Do Dogs Have a Theory of Mind?

    • Gold Top Dog

    Do Dogs Have a Theory of Mind?

    Dogs can read visual cues from us better than chimpanzees can. Does that mean they have a 1st-level Theory of Mind? 

    "Do Dogs Have a Theory of Mind? Yes, but Whose Mind Is It?"

    LCK

    BTW, Jupiter, this article is not specifically about telepathy, but you might be interested to know that when Rupert Sheldrake did his studies on "dogs that know when their owners are coming home," he set them up in such a way as to discount all the arguments you have against the dog being able to determine their arrival through sensory cues, or through past experiences.

    • Gold Top Dog

    I think that you have assumed far too much from a short home grown experiment.  There is ongoing research taking place at Harvard right now that is incomplete.  I must say that I saw some flaws in it, but am still going to wait to see what conclusions are reached, since my dog is one of the participants.

    • Gold Top Dog

    spiritdogs

    I think that you have assumed far too much from a short home grown experiment.  There is ongoing research taking place at Harvard right now that is incomplete.  I must say that I saw some flaws in it, but am still going to wait to see what conclusions are reached, since my dog is one of the participants.

     

    Hi, SpiritDogs,

    This is great news! Congratulations!

    The main purpose of my home-grown experiment was to see how dogs would react in situ, without any context related to the dog's hunting instincts (rather than in a clinical, i.e., unnatural setting where the experiment is set up as part of a hunting game). So far, not a single dog has shown any ability to look behind them. (One of my readers has posted a comment saying that her corgi can do this very reliably; though I haven't seen any video evidence of it.)

    I also disagree with the conclusion found in Hare, et al (2002)* that this behavior isn't inherited from the wolf's prey drive. That's true on one level: the behavior isn't inherited directly from the wolf, but it is directly related to the group coordination necessary for hunting large prey. That emotional flexibility necessary for working as a team, and the ability to pick up cues from one's packmates while hunting, form the bedrock for the dog's social and emotional flexibility as well as their ability to pick up cues from us.

    I also recall seeing something to the effect that the study shows that dogs are capable of understanding the mental representations of human beings (I think Brian Hare himself might have said this), which is one way of suggesting that they have a first level theory of mind.

    At any rate, I can't wait to hear more about your dog's studies!

    LCK

    * "these results do not support the predictions of ... the canid generalization hypothesis (dogs have inherited their skills from wolves)"

     

    • Gold Top Dog

    *REMOVED BY MODERATOR - Board Wars are not allowed.... and removing Moderator warnings/edits is not allowed, either.*

    ToM and many other things that LCK denies are all affirmed by Colin Allen, Center for the Integrative Study of Animal Behavior, in a podcast http://www.philosophytalk.org/pastShows/AnimalMinds.html

    When my dog whines or brushes my hand to wake me at night to let her out for an unscheduled poop, she is a) communicating, b) exercising conscious intentionality, and c) waking up another's mind to get what she wants = ToM. Somehow, LCK denies all such abilities for dogs.

    I would hate to be the kind of dog that Lee thinks all dogs are like - thoughtless slaves to uncontrolled emotional charges. How could such a tormented creature even try to navigate their environment.

    Enjoy the podcast, Lee, and get some good knowledge on.

    • Gold Top Dog

    It may be of interest to recall that Ted Kerasote's dog, Merle, in the book _Merle's Door_ did not recognize himself in a mirror _until_ Ted stood next to Merle, and then Merle recognized himself as the companion of Ted, Thereafter, Merle looked at the mirror to see himself w/ or w/o Ted around.

    Regardless, recent thinking has it that the mirror test is probably not very useful in determining a ToM. I would think research methods of infant psychologists like Piaget would be useful in studying dog cognition.

