The business of memory

Memory's a funny thing. It's inherently subjective—there’s no way to tell exactly what someone else sees with their mind's eye when they recall a moment from their past. And yet memories are at the core of so many of our everyday interactions, our conversations with friends, family and coworkers, that generally we have a pretty good idea of how our memory compares to others. Except when we don't.

Dr. Brian Levine is a neuropsychologist and cognitive neuroscientist at Baycrest’s Rotman Research Institute and at the University of Toronto, who's specialized in memory for over 20 years. Most of his work has been with patients with severe memory issues, often the result of brain injuries and diseases. But in 2006, Levine and his colleagues were the first to describe the syndrome of "Severely Deficient Autobiographic Memory," or SDAM. People with SDAM are healthy but unable to recollect the autobiographic details or "episodes" of their life—they understand intellectually the events they've experienced in the past, they just lack the first-person recollection of it. A key difference between someone with SDAM and someone who suffers from amnesia due to brain injury is that people with SDAM have never had the ability to recall those details. In many cases, Levine says, they may not even realize that their minds work differently from most others.

At the other end of the spectrum are people with Highly Superior Autobiographic Memory, (HSAM), also known as "super-rememberers." People with HSAM are able to recall an enormous amount of detail about their past—in many cases they know not only what they had for lunch last Tuesday, but also what they had for lunch on August 4th of last year. But most of the research Levine does is with subjects who are somewhere in the middle, conducting extensive autobiographical interviews and questionnaires while simultaneously mapping people's brains to get a sense of the areas responsible for different kinds of memories, and how we process and retain information. Understanding why and how we remember the things we do, Levine says, has enormous implications for our everyday lives—an idea that seems doubly true for research.

dscout sat down with Levine to chat about memory, good stories, and the power of the mind.

dscout: You're a neuropsychologist and cognitive neuroscientist. What drew you to studying memory?

Brian Levine: I'm trained as a clinical neuropsychologist, which means I have training in the assessment of brain functioning and behavior. And memory is part of that. But my interest in it probably started with my interest in stories, which has nothing to do with science. I like to talk to people, find out their stories. Even just friends or my family, when they tell stories I always want to know the details. “What were you thinking at the moment?” “What was the exact sequence of events?” That sort of thing.

And I've always been particularly in awe of how the great writers, people like Richard Ford and Alice Munro, tell stories. There's a level of detail and a way they embellish a scene that makes it feel real. They bring color and life to it.

They can put you right in the scene.

Exactly. I’m also hugely interested in biographies—I just read Patti Smith’s Just Kids. She has so many incredible stories about things like how she met Robert Mapplethorpe in Washington Square Park, running into him by coincidence.

It would be interesting to look at a writer’s memory. To look at that ability to describe the internal narratives going on in people’s heads, the things that we think that we can’t even really articulate. It’s something that helps us understand the human experience more, but it’s hard to understand how someone conjures that.

Yes, with fiction especially. The ability of the fiction writer to describe an experience of someone that never existed, with such incredible detail that it can make you cry? It’s astonishing.

When you think about those kinds of stories as it relates to memory, though, the research has historically been pretty lacking. Diagnostic memory tests tend to be based on lists of words or pictures. Which is fine for what it is, and it’s an important thing to do. But it doesn’t give you that level of detail I was talking about, or help you understand what’s meaningful to someone when they tell stories about their past. It doesn’t get at the humanity of it.

When I started doing this twenty-five, thirty years ago, I found there were people who could do well on lab tests, but were actually impaired in their ability to tell stories about their past. The memory clinic could send them home saying “you’re fine,” when in fact that wasn’t the case.

In the late nineties I had a patient we called “Patient ML,” who was part of a study the lab was doing on traumatic brain injury. He’d had a bicycle accident and had a severe traumatic brain injury. When he recovered he had a “good recovery,” except that he actually lost all these details and episodes from his past. He’d lost his episodic memory, which is what we call personal, autobiographical memories of past experiences.

That’s the ability to be able to recreate or visualize a scene in your mind, and recollect the details of an event or something that’s happened to you?

Yes.

So Patient ML, he remembered what his memory had been like before, and he could tell the difference between the before and after?

Actually at first, he kind of covered it up. Waking up from a coma, it’s hard to imagine what that’s like. I think he was more interested in appearing functional. But then his family noticed the problem. (There's a documentary you can check out called “Plastic Fantastic Brain” that tells his story.) By the time I saw him he was fully aware he had this problem.

It was also right around the time that functional brain imaging was getting started. At the time it was PET scans, positron emission tomography, though now it’s usually fMRI, functional Magnetic Resonance Imaging. But the PET was a great tool because it allowed us to look into the brain while people were doing tasks. I got really interested in probing the brain that way and helping people.

In thinking about your work in the context of research, it seems to present a whole host of challenges. Researchers rely on people’s memories to be accurate, but a lot of your work is with people whose memories don’t work the same way as most. How do you even begin to approach that?

