Do Dogs Have a "Mind's Eye?"
Can We Communicate With Dogs Using Mental Images?
Originally published in slightly different form on January 22, 2013 at PsychologyToday.com.
The Dog’s Visual System One of the ways cognitive scientists describe how the human and animal mind both process visual information is through mental representations or “cognitive maps” of physical objects. You show the dog a ball. He sees the ball, but how does he see it? Does the dog’s visual system create an internal representation of the ball in his mind? Can a dog also see pictures of things in his mind that lie outside his visual field? For instance, if you throw a ball and it goes under the sofa, you can usually tell from the dog’s behavior that he “knows” the ball is there, even though he can’t see it. But does some part of his mind also picture the ball being there? Another example: when a wolf is chasing a deer and deer disappears momentarily behind some trees an outcrop of rock, does the wolf “think” the deer has vanished, or does he follow the same trajectory as before, as if mentally tracking the deer’s movements in her mind? Of course we can’t know for certain if the dog actually has a picture of the ball in his mind or if the wolf visualizes the deer’s trajectory, but we do know that dogs dream in much the same way we do because—as Dr. Stanley Coren has pointed out—they exhibit REM sleep: why would the dog’s eyes be moving rapidly if there was no internal visual component to his experience? And if his eyes are closed, then such images must exist within the “mind’s eye.” Find the Toy! Does your dog have a favorite toy? When you say, “Get the [fill in the blank]!” does the dog bring the toy you named? Most dogs do. They’ve formed a memory-bank, through a combination of pleasurable emotions and pattern recognition: the aural pattern of the word “ball” is matched in the dog’s mind with the visual pattern of the ball’s shape, along with the pleasurable feeling of having the ball in his mouth. Two border collies in the past 10 years or so have shown an amazing ability to remember the names of their toys. Rico knew the names of over 200 toys. Chaser knows the names of more than 1000. (It’s interesting that canine memory tests usually involve toys, not household objects like furniture, clothing, or cooking utensils.) Stanley Coren tells of a study showing that dogs may be able to make mental connections between symbolic [iconic] representations of physical objects and the objects themselves. The suggestion is that this goes beyond pattern recognition and puts a dog’s level of cognitive development on par with that of a 2 -3 year-old child. “It is usually not until after a child’s second or third birthday,” writes Coren, “that they can infer what an adult wants when they are shown a replica of an item as an example. … nonhuman animals find this very difficult, if not ... impossible. However dogs seem to be a special case…” In the study, “Domestic dogs comprehend human communication with iconic signs,” 5 dogs were tested (all border collies, by the way, including the famous Rico) separately in their own homes where, in separate tests, they were shown replicas of 3 objects: one an exact replica of a toy, another a miniaturized version of another toy, and the third a photograph of still another toy. All 5 dogs lived as pets with their owners. 3 were experienced at playing fetch, the others were not. When the dogs were shown a replica of an object, they were then asked to go into the next room, find the object in question (out of 8 possible choices), and bring it back. The command given was “Bring here.” The object’s names were not mentioned. According to the researchers, the 3 experienced fetchers brought the exact replica and the life-size version of the miniature at a percentage rate that was well above pure chance. The 2 inexperienced dogs didn’t do as well. When it came to fetching the object shown in the photograph, that was more difficult for all but one of the dogs. Kaminski et al conclude that “the most reasonable interpretation of dogs’ success in the replica tasks is that they understood that by showing the iconic sign the human was trying to communicate something to them.” I think that would depend on what definition of communication is being used. Certainly there are a number of tests showing that dogs (and in some cases, even wolves) show a sensitivity to human gestures and eye contact, and may be able to take cues from humans in this way, though this works best if the cues are part of a game involving the dog’s (or wolf’s) hunting instincts. Either way it seems clear that the dogs felt (not understood) that the humans wanted them to do something—fetch a toy.
Is Something Else Going On? Former Cambridge biologist Rupert Sheldrake writes, “My research on telepathy in animals (published in detail in a series of papers found here) led me to see telepathy as a normal, rather than paranormal. ... aspect of communication between members of animal social groups.” Even Lloyd Morgan (of Morgan’s canon) entertained the possibility that animals communicate via mental images. In The Limits of Animal Intelligence—when describing a dog coming home from a walk, tired and hungry—Morgan wrote, “I for one, would not feel disposed to question that he has in his mind’s eye a more or less definite idea of a bone.” (1892.) Morgan goes on to define “idea” as a visual image. If Morgan is right, that dogs can communicate through visual images, then this could mean that the dogs in the 2009 study might not have been responding to the iconic representations of toys they were being shown by their owners, but to the pictures of those objects held in the owner’s mind! (The lack of success with the photos could be explained by the fact that they were 2, not 3 dimensional, making it slightly harder for both dog and owner to translate the images to real objects.) A 2011 study shows that detection dogs will sometimes give false positives when their handlers think they know where certain substances have been hidden. The conclusion drawn there was that the handlers unconsciously tipped off the locations to the dogs. But I think this too could have been the result of unintentional telepathic communication via mental images. In order to test whether mental images were at play in the iconic communication study, we would repeat it, as it was initially done, but add 2 extra controls: neither the researcher nor the owner would be allowed to see the icon shown to the dog. This could be accomplished by having each object (or photo) put into a box with a door that could be opened and closed. A third party would place the object (or photo) in the box, and would assign that box with a number to be checked later against the object that the dog retrieved. The owner would then open the door and show the dog the object without being able to see it for him or herself. If the dogs still retrieved the toys at a rate higher than chance, we would know that the dogs were responding only to the image of the object they had viewed, not whatever mental images the owner and researcher may or may not have had in mind. If not—if the dogs’ retrieval rate fell below pure chance—we would have evidence that something else (perhaps of a somewthat telepathic nature) was going on in this study. LCK “Life Is an Adventure—Where Will Your Dog Take You?”