    • Gold Top Dog

    I have had my own experience with dogs and mirrors.  Two of my dogs seem not to recognize themselves, but they also do not regard the image as a threat, so perhaps they have at least learned that unless the image is three dimensional it need not concern them.  Sequoyah, on the other hand, barked viciously at her own image in a mirror (at age ten weeks), but stopped as soon as she began, I think, to notice that the image was moving whenever she moved.  Now, that may not be self-recognition, but it's certainly something that told her it was not a threat.  She has never barked at a mirror ever again and she is now five years old.  Oddly, she is also the dog who has most easily learned to mimic me:-)  I find this all fascinating, and I hypothesize that while we may have greater executive function than dogs, because of our larger pre-frontal cortices, they almost certainly have some function that is beyond the capacities that LCK has assigned to them.

    • Gold Top Dog
    spiritdogs

    I have had my own experience with dogs and mirrors.  Two of my dogs seem not to recognize themselves, but they also do not regard the image as a threat, so perhaps they have at least learned that unless the image is three dimensional it need not concern them.  Sequoyah, on the other hand, barked viciously at her own image in a mirror (at age ten weeks), but stopped as soon as she began, I think, to notice that the image was moving whenever she moved.  Now, that may not be self-recognition, but it's certainly something that told her it was not a threat.  She has never barked at a mirror ever again and she is now five years old.  Oddly, she is also the dog who has most easily learned to mimic me:-)  I find this all fascinating, and I hypothesize that while we may have greater executive function than dogs, because of our larger pre-frontal cortices, they almost certainly have some function that is beyond the capacities that LCK has assigned to them.

    As I recall, one or two of ours did the same thing you report about a mirror. We humans need to see ourselves (egocentrism) to know ourselves. Non-human animals live in an eco-sphere, not an ego-sphere, so they want to feel themselves at home - connected. The mirror image was not a threat, it was even perhaps a comfort of some sort. Merle reportedly enjoyrd seeing himself.

    I agree completely w/ your comment on our higher executive function is not negating any such function for dogs. On this, Kerasote makes the point that as a stray, Merle had to use intelligence to survive, and that this made hime a better than average companion.

    • Gold Top Dog

    spiritdogs
    I hypothesize that while we may have greater executive function than dogs, because of our larger pre-frontal cortices, they almost certainly have some function that is beyond the capacities that LCK has assigned to them.

     

    Hi, SpiritDogs,

    This is a fascinating discussion. I've also had dogs who barked at their mirror image when they were younger, but lost all interest over time.

    The whole phenomenon of how dogs do or don't interact with mirror images may relate, in part, to mirror neurons, which are quite probably part of a dog's cognitive palette. There was a recent article by Greg HIckock (Professor of Cognitive Sciences, and Director of the Center for Cognitive Neuroscience at UC Irvine), in which he articulates a very compelling position that mirror neurons are not about action "understanding" but action "selection." There's a huge difference in terms of cognition.

    It's also important to note that when scientists perform the mirror test on animals, they do so by "surreptitiously marking the animal with two odourless dye spots. The test spot is on a part of the animal that would be visible in front of a mirror, while the control spot is in an accessible but hidden part of the animal's body. Scientists observe that the animal reacts in a manner consistent with it being aware that the test dye is located on its own body while ignoring the control dye." (From Wikipedia: see also dog barking at mirror video.)

    I agree with you that dogs have a precursor to the kind of executive function found in the human brain. In humans this is variously called "self control," "delayed gratification," "impulse control," etc. I go into this in more detail in my articles on Freud, which reference studies on how exerting impulse control, i.e., suppressing some of the energy coming from the Id, depletes energy in the prefrontal cortex, which is where Freud located the Ego, or conscious mind. (By the way Freud coined the term Ego primarily because the so-called "conscious" mind isn't necessarily all that "conscious";).

    I think the genesis of this ability in dogs comes from the way wolves are able to sublimate their individual needs for the needs of the group as a whole (which goes back to a kind of group consciousness, or a group desire).

