Dogs are wonderful creatures. Our dogs recognize me and are always happy to see me. Dogs are also smart and successful creatures. Our dogs have learned several cute tricks. But dogs (and other non-human animals) are missing something we take for granted: Episodic memory. Dogs don't remember what happened yesterday and don't plan for tomorrow.
In defining episodic memory, Endel Tulving argued that it is unique to humans. Experience influences all animals. Most mammals and birds can build complex sets of knowledge or semantic memory. You and I also remember the experience of learning these complex sets of information. Dogs don't.
Episodic remembering is mental time travel and depends on a few crucial cognitive capabilities. First, to experience episodic remembering, an individual must have a sense of self. Most non-human animals have a dramatically different experience of self than we do. For example, most animals (and young humans) fail to identify themselves in mirrors. If I look in a mirror and see that I have something stuck between my teeth, I try to correct the problem. (I also wonder why my friends didn't tell me I had something stuck between my teeth.) In contrast, put a red dot on a child's forehead, put the child in front of a mirror, and watch what happens. Young children are more likely to reach for the baby in the mirror than for their own foreheads. Dogs treat the dog in the mirror as another dog; not as themselves. Most animals fail at the red dot mirror task.
Self-concept is not, however, enough to ensure episodic remembering. Mental time travel is the other critical cognitive capability. I understand that yesterday is different from today and that tomorrow will be different as well. We realize that when we remember, the mental experience is a disjointed slice of time. Thus episodic remembering is the combination of self-concept and mental time travel: recollecting the self in that other time period. Mental time travel also enables planning for the future. Dogs don't plan for particular future events although they have a general expectation of when dinner will appear.
Tulving also argued that since episodic memory is a recent evolutionary development, it is particularly likely to suffer damage and loss. Anterograde amnesia is the failure to encode and remember new episodic memories. Anterograde amnesiacs can learn from single experiences without recollecting the experience. They retain a clear sense of self, but they have difficulty with time as personally experienced. Because they lack episodic memory, they can't recall what occurred just before the present moment and constantly feel like they just woke up. If you meet an anterograde amnesiac, leave the room, and return after 10 minutes, you'll remember having met the individual, but the amnesiac won't remember having met you.
My dogs display this particular failure of episodic remembering. If I walk into the backyard, the dogs are overjoyed to see me and act like they haven't seen me for days. If I stay in the backyard, they quickly become bored with me. If I go inside and return after 10 to 15 minutes, my dogs are overjoyed to see me and act like they haven't seen me in days. They don't remember that I was in the backyard just a few minutes ago.
Arguing against Tulving's notion that episodic remembering is unique to humans is hard. Showing the impact of a single experience is not enough. Even without episodic memory, humans can show the impact of single events. Anterograde amnesiacs can learn fear, learn new skills, and gain new conceptual knowledge. Normal humans also gain knowledge without remembering when and where they learned the information (see my earlier post on Haven't I Seen You Somewhere Before).
Although I appreciate Tulving's conception of episodic memory, I've always been troubled by the difficulty of documenting that other animals have episodic memory. Episodic remembering hinges on the conscious experience of the self in some other time and place. Episodic memory is thus hard to demonstrate without the verbal ability to describe conscious experience.
Nonetheless, in a recently edited volume (The Missing Link in Cognition: Origins of Self Reflective Consciousness, edited by Terrance and Metcalf), several individuals have taken up the challenge. In my next post, I'll present the counter-argument: Dogs don't remember, but maybe chimps do. Since some non-human primates can perform self-recognition with mirrors, they may perform episodic remembering. Even if they can't describe their memories, chimps may engage in mental time travel. My dogs, however, are stuck in an eternal present.