Learning, Memory and the Brain Can you can learn something without being able to remember that you remember it? New brain research sheds some light on the answer to that puzzler. Joe Palca leads a discussion on the issue.
NPR logo

Learning, Memory and the Brain

  • Download
  • <iframe src="https://www.npr.org/player/embed/4777024/4777025" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript
Learning, Memory and the Brain

Learning, Memory and the Brain

  • Download
  • <iframe src="https://www.npr.org/player/embed/4777024/4777025" width="100%" height="290" frameborder="0" scrolling="no" title="NPR embedded audio player">
  • Transcript

JOE PALCA, host:

From NPR News, this is TALK OF THE NATION/SCIENCE FRIDAY. I'm Joe Palca.

When you learn something, do you know that you're doing it? You can tell you're using your memory when you try to call up some piece of trivia, like: Who was that quarterback? It was--it was the third play--no, it was--I think it was the Super Bowl. No. You can remember that kind of stuff sometimes, but what about those things you do almost unconsciously, like navigating the roads to your house? In a work published this week in the journal Nature, researchers looked at that kind of learning, the kind of learning by habit. They studied two people with brain injuries that prevent them from forming new conscious memories and tried to teach them a simple task in which they picked out one object from a set of others. They repeated it many times, and each time the patients had no memory of what the task was or why they were doing it. But when they tested them after a period of time practicing this task, they got some very surprising results.

And joining me now to tell us what those results were is Larry Squire. He's a professor of psychiatry and neuroscience at the University of California School of Medicine in San Diego and a research career scientist at the Veterans Affairs Medical Center in San Diego. He's also one of the authors of the Nature paper.

Welcome to the program, Dr. Squire.

Dr. LARRY SQUIRE (Professor of Psychiatry and Neuroscience, University of California School of Medicine): Hello.

PALCA: Hello. And we'd like to hear your questions for Dr. Squire, as well. Our number is 800-989-8255. That's 1-800-989-TALK.

And, well, maybe I should start and make sure--did I describe the experiment properly? What exactly were these people asked to do?

Dr. SQUIRE: Yeah, you described it nicely. What they were asked to do was to learn a standard task that's been used to test experimental animals for over 50 years. What the task involved was eight pairs of objects which the patients saw one pair at a time. And on each pair, as it was presented, the task was to pick one--the object of the pair that had been designated as correct. So they saw all eight pairs of objects five times each day for a total of 40 trials. We tested them twice a week, 40 trials each visit. And as you described, over a period of several months they gradually and relentlessly improved their performance until they were performing almost perfectly in picking the correct object every time. Although, as they worked on the task and even as they completed the task, they--neither patient had any memory of having done the task before and couldn't describe what they were about to do that day.

PALCA: So they were learning, but they had no conscious understanding or appreciation of the fact that they were learning.

Dr. SQUIRE: Yeah, that's exactly right. The key idea is that the memory is not a single faculty of the mind and the memory is composed of several different systems that are supported by different brain systems. And the kind of memory that we ordinarily mean when we use the term `memory,' we call declarative memory or conscious memory, and that's dependent on the activity and the function of one part of the brain. But this other kind of memory, unconscious memory, non-declarative memory, is acquired more automatically, more in a trial-and-error kind of way, as a habit, and that depends on a different part of the brain.

PALCA: So when you asked these patients, `How did you know which one to pick?,' what did they say?

Dr. SQUIRE: Well, some of them are pretty surprised that they are doing it. At one moment during the late part of the testing, one of the patients said in the middle of testing. `How am I doing this?' And the--our co-author, Jennifer Frascino, asked him, `Well, are you doing it because you remember from the past which object was correct and which object was incorrect?' And he said, `No, it just goes from here,' pointing to his head, `from there,' pointing to the table.

PALCA: Wow. So are there--I mean, that was the other question I had. Are there--I mean, do we--is there learning that normally takes place in this habit way or is this something that, you know--that is the way animals learn, for example?

Dr. SQUIRE: Yeah, well, this is really the motivation for the study in the first place, because we've known for some time that experimental animals, rodents and monkeys being the animals most often studied, have this ability to learn gradually over hundreds of trials by trial and error in a way that depends on this part of the brain that we've called habit learning, the part of the brain called the basal ganglia that's impaired in Parkinson patients. So we've known for a long time that experimental animals can learn this way, but in the case of humans, we've never been sure whether they have this ability or whether we just do everything by memorization. You know, we--it's possible that this habit system in humans is rather diminished or overwhelmed by the tendency to memorize everything. So the question was: If we have that system, we should be able to employ it in the way that monkeys do. And, indeed, when these subjects learned the task, they learned it in about the number of trials that a monkey would take to do it, several hundred trials, about a thousand trials.

PALCA: So is this--I mean, this is fascinating, because it sounds like you're sort of getting at the brain structural differences between monkeys and human beings.

Dr. SQUIRE: Yeah, well, that's exactly right. It tells us that we do have this robust system for learning habits. It's going on all the time in us, but it's behind the scenes, so to speak, behind our conscious memory. We're responding all the time to the contingencies in the environment, to the rewards and punishments that we encounter. And gradually, our behavior changes, but we're not necessarily conscious of it. In fact, as we approach the world with our habits and our dispositions and our preferences, one might not be sure whether what one is doing is based on the way we always were or whether it actually is based on training and on having--the fact that we've responded to the rewards and punishments gradually given to us over the years.

PALCA: Is this--I'm just--now I'm just curious. Does this have any implications for how to help patients with this kind of strange memory loss in terms of coping with their circumstances?

