obvious to those of us who spend a lot of time around them. When my cat does (ears go back), I give her a stern, but not loud, “no”, and she immediately stops the threat display. She usually follows up with a head butt and wanting attention. :)

  24. opposablethumbs says

    Exactly, Wylann – cues are what it’s all about, isn’t it!

    Used to know a bloke who’s a bit of a dog-whisperer (except in a totally non-glamorous, non-TV, East End sort of way) who always said you have to tell the dog off before it’s done something wrong.

    After several years of thinking he was advocating something literally impossible without a time machine (but noting that he was remarkably good at getting good results with dogs, so deciding to pay attention anyway) … eventually, eventually it sank in that what he actually meant was – if you pay enough attention and are sufficiently observant, you’ll see the cues. So you can say “no” or make a loud noise with something before the dog actually jumps up/lunges at another dog/nicks the food/whatever. Made total sense. Couldn’t believe I was so slow on the uptake.

    (Every now and then, he’d get someone go to him with aggression problems, usually a dog with a macho owner who thought he needed to encourage it to be aggressive and then found he couldn’t get near the dog himself. He was very good at re-training the owners to drop the macho bullshit, at least as far as their dog was concerned. Had one guy turn up with a rottweiler called “Tyson” (yeah, no guesses) that he couldn’t control, and had my then 12-year-old daughter handling the dog within about 20 minutes. The owner was gobsmacked. Part of the programme for this dog included changing its name, to get rid of previous associations and also to make it something the owner couldn’t bellow aggressively; the bloke went off with his dog eventually … real tough-guy sort of bloke with a great big male dog now answering to the name of Daffodil.)

  25. A Masked Avenger says

    But do they have an elaborate enough theory of mind that they think about the feral cat outside who might be hungry and thus might want some of their food? Do they have an elaborate enough theory of mind that they then decide to leave some of their precious food behind for that possibly hungry cat outside who might come inside and eat it?

    Uh, no.

    Right. But this is interesting on a couple of levels.

    First, it’s interesting in light of fundies’ conviction that humans are separated from the rest of the animals by an uncrossable gulf. It’s evidence that our differences are in degree rather than kind.

    But it’s also interesting from the general viewpoint of evolution of consciousness, or inquiries into what the heck consciousness even is in the first place. One of the things that fatally undermined my own human essentialism was Dawkins’ suggestion that consciousness might simply be an emergent property of any theory of mind applied to itself. I.e., as soon as I become able to forecast my own mental state after X, Y, or Z hypothetical event, I’m now “self-aware,” and somewhere in the center of that self-referential spiral is my “self.” I dunno if that turns out to be right, but it’s sufficiently plausible to destroy the incredulity that I used to appeal to.

    It’s also interesting in conjunction with other research about dogs. Dogs, for example, can find objects that you point to. Cats, on the other hand, will sniff your finger and stare at you. The Russian domesticated foxes share with dogs an ability to read human facial expressions. It seems fairly clear that they have some sort of model that describes you and allows them to make useful predictions.

  26. Wylann says

    It’s also interesting in conjunction with other research about dogs. Dogs, for example, can find objects that you point to. Cats, on the other hand, will sniff your finger and stare at you.

    This is a bit of a generalization. I have known (and owned) dogs who never got the idea of what pointing was all about. The flip side of that is that some cats (my current cat, who is very people oriented) does have a tentative, but not 100%, grasp of what pointing means.

  27. AnotherAnonymouse says

    We used to joke that our dog could read emails; a typical afternoon for me consisted of coming home from work, quickly checking my email, then taking the dog for a walk. Clearly the dog picked up on whatever cues I was subconsciously giving off, because all I had to do was type, “I’m logging off to walk the dog” and he would run to the cupboard where we kept his leash. He also understood line-of-sight; when he wanted an additional walk, he’d stand halfway between his leash and the nearest human, catch the human’s eye, then stare at the leash. We didn’t teach this to him; he figured out on his own that humans understood line-of-sight.

  28. A Masked Avenger says

    This is a bit of a generalization. I have known (and owned) dogs who never got the idea of what pointing was all about. The flip side of that is that some cats (my current cat, who is very people oriented) does have a tentative, but not 100%, grasp of what pointing means.

    Sure. But I’m alluding to actual studies (which I apologize for not tracking down, but I don’t have time right now), which I learned about secondhand, in which dogs demonstrated an ability to retrieve objects that were pointed out, while cats could not. Although I don’t know the success rates or correlation coefficients, I’m sure it was less dramatic than “all dogs always can and no cat ever can.”

Leave a Reply

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

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>