Can a dog help to replace a person's failing memory? I recently wrote an article describing some new programs designed to train dogs to assist people with Alzheimer's disease and dementia (click here to see it). Only a day or two after that article appeared I received a phone call from an academic colleague who protested that dogs simply could not successfully serve in that capacity because they lack a particular form of memory that is needed. My colleague pointed out that Daniel Dennett, a philosopher and cognitive scientist from Tufts University has stated unambiguously that he was convinced that dogs have no episodic memory, and this is a viewpoint which is shared by a number of other philosophers and some behavioral scientists.
To understand what he meant you must first know that there are many types of memory. Psychologists often start by dividing memory into to large groupings that they call “explicit” or “implicit” memory. The easiest way to distinguish these is to note that explicit memories are the ones that you can describe or call into your mind at will, while implicit memories are automatic and not really conscious. Learned skills are good examples of implicit memories. Thus although you might remember how to ride a bicycle (since you can easily do it) trying to describe to someone else what you have to do to stay upright on a bicycle is virtually impossible. You know what to do, but you can’t make these actions conscious in such a way as to communicate them to others.
Explicit memories are memories are the ones which are easily brought into consciousness and which we can describe verbally. When we consider explicit memory it comes in two varieties, namely “episodic” and “semantic” memory. Episodic memory is memory for what you have personally experienced. When you answer a question about what you had for dinner last night, which clothes you wore yesterday, or you describe your first romantic kiss, you are recalling episodic memories. This is different from semantic memory which involves memory for facts. To answer a question such as “Who was George Washington?” or “What is the climate like on the moon?” will involve semantic memory. It is not episodic memory since you never met George Washington, nor have you visited the moon. Some people say that episodic memory is a sort of mental time travel in which you revisit events that you experienced by bringing them into consciousness. Episodic memory is not based upon practice or repetition, since most life events occur only once, and nonetheless are remembered. An important characteristic of episodic memories is that each contains specific data about “what,” “when,” and “where.”
I knew about Dennett’s beliefs since, in fact, we had once had a brief discussion about dogs' memory capacity and the nature of canine consciousness when he had visited my university to give a series of lectures. His beliefs about this matter puzzled me, even in light of only casual observation of dog behaviors. For example, many dog owners have a variety of "Find the object" phrases that they use and to which the dog responds appropriately. For example my dogs respond to, "Where's your ball?" by dashing around to find the ball and then bringing it to me. If it is inaccessible, they will usually stand near where they believe it to be and bark. My dogs also respond to "Where's Joannie?" which is a convenient phrase which helps me to locate my wife. On hearing it the dog goes to the room where they last saw my wife. If she is upstairs on in the basement, the dog will move the appropriate set of stairs and wait there. If she is out of the house the dog will usually go to the door that she used when she left. If the dog doesn't know where she is, or if she has moved since he last saw her, he will usually start to search for Joan. Each of these is an instance of episodic memory since the dog must remember where he last saw the item. This memory clearly has the required “when,” “what,” and “where” components, since we are asking for a location of a particular object when it was most recently seen based upon the dog's personal experience.
I pointed out the “Where’s Joannie?” situation to Dennett who did not appear to be much impressed. He would only grant that this was “episodic memory-like behavior.” He went on to note that he would have to think about this further and would get in touch with me when he found the flaw in my reasoning. He never did contact me again about the matter.
My colleague's protests that we simply could not have something like a "memory assistance dog" reminded me of what might have been the first dogs actually used in such a capacity. Around 2003 I interviewed John Dignard who lives in Wetaskiwin, a town in Alberta, Canada. Dignard was hit by a car at the age of five, and the accident caused brain damage. In the end he was left with learning difficulties and a very unreliable short-term memory. Before anything makes it into his long-term memory, it must be repeated and relearned many times. Early memories are still there, so that he can remember his phone number from when he was four, but new ones that are a problem. For instance it took him a year after his marriage to remember his wife's name. He told me "When you ask someone's name 600 times because you can't remember, it's very frustrating."
At a very pragmatic level, in the absence of short-term memory, simple tasks became nightmares. If Dignard went to a shopping mall, by the time he came out he usually had completely forgotten where his car was parked. It is in such situations where the episodic memory ability of a dog becomes important. Dignard can now go shopping with confidence because of a German Shepherd Dog named Goliath, who is his memory aide. Goliath is the third such memory assistance dog that he has had. Obviously Goliath can't help with names, phone numbers, or shopping lists, but the dog does serve the same purpose as the ball of string that Theseus let out as wended his way through the labyrinth in order to find his way back out after he slew the Minotaur. Goliath’s task is to lead his master back to the places that he can't remember, such as the way out of a building has only been in this one time. In other words, the dog must use his episodic memory to remember where an exit is, or where his owner's car was parked.
Dignard says “I'd be lost all the time without him. Now I just tell him ‘go to the exit door,’ or I tell him ‘back to the car,’ and he takes me there.” It is thus Goliath’s episodic memory that substitutes for the episodic memories that his master has such difficulty retrieving. It would be interesting to hear what Dr. Dennett might have to say about this.
Stanley Coren is the author of many books including: The Wisdom of Dogs; Do Dogs Dream? Born to Bark; The Modern Dog; Why Do Dogs Have Wet Noses? The Pawprints of History; How Dogs Think; How To Speak Dog; Why We Love the Dogs We Do; What Do Dogs Know? The Intelligence of Dogs; Why Does My Dog Act That Way? Understanding Dogs for Dummies; Sleep Thieves; The Left-hander Syndrome
Copyright SC Psychological Enterprises Ltd. May not be reprinted or reposted without permission