    Also, I'd like to point out that I'm not "assigning" dogs any cognitive function of any kind, nor am I restricting them in any way. Dogs have whatever cognitive functions have been "assigned" to them by Nature. I'm simply observing and interpreting their behaviors from what I think is a different, more parsimonious perspective than what most mainstream scientists are doing; I try to confine my explanations of behavior from the perspective of energy, emotion, and a remarkable ability for group consciousness. At this point in time, most mainstream scientists see dogs as "thinking" (which violates the principles of both evolution and neuroscience), or mechanical organisms (which is a very old-fashioned, Cartesian view). I believe that by not framing their questions about dogs around the principles of emergence theory (just as an example), these scientists are missing what amazing and wonderful creatures dogs truly are.

    I'm been discussing these issues openly for over ten years and no one has yet given me an example of canine behavior that can't be explained more parsimoniously, i.e., through Ockham's razor, Morgan's canon, and the law against anthropomorphism. There have been times when I felt stumped by a certain scenario. But once I got away from the computer, and spent a little time hanging out with dogs, the answer always rang clear as a bell.

    LCK

    • Gold Top Dog

     

     

    Lee Charles Kelley
    I'm been discussing these issues openly for over ten years and no one has yet given me an example of canine behavior that can't be explained more parsimoniously


    Burl
    When my dog whines or brushes my hand to wake me at night to let her out for an unscheduled poop, she is a) communicating, b) exercising conscious intentionality, and c) waking up another's mind to get what she wants = ToM.  

     

    • Gold Top Dog

    Nice try. But we both know I've already explained this at least twice.

    LCK

    • Gold Top Dog
    Here is what you said:

    Burl said: When one of the mutts needs to go outside and we are at the computer or asleep, they signal us to alertness with a yelp or a nudge. This is also an example of conscious intention and communication.

    Lee said: To me this is clearly an example of a behavior performed by a dog to try and get its owner to do something, not performed with the conscious intent to report information.

    Do you not see the difference yet?

     

    Do you not see there is no difference... yet?

    Lee, you have not taken even a small step towards making any meaningful point here. This is a simple case of conscious reasoned action and intent to communicate with another mind. What would you call it when a child does the same type thing to let a parent know he/she has to go? Conscious reasoned action and intent to communicate with another mind, maybe?

    • Gold Top Dog
    Newborn babies have a ToM http://www.philosophytalk.org/pastShows/MindsofBabies.html I wonder who a puppy thinks will come pick her up when she cries?

    In latest Charlie Rose episode, an animal writer says a human gets the ToM at age 3 and also that dogs have a ToM. This would mean about a year old dog is more socially advanced than a 2-3 year old baby. I can see that, I guess.

    • Gold Top Dog

    Lee Charles Kelley

    spiritdogs
    I hypothesize that while we may have greater executive function than dogs, because of our larger pre-frontal cortices, they almost certainly have some function that is beyond the capacities that LCK has assigned to them.

     

    Hi, SpiritDogs,

    This is a fascinating discussion. I've also had dogs who barked at their mirror image when they were younger, but lost all interest over time.

    The whole phenomenon of how dogs do or don't interact with mirror images may relate, in part, to mirror neurons, which are quite probably part of a dog's cognitive palette. There was a recent article by Greg HIckock (Professor of Cognitive Sciences, and Director of the Center for Cognitive Neuroscience at UC Irvine), in which he articulates a very compelling position that mirror neurons are not about action "understanding" but action "selection." There's a huge difference in terms of cognition.

    It's also important to note that when scientists perform the mirror test on animals, they do so by "surreptitiously marking the animal with two odourless dye spots. The test spot is on a part of the animal that would be visible in front of a mirror, while the control spot is in an accessible but hidden part of the animal's body. Scientists observe that the animal reacts in a manner consistent with it being aware that the test dye is located on its own body while ignoring the control dye." (From Wikipedia: see also dog barking at mirror video.)

