Theory of Mind

 My friend has just completed her Master's looking at this with dogs and obtained quite positive results (however, it isnt published yet

Generally research on Theory of Mind looks at the ability of nonhuman animals to attribute knowledge and perspective to other individual

One of the common type of experiments is a knower-guesser paradigm- the animal watches someone hide some food under one of several cups, while there are 2 other people present - one of these watches the food get hidden (the knower), while the other one doesn't (either they leave the room, or look away, or cover the eyes, etc.- this is the guesser). Then, the knower and the guesser point at the cup they think the food is in- if the animal has a theory of mind, then they should reliably follow the knower's pointing and ignore the guesser, because they would realise that the knower has seen the location of the food (and therefore knows where it is and is pointing to it), whereas the guesser hasn't, and is consequently pointing at the wrong cup.

The major problem with this kind of research is that there are lots of other cues (are the eyes uncovered? is the person present or absent?)  the animal could be using to determine which is the correct response, and often experiments don't control for them, or when they do, performance is at about chance levels.

I do believe it is possible for social animals to co-exist without a theory of mind- interactions could be based merely on the behavioural cues, rather than beliefs about another individuals' thoughts, emotion or knowledge.

 Ravens and possibly Pinon jays have a theory of mind, as well. With ravens the experiment was one in which ravens were observed caching food in the presence of another raven. The raven with the food would often cache the food behind something that prevented the other raven from seeing where it was caching the food. Interestingly, they also watched some ravens doing that kind of thing around wolves. If I remember correctly, the wolves were not cottoning on and paid no attention to the ravens unless the raven got too close to them, even when the ravens watched where the wolves were caching food and went in after to steal the food. Wolves never made an attempt to hide caching behaviour from the ravens, but the ravens made attempts to raid wolf caches and hide their own caches from other ravens. I'm not sure if the ravens hid caching behaviour from wolves, though.

I thought it was interesting, though, because the wolves apparently saw ravens raiding their caches sometimes and attacked the ravens when they saw a raid, but still never tried to hide caching from the ravens, whereas the ravens' caches were being raided by other ravens, and so the ravens started hiding their caching.
 

 That is definitely interesting! You can certainly see how having a theory of mind would be an evolutionary advantage, particularly for prey animals. Like the stereotypical cartoon of the large animal/person/whatever hiding behind the teeeeeeny tiny skinny tree - not a very safe hiding place! If the animal were capable of thinking, "Hmm, even if I can't see the lion, if I hide with just my head under this rock, then surely the lion will still be able to see me" then they would be much more likely to keep themselves out of harm's way (or jaws).

Vinia

One of the common type of experiments is a knower-guesser paradigm- the animal watches someone hide some food under one of several cups, while there are 2 other people present - one of these watches the food get hidden (the knower), while the other one doesn't (either they leave the room, or look away, or cover the eyes, etc.- this is the guesser). Then, the knower and the guesser point at the cup they think the food is in- if the animal has a theory of mind, then they should reliably follow the knower's pointing and ignore the guesser, because they would realise that the knower has seen the location of the food (and therefore knows where it is and is pointing to it), whereas the guesser hasn't, and is consequently pointing at the wrong cup.

The major problem with this kind of research is that there are lots of other cues (are the eyes uncovered? is the person present or absent?)  the animal could be using to determine which is the correct response, and often experiments don't control for them, or when they do, performance is at about chance levels.

Is the ability to follower the knower's cue learned?  That is, do they have a few practice runs so they learn to follow the knower but not guesser?  In terms of the other cues, what do you mean by "are the eyes uncovered"?  Are you saying that dogs are more less likely to follow the cues of the guesser if the guesser cover their eyes with their hands or a cloth than if they just shut their eyes?

The other question I would have is whether using humans to be the guesser and knower is the best test (and yes, I know I said that the information I read stated that dogs had a "Theory of Mind" as far as humans were concerned), but if dogs do have a "Theory of Mind," then I think they would understand that humans communicate with each other in ways dogs do not understand just as dogs can communicate with each other in ways humans do not understand.  Or in other words that dogs know that humans know things they do not understand (i.e. driving the car).  Therefore, the dog may assume (can dogs assume?, please just follow my attempt at logic) that the human may have knowledge of where the treat is that the dog do not have access to.  For example, that that the location of the treat is written somewhere.  Maybe if the knower and the guesser were dogs?  It seems that if a dog did have a "Theory of Mind" these issues might skew the test to make it seem like they did not.