It’s actually not as complicated as you might think. The first thing is just to listen to people. People actually are very articulate about their experiences. I’m a trained psychologist so I speak technically, but if you just let people talk, you’ll find they actually give pretty good descriptions of their inner state. Ultimately a lot of it is subjective, so our tests are really proxies. They're a barometer for a subjective state.

Typically we’d just say, tell me a story about an event—we don’t need someone to tell us about their entire vacation in Greece, just focus on something that happened in one day. Then we would ask eight to ten follow-up questions: “Are there any visual images that come to mind?” “Where did this take place?” “What sort of thoughts or feelings did you have during this event?”

And we look at the details in these autobiographical stories, and categorize those details as either episodic memory, which is re-experiencing, versus more factual statements. There can be great heterogeneity in expression within a single story. I could tell you a story about World War II that contains no episodes, just a factual statement of who the parties were and what they were trying to achieve. Compare that to a story full of details about people, happenings, events that are specific in time.

The hardest part was getting everyone on the research team to agree what kinds of details constituted episodic memory, vs just an accounting of the facts. For instance, if someone says something like “Alice was wearing a red hat.” You could say that because you know that Alice likes red hats, but you may not have a specific recollection that in that instance she was wearing a red hat. Or you could be seeing it in your mind. One big question was, what do we do with that kind of detail? Do we give the person credit for seeing the hat in their mind? What if they just knew that she liked red hats? So we decided every time someone said something like that, we would apply what we call the benefit of the doubt rule. Because if we tried to parse how much the person knew about Alice's preferences or whatever, that's a waste of time. That could never work. So we just said, any time someone says something that could reasonably be assumed to reflect re-experiencing, we just give them credit for it.

It seems to take the importance of the methodology to a new level. You aren’t just trying to get information, you’re trying to decide if what someone is telling you is reliable—and they may not even be a hundred percent certain if it is.

One thing you learn in the business of memory is that false memory does exist. Things we remember are not necessarily accurate, sometimes the details aren’t even the right details. Memory confidence is not necessarily the same as memory accuracy.

A lot of your research has focused on people who, like Patient ML, don’t have an ability to visualize or form autobiographical memories. But unlike Patient ML, it isn’t because they’ve suffered a trauma, it’s just how they were born, it’s the way their brain has always worked.

That’s what we call SDAM, or Severely Deficient Autobiographical Memory.

It seems like the discovery of this condition has opened up a lot of new avenues into looking at how our memories work.

Absolutely. There are pretty vast differences in the way people remember things, how much detail they can recall. Most people have a sense of how their memory stands relative to others, how well they’re able to remember events compared to their friends and family. Personally my memory isn't great, it's not particularly bad, it's probably around average. My reactions are often driven by images. For instance, I’m a dad, and that’s true when it comes to dealing with safety issues, like your kids running out into the street. My reaction to those kinds of events are often image-based.

The differences in our memories is something that’s sort of self-evident, but has never been studied in depth scientifically. But these extremes, SDAM, and its cousin HSAM, which is highly superior autobiographical memory, point to the fact that these differences exist and that they could be meaningful in all sorts of ways that have implications for day to day life and health.

What are some of the biggest ways those differences tend to resonate through people's day to day life?

One idea that’s been very influential in the field of memory is that good episodic autobiographical memory is necessary for forming plans and thinking about the future. That you use recollection of past events to plan future events. So if I'm packing my suitcase to go to a hot climate, I might reflect on the last time I was in such a place and that would influence how I pack. But people with SDAM, who can’t recollect, don’t seem to have any problems planning. So it calls into question the idea that you have to have episodic memory and that it's even necessarily always beneficial. I can think of instances when I’ve made bad decisions because I was relying too much on a single episode or an event that previously happened to inform my thinking. And in fact it may be more important to have the ability to abstract information that cuts across episodes, like factual information.

People who don't have the ability have adapted and learned other ways to do things.

Exactly. In fact many of them are very high functioning and in jobs that require a lot of reasoning—if/then reasoning. There’s a very well-known blog by the software engineer Blake Ross, who is famous for developing the web browser Firefox. He has something called aphantasia. It’s related to SDAM, it’s an inability to form visual images and recollect past events. It seems like his thinking style may relate to his technical software skills.

Many people with SDAM don't even realize they have it until they’re middle-aged. I've gotten emails from people who’ve read about my lab’s work and research into SDAM and said, “Oh my God, I can't believe this. That's me, and not only did I have no idea that there was anyone else like this, I had no idea that I was like this.”

Now other people do know, I should be clear, that's not everybody. Other people know and it bothers them. So there's a whole range of subjectivity on this.

It’s striking that people who don’t have the same memory ability as others don’t necessarily seem to know it. Are there ways people can start to better understand the kind of memory that they do have?

Right now there really isn’t, other than just talking to people and comparing your experience to others. But you can help contribute to our ongoing research by taking a test we’ve developed called the Survey of Autobiographical Memory: it’s at memoryinventory.com. By doing that you’re helping us understand more and more. The discoveries that are happening right now are really exciting. They’re opening up a whole line of questions that could keep me busy for many years.