    I agree with you that dogs have a precursor to the kind of executive function found in the human brain. In humans this is variously called "self control," "delayed gratification," "impulse control," etc. I go into this in more detail in my articles on Freud, which reference studies on how exerting impulse control, i.e., suppressing some of the energy coming from the Id, depletes energy in the prefrontal cortex, which is where Freud located the Ego, or conscious mind. (By the way Freud coined the term Ego primarily because the so-called "conscious" mind isn't necessarily all that "conscious";).

    I think the genesis of this ability in dogs comes from the way wolves are able to sublimate their individual needs for the needs of the group as a whole (which goes back to a kind of group consciousness, or a group desire).

    Also, I'd like to point out that I'm not "assigning" dogs any cognitive function of any kind, nor am I restricting them in any way. Dogs have whatever cognitive functions have been "assigned" to them by Nature. I'm simply observing and interpreting their behaviors from what I think is a different, more parsimonious perspective than what most mainstream scientists are doing; I try to confine my explanations of behavior from the perspective of energy, emotion, and a remarkable ability for group consciousness. At this point in time, most mainstream scientists see dogs as "thinking" (which violates the principles of both evolution and neuroscience), or mechanical organisms (which is a very old-fashioned, Cartesian view). I believe that by not framing their questions about dogs around the principles of emergence theory (just as an example), these scientists are missing what amazing and wonderful creatures dogs truly are.

    I'm been discussing these issues openly for over ten years and no one has yet given me an example of canine behavior that can't be explained more parsimoniously, i.e., through Ockham's razor, Morgan's canon, and the law against anthropomorphism. There have been times when I felt stumped by a certain scenario. But once I got away from the computer, and spent a little time hanging out with dogs, the answer always rang clear as a bell.

    LCK

     

    I believe that the reason some mainstream scientists argue that dogs think is because they have designed some experiments that show that, not because they are deliberately trying to find reasons to support the hypothesis that dogs don't think.  Years ago, it was argued that dogs don't feel pain.  I still think the subject is ripe for research, and I don't close the book on any of it, as you seem to be doing.

    Anthropomorphism?  Interesting take on that subject: http://www.barnard.edu/psych/faculty/Horowitz/pdfs/Horowitz_Bekoff.pdf

    I don't know if you were trying to imply that those of us who don't necessarily share all your views are somehow NOT hanging out with dogs, and spending all our time on our computers, but I assure you that is not the case.  Some of us hang out quite regularly with them, and with more than just our own dogs.

    I don't necessarily think that scientists are missing the wonders of the canine species by trying to evaluate their cognitive processes, but I do think that some research may be flawed by their lack of knowledge of dogs - in the recent experiment that Sioux participated in, her reactions were muddied by the fact that the "ready" signal they used was a bell that hurt her ears.  Once the bell was gone (it was not part of the experiment, and could easily have been supplanted with something quieter) she was still leery of the research assistant.  So, she made a quick association that he was a problem, and was not as able to focus on the exercise at hand.  She eventually got over it, but not until some of the responses had been affected.  So, unless you can control for those things, maybe your research isn't accurate...

    I think that parsimony is not always the best course - we need to be open to complexity as well;-)   Also, define "energy" - so often loosely or unscientifically defined, yet offered as a reason for all sorts of things.

    • Gold Top Dog
    spiritdogs

    I believe that the reason some mainstream scientists argue that dogs think is because they have designed some experiments that show that, not because they are deliberately trying to find reasons to support the hypothesis that dogs don't think.  Years ago, it was argued that dogs don't feel pain.  I still think the subject is ripe for research, and I don't close the book on any of it, as you seem to be doing.

    Anthropomorphism?  Interesting take on that subject: http://www.barnard.edu/psych/faculty/Horowitz/pdfs/Horowitz_Bekoff.pdf

    I don't know if you were trying to imply that those of us who don't necessarily share all your views are somehow NOT hanging out with dogs, and spending all our time on our computers, but I assure you that is not the case.  Some of us hang out quite regularly with them, and with more than just our own dogs.