Vinia

One of the common type of experiments is a knower-guesser paradigm- the animal watches someone hide some food under one of several cups, while there are 2 other people present - one of these watches the food get hidden (the knower), while the other one doesn't (either they leave the room, or look away, or cover the eyes, etc.- this is the guesser). Then, the knower and the guesser point at the cup they think the food is in- if the animal has a theory of mind, then they should reliably follow the knower's pointing and ignore the guesser, because they would realise that the knower has seen the location of the food (and therefore knows where it is and is pointing to it), whereas the guesser hasn't, and is consequently pointing at the wrong cup.

This may have been what the scientists doing these studies have referred to as, fast mapping. Associations resulting in behavior and learning to differentiate.

I am amazed how some of these scientists seem to get baffled by what they find out. At least for me, this is what it feels like they project onto the television shows. Many people have known that animals other than humans have minds. Dogs certainly have feelings which are an attribute to mind. As humans, we feel that dogs feel and we can "see" that they can feel and utilize their minds. Maybe some of these scientists can hone into their own department of feelings and use this feature of mind to complement the sheer logic compartment of mind that drives their research.

 This does not mean to say that there isn't good information to be discovered and raved about. For example, in that show Dog Genius, I did not know that left/right wag of a dog's tail is mapped into right/left brain hemispheres. That was an interesting finding. But to find out that dogs can know when we close our eyes, they can get away with sneaking the food behind the masters back. Heck, some of us have known that for ages - LOL. It is almost as if so many humans think that we're so special, that features of mind have only applied to us. How egocentric is that?

Btw, Good topic GoldenAC

GoldenAC

Is the ability to follower the knower's cue learned?  That is, do they have a few practice runs so they learn to follow the knower but not guesser?  In terms of the other cues, what do you mean by "are the eyes uncovered"?  Are you saying that dogs are more less likely to follow the cues of the guesser if the guesser cover their eyes with their hands or a cloth than if they just shut their eyes?

I would think that the researchers would have attempted to prevent the possibility of learning the knower's cue over trials, I'm pretty sure they didn't have practice runs, and only single trial testing ( only one turn per experimental condition), though I am not completely sure of this. There was one condition (and also in an experiment with chimps) where the knower put their hands on their cheeks, while the guesser covered their eyes- I can't quite remember what the results were though LOL.

I'll do some searching and see if I can locate any articles on ToM- then we can discuss an experiment properly.
 

GoldenAC

The other question I would have is whether using humans to be the guesser and knower is the best test (and yes, I know I said that the information I read stated that dogs had a "Theory of Mind" as far as humans were concerned), but if dogs do have a "Theory of Mind," then I think they would understand that humans communicate with each other in ways dogs do not understand just as dogs can communicate with each other in ways humans do not understand.  Or in other words that dogs know that humans know things they do not understand (i.e. driving the car).  Therefore, the dog may assume (can dogs assume?, please just follow my attempt at logic) that the human may have knowledge of where the treat is that the dog do not have access to.  For example, that that the location of the treat is written somewhere.  Maybe if the knower and the guesser were dogs?  It seems that if a dog did have a "Theory of Mind" these issues might skew the test to make it seem like they did not.

 Very good points!  I am not sure whether they have done dog-dog ToM testing- from my impression, it seems that not very many scientists are willing to venture into this area because of the difficulty drawing solid conclusions from the data. 
 

Okay, I am going to play devil's advocate here...

tssst

Many people have known that animals other than humans have minds. Dogs certainly have feelings which are an attribute to mind. As humans, we feel that dogs feel and we can "see" that they can feel and utilize their minds. Maybe some of these scientists can hone into their own department of feelings and use this feature of mind to complement the sheer logic compartment of mind that drives their research.

How can someone  know that other animals have minds? How do you know that I, or any other person, have a mind?  For that reason, what is a mind? A brain? Feelings? Neuronal activation levels

Some scientists believe that assuming other animals have minds like ours is pure anthropomorphism and unhelpful for research- how can we appreciate the true mental capabilities of nonhuman animals for what they are if we're too busy comparing them to ourselves?

tssst

Dogs certainly have feelings which are an attribute to mind

 Actually, feelings can be explained by a purely biological response (this was theorised very early on- in the 1800s, by Henry James). E.g. you see a huge bear and because you run, and your heartrate increases, and you break into a cold sweat, you feel afraid - rather than because you feel afraid, your heart rate increases, you break into a cold sweat, and you run.

Emotional responses do not necessarily reflect the ownership of a mind, whatever that may be...
 

Vinia

 Actually, feelings can be explained by a purely biological response (this was theorised very early on- in the 1800s, by Henry James). E.g. you see a huge bear and because you run, and your heartrate increases, and you break into a cold sweat, you feel afraid - rather than because you feel afraid, your heart rate increases, you break into a cold sweat, and you run.

Was that Henry James or his brother William James (of "The Will to Believe" and Pragmatism fame).  As a pure tangent, William James wrote in one of his books (can't remember which one right now) that dog could only think about something when it was in their sight, that if a dog saw a ball, it could think about "know" about the ball, but as soon as the ball was out of its sight it was like the ball did not exist.  I think we all know this is not true.  If a dog likes a ball and it disappears under the couch the dog will look for it endlessly.  

tssst

I am amazed how some of these scientists seem to get baffled by what they find out. At least for me, this is what it feels like they project onto the television shows. Many people have known that animals other than humans have minds. Dogs certainly have feelings which are an attribute to mind. As humans, we feel that dogs feel and we can "see" that they can feel and utilize their minds. Maybe some of these scientists can hone into their own department of feelings and use this feature of mind to complement the sheer logic compartment of mind that drives their research.

"Theory of Mind" is actually looking at what is considered a more complex ability of understanding than just having a mind.  It is whether an animal understands that other animals have their own minds and that their actions are controlled by their own emotions, desires and needs.  I believe in humans a child learns a "Theory of Mind" sometime around 3 years old.  But it is something that needs to be taught and then it needs to be extended to other animals, i.e., "do not pull the dog's hair because it will hurt the dog, you don't like it when someone pulls your hair."

But, I do agree with you, the idea that anyone can doubt that animals do not have minds or emotions is crazy.  I was at Amazon.com the other day  reading reviews of Marc Bekoff's "The Emotional Life of Animals" and it appears his central argument for believing that animals have emotions works well, I think, for this question and it is (from the review) 'if we don't know for sure, the default assumption should be that animals do have emotions until proven otherwise. To paraphrase: If I assume animals feel pain and pleasure and love, and act accordingly, and then it turns out my assumption was wrong, I will have done no harm. However, if I assume that animals don't have feelings, and then it turns out I was wrong, I may have caused immeasurable damage"

I do think that, more and more, science is coming around to this viewpoint. 

Hi there Vinia,

I have heard people often say that the brain is like the hardware and the mind is like the software in action.

The mind is so very very complex in that it involves not only billions of interwoven neural networks but these neural networks are adaptive through our sensoral inputs, neural network correlation, and a whole slew of other things we barely have a handle on, if at all.

So how do we know that a dog has a mind? Going with that question, how do we know humans have a mind... or how do we think we know that thay don't have one?

I was not comparing a dogs mind with a humans mind. They are just plain different (going with the assumption here that a dog has a mind) and the level of complexity is vastly different as well. We've got a much larger neocortex for one thing... but that's getting technical and off topic. The question was how that I iknow a dog has a mind. I can certainly say because I feel like they do and be done with it, but there is more. I can see it work with the way my own dogs solve problems presented to them, by the way they engage in play. can I prove it scientifically? Nope.

 What do you think it takes to claim that a being possesses a mind? The ability ro reason? the ability to create? The ability to rationalize? The ability to bond socially? The ability to know where it should go potty?

Cita

You can certainly see how having a theory of mind would be an evolutionary advantage, particularly for prey animals.

Actually, it wouldn't be an evolutionary advantage.Too much thought gets in the way of taking evasive action.

 "In all likelihood, the human brain evolved with a strong bias towards pattern recognition rather than deductive reasoning. This natural environment is comprised of two systems: the physical and the biological. Many aspects of the physical world can be usefully described by deductive axiomatic systems, and an information-processing system operating solely in a law- governed world would be able to survive with purely deductive reasoning; examples would include computer viruses and programmed trading systems.

"The biological world, in contrast, is exceedingly complex and arbitrary. It is a world of individuals constructed from complex feature vectors made of DNA, with billions of components, and selected solely by the ability of their ancestors to reproduce, oftentimes in unusual circumstances such as the aftermath of asteroid collisions. Such a world cannot be described deductively in any practical sense, but because it is very repetitive, pattern recognition is an effective information-processing strategy. If one Tyrannosaurus Rex tries to devour you, the next one is likely to as well. Since critical decisions must be made in real time ("Is the object approaching me sometime I can eat, something that will eat me, or something I can ignore?";), evolution will select for high recall speeds under noisy environmental conditions. It does not select for theorem proving or the minimization of quartic polynomials.