    I don't necessarily think that scientists are missing the wonders of the canine species by trying to evaluate their cognitive processes, but I do think that some research may be flawed by their lack of knowledge of dogs - in the recent experiment that Sioux participated in, her reactions were muddied by the fact that the "ready" signal they used was a bell that hurt her ears.

     

    Hi, SpiritDogs,

    Thanks for the response. It's cool to hear about Sioux' experiences, teaching those researchers a lesson in dognitive science! (It's also interesting about the bell, since there's a new study ostensibly showing that dogs have a Theory of Mind, which involves putting bells on containers of food to see if dogs will steal food from the quieter of two containers.)

    Maybe I have closed the book, as you put it, on how dogs think. I like to think I'm open to any possibility as long as it makes sense and passes the "smell test." As I said, I've been discussing these issues for many years, and every so often someone will come up with a behavior that has me stumped*; one whose only explanation would be some form of rational thought. But it usually only has me stumped (as I said) while I'm at the computer. When I'm out with dogs, I can see things more clearly, i.e. from their point of view. This is my own thought process, and is not meant to imply anything about how anyone else approaches their ruminations on the topic. *(A behavior that has me pondering, if not stumped, is the reports I'm getting that some dogs react to two-dimensional images of other dogs, either on TV, or in paintings or posters.)

    As for whether some scientists are deliberately ignoring simpler, more parsimonious explanations for certain behaviors, that became very clear to me when I investigated the idea of "inequity aversion."

    This concept, which is probably quite true in humans, was initially brought forth by two experimental economists. Their paper on the subject was not based on their own research. They culled their data from various sources, set up a rigorous list of criteria on how the data should be interpreted, which they ignored if doing so gave them the results they wanted.

    Another economist wrote, "It seems that, at least in part of the field of experimental economics, the scientific standards may not be as rigorous as one may desire. ... the sad point is not only that two distinguished economists allow themselves to write in this style, it is also ... that these papers have caused no outcry or protest in the experimentalist community. Could it be that ... no one has ever carefully read the details of the paper until now?" "The Rhetoric of Inequity Aversion,"

    Then Frans de Waal and his colleague Sarah Brosnan took this idea and applied it to captive capuchin monkeys, and then to chimpanzees. In each case, they, too, got the results they were looking for. (de Waal admitted that his results probably couldn't be applied to wild monkeys.) Later, in a New York Times online interview he said, "I expect the same inequity aversion [to be found] in dogs and wolves..."

    So Fredericke Range, at the Clever Dogs Lab in Austria, following de Waal's expectation, armed with some pre-determined ideas of her own, set up a very badly designed study (i.e., one with heavy confirmation bias built into it) to prove that this tendency did, in fact, exist in dogs. Her conclusions that dogs "know when they're being treated unfairly" was based almost solely on her interpretation of the look on a dog's face. (Here's my take on that study: "Tuning In to Your Dog's Emotions.";)

    Meanwhile, another group of researchers tested this idea with chimpanzees and came to a far different conclusion: "These results are exactly the opposite to that of Brosnan et al. (2005)."

    On Charlie Rose the other night, Jeffrey Kluger, a science editor at Time Magazine (formerly an attorney, not a scientist) was on to discuss the cover story of tomorrow's issue. He gave a litany of cognitive abilities that "science is now telling us" animals have. And each and every one of them has been questioned, disproven, or debunked by other scientists. One example is the idea that scrub jays will re-cache food only when rival jays "aren't looking." This caused the initial researchers to conclude that these jays must have a Theory of Mind. But in Penn & Povinelli 2007b; Povinelli et al 2000; and Povinelli & Vonk, 2003; 2004, lucid and alternative explanations were given. (See "Darwin's Mistake," Penn, Holyoak, and Povinelli, p. 12)

    On Charlie Rose, Kluger offered video-taped evidence that Kanzi the bonobo can understand human language. And it's pretty compelling stuff, up until the last few seconds.