"This neural bias would emerge early in the biological record, well before the development of primates, or mammals, or even vertebrates. Homo sapiens is endowed with sophisticated pattern recognition capabilities honed through eons of evolution, and it is unsurprising that this capacity is put to use in social behavior. Deductive reasoning, in contrast, is a comparatively recent development and is much more difficult. While we are very proud of deductive reasoning, it is not necessarily more useful, particularly when dealing with social behaviors which may also have some evolutionary roots."

"A New Kind of Social Science,"  By Valerie M. Hudson, Brigham Young University, Philip A. Schrodt, University of Kansas, Ray D. Whitmer, Jhax.Ltd. Version 0.8B2, Last Update: 26 March 2004 Paper prepared for delivery at the Annual Meeting of the International Studies Association,
Montreal, Quebec, Canada, March 2004.
 
Granted, a theory of mind and deductive reasoning are not synonymous, but I think the principle holds true that pattern recognition is more advantageous for survival purposes (evading predators) than thinking about what the other animal is thinking or experiencing.

Besides, it's doubtful that most animals would even be capable of knowing what they themselves are thinking, feeling, or experiencing. They can only experience it, not put it into words...

Anyway, that's how I see it,

LCK

 

LCK 

Vinia

I do believe it is possible for social animals to co-exist without a theory of mind- interactions could be based merely on the behavioural cues, rather than beliefs about another individuals' thoughts, emotion or knowledge.

I agree in part. I don't think behavioral cues are the whole picture with dogs, though. My sense of things, based on my studies and observations, is that dogs can feel the emotional states of other dogs and humans, especially those they know and love, as if they were there own emotions. I think this ability comes from wolves, and how they have to be in tune with one another emotionally when they hunt as a pack.

Anyway, that's how I see it,

LCK 

tssst

But to find out that dogs can know when we close our eyes, they can get away with sneaking the food behind the masters back. Heck, some of us have known that for ages

Hey, Tssst,

Actually, that's probably a learned behavior, not that the dog "knows" we can't see him if our eyes are closed. For instance, another common phenomenon is how a dog will bring a tennis ball back to you and drop it where he "knows" you can see it. The trouble is, we tend to forget all the times that he brought it back and put it where we couldn't see it, or what's more likely, where we couldn't reach it. And if we can't reach it, then sometimes it's probably because it's not in a spot where we can see it. So the dog learned to put the tennis ball in the spot that would provide the quickest payoff of another throw, not because he knew that's where we could see it. (I'm working with a dog right now who still hasn't figured out that I can't reach the ball if I'm sitting on the bench at the dog run and he puts it under my legs--he just barks and barks at me.)

It's probably the same thing with stealing stuff when we have our eyes closed. Although since dogs are very sensitive to our emotions and our energy levels, it could very well be that they sense our inertness at a specific moment, which just happens to coincide with use having our eyes closed at that particular time. Do you see what I mean? And if they're successful at grabbing stuff when they sense that lack of energy on our part, they learn to do it at those times, and may even associate that lack of energy or focus on our part with those times our eyes are closed. But do they know that when our eyes are closed we can't see them? I think there are a couple of simpler steps involved in that behavior.

If anyone's really interested in looking deeper into this subject, Susan Friedman, a behaviorist on another board posted this link, which is to the first chapter of a book on cognitive science. The book is a kind of anthology, and I think the first chapter basically outlines the basics of what is and isn't cognition, reasoning, theory of mind, etc., from the points of view of various cognitive scientists, most of whom disagree on many issues. I found some of the studies and scientists (sorry) to be stuck in the last century. But overall I found the 1st chapter to be very helpful.

Here's the link: http://www.tiny.cc/rationalanimals (cut and paste if it doesn't automatically click on)

LCK
 

Lee Charles Kelley

The trouble is, we tend to forget all the times that he brought it back and put it where we couldn't see it, or what's more likely, where we couldn't reach it.

Another excellent point. And I am always first to admit when I might be wrong. I took away from that tv program the notion that dogs can tell when we can't see something when our eyes are closed. And it's an easy deduction to make. However, in other tests, we can see that dogs do look to humans, a part of their natural environment, for cues. But I agree, a dog may, after a while, figure out what eyes closed means. In which case, it would be learned behavior, rather than an proof that the dog knows we have a mind separate from theirs.