    In the video (which I found, on a certain level, to be very sad), the researcher -- who is oddly wearing a welder's mask (ostensibly to prevent Kanzi from reading her eyes) -- asks Kanzi to do a series of simple tasks involving two nouns, each for a different object, and a verb, connecting them.

    "Kanzi, could you cut the onions with your knife?"

    Kanzi seems to think about it, looks around, locates the knife, and mimes cutting the onions.

    "Could you put some soap on your ball?"

    Kanzi finds the dish soap and squeezes some onto a ball.

    This goes on and on. But look very closely at what happens at the end, starting at about 1:30.

    The researcher says, "Kanzi, pour the Perrier water ..."

    Kanzi reaches for a jar of jelly.

    "... into the jelly."

    Wait! What? Kanzi's response is out of sequence with the researcher's words! This means that either this particular behavior (and perhaps each one of them) was rehearsed over and over -- which removes some of the mystery of how he supposedly "understands" the meanings of words (they're not words to him so much as they're audio cues) -- or else Kanzi was picking up a mental image of the jelly jar from the researcher before she actually spoke the word "jelly." (Another possibility is that it's just a coincidence, though that seems unlikely.)

    Why did Charlie Rose and his animal cognition expert miss this? It's right there on the video, clear as day. And why does the science media, such as Time Magazine, always report the amazing news about scrub jays, or that dogs know when they're being treated unfairly, and you never hear any info about the rebuttals coming from serious cognitive scientists?

    This is a long-winded reply to your statement that you don't think that "scientists are missing the wonders of the canine species by trying to evaluate their cognitive processes." That statement is partially true; any serious attempt at evaluating a dog's cognitive processes is laudable, and affords us a marvelous window into the minds of these amazing animals. What isn't true (at least in most cases) is that these scientists (or most of them) are actually trying to evaluate the cognitive processes of dogs when they're not. In fact, they seem hell bent on setting up their experiments and interpreting their data in a way that automatically guarantees the results they're looking for.

    Yes, science should be open to possibilities, but it should also be parsimonious, skeptical, and -- well -- scientific. From my observations, that's not happening often enough.

    LCK

    PS: Marc Bekoff wrote an article for PsychologyToday.com, lauding anthropomorphism (a kind of precis of the paper you cited). For those interested, here's my reply: "How Dogs Think: The Debate Between Emotion and Logic." As I write in the article, "When people tell me dogs have the ability to reason I say, 'Hold on, let's understand their emotions first before we start giving them intellectual faculties.'"

    It's true that to some people, I seem to be throwing a damp rag onto the fire. But, hey, somebody's got to do it!

    Seriously, though, which do you think is more scientific: jumping straight to an interpretation based on higher cognitive function (which will also bring you media attention, more grant money, and possible book deals), or looking for an explanation, a la Morgan's canon, which "stands lower in the scale of psychological evolution and development." I think too many dognitive scientists are ignoring the laws of parsimony, Morgan's canon, and are acting almost as if they feel compelled to anthropomorphize animals. (Marc Bekoff, God bless him, certainly does.) Their interpretations may prove to be justified in the long run, but at present, I think it's best to be more scientific, not jump to any conclusions, and look at all sides of the issue.

    That's all I'm saying.

    LCK

    • Gold Top Dog

    I've seen experiments that show that dogs have a theory of mind. You close your eyes and they will do something you would normally caution against because they know you can't see with your eyes closed. They will also follow your eyes for a cue. This was proven in a double blind. My friend, Lee, put it best. Point your finger and a cat will look at your finger. Point your finger again and a dog will look to where you are pointing. That is, they understand that you have data and a vantage point different than their own that they can gain from.

     People that can't accept that need to examine their own emotional reasons for need to believe that dogs do not have a theory of mind. That's something that is often missing from this debate. For those who simply will not accept observable evidence and argue against it might have an emotional, religious, or psychological reason for needing it to not be true.

    As Freud might have said, "Sometimes, a banana is just a banana."