Dr. SQUIRE: Well, in one sense it's limited, because the patients don't have the conscious knowledge that they have any information, so--but it does tell us that they have this ability to learn, through repetition, sequential acts, things that they have a chance to do a lot of the time. And in the case of the patients that we study, it really came like a lightning bolt and it explained to us how they can do as much as they do do in their daily life. One of the patients takes a little L-shaped walk around his house every day and never gets lost, and they find their way around their house all right, even though they've moved to the house after they developed their memory impairment and live with their families.

So it does tell us that we have this ability and that patients that have damage to their conscious memory system can acquire new habits gradually through repetition, and that system is available for us.

PALCA: But you're not saying that the conscious brain is overrated.

Dr. SQUIRE: Well, the conscious brain is what we really make use of as humans to develop...

PALCA: Yeah.

Dr. SQUIRE: ...our educational system and to pass our culture from one generation to the next.

PALCA: So we keep it around for a while.

Dr. SQUIRE: Yes.

PALCA: I think that's probably a good idea.

We'd like you to join us with the conscious part of your brains, if you care to. Our number's 800-989-8255. That's 1-800-989-TALK. Although, I'm sure there are people who call in to call-in shows just by habit, but I doubt--let's see if--Anthony from Chapel Hill, welcome to the program. You're not a habitual caller, are you?

ANTHONY (Caller): No, I'm not.

PALCA: Good. OK. So what's your question?

ANTHONY: I was--my question concerned the, I guess, sort of the neural underpinnings of this phenomena. Were the patients diagnosed with some sort of amnesia? And what brain areas are responsible for the habit-forming behaviors that you saw?

Dr. SQUIRE: Yes. So these patients both developed profound amnesia as a result of viral encephalitis. This is a virus that attacks the brain. It's quite profound and tragic. It often attacks the frontal lobes of the brain. It attacks different parts of the brain. But in these particular cases, the virus destroyed parts of the brain important for conscious memory. And those parts of the brain lie in the inner surface of the temporal lobes involving areas called the hippocampus and some adjacent structures that are anatomically related to the hippocampus. So tho--that's the system that's important for conscious memory and that's the system that was nearly completely destroyed in these two patients by the virus. The habit-learning system, on the other hand, depends on a completely different part of the brain called the basal ganglia that includes the caudate nucleus and other structures. And these structures are well known in medicine because they are the structures that are damaged in patients with Parkinson's disease. Parts of that system are damaged in Parkinson's disease.

ANTHONY: Is it...

PALCA: Does that answer--go ahead, Anthony.

ANTHONY: Yeah, is it believed that these two--I guess two systems are completely separate?

Dr. SQUIRE: As far as we can see, they're pretty independent of each other. That is that, you know, they work in parallel, would be the best way to think about it.


Dr. SQUIRE: And they're all--they're both working all the time and our behavior is a composite, you might say, of our habit systems and our skill systems on the one hand and our conscious memory systems on the other hand.

PALCA: OK, Anthony. Thank you for the call. Let's go to John in Louisville, Kentucky.


JOHN (Caller): Hello. Thank you. I'm a bass player. I play in several bands, but there are tunes that I've been playing for years and years. And if for some reason I have--I'm required to play it a different way with a different rhythm instead of walking or something, I can't remember the chords. It's as though my fingers have a memory of it outside from my conscious mind.

PALCA: Yeah, is that the kind of...

Dr. SQUIRE: Yeah.

PALCA: ...habit learning you're talking about, Larry Squire?

Dr. SQUIRE: Yeah. You know, that's very interesting. And, in fact, one of the features of habit learning is that it tends to be rigidly organized and is best expressed through the same response systems, in this case through the same, you know, hands and fingers that one used to acquire the information in the first place. And I might just mention that in this study one of the extraordinary things that we found was that when we took these eight pairs of objects involving all 16 objects, put them on on a pile in the middle of the table, we asked the subjects, `Now you've been doing this for a long time. Some of these objects have been called correct and some of them have been called incorrect. We'd like you to sort the objects into the two piles, a correct and an incorrect pile.'

Well, for our healthy subjects, that was just a trivial exercise because they had memorized the answers and then they could do it one way or they could do it another way. But the patients that had acquired the task through habit learning completely collapsed their performance when they were asked to do this. They actually performed no better than a flip of a coin in sorting the objects. And then when the task was reconstituted in its normal way, they shot back up to 90 to 100 percent correct. So the way that we understand the habit system is--very much, you know, coincides with how you're describing your experience in music, that these things we learn as habits are very specialized, they're rather inflexible and they're best expressed when you do it exactly the same way you learned it.

PALCA: John.

JOHN: It a different part of the brain then.

Prof. SQUIRE: But does not...

PALCA: He--John...

JOHN: It's a different part of the brain, then.

PALCA: A different part of the brain, yeah.


Prof. SQUIRE: Right, a different part of the brain.


JOHN: All right.

PALCA: Hey, John, thanks very much.

You know, I had an experience similar to John's. I had a locker in high school, and if you asked me, I couldn't tell you what the combination was for the lock on the locker--it was one of those spin, you know, numbers on it. But I walked up to the locker and I spinned the numbers, and it opened. And my fingers knew how to open the lock, but my brain--I couldn't tell you what the numbers were. Very--I guess that's something similar.

What would happen--you mentioned the sorting part of the experiment. What would happen if you, you know, adjusted the conditions or changed things slightly? Could these amnesiac subjects adapt to that, or did it have to be exactly the same each time for them to learn it?

Prof. SQUIRE: Yeah. It didn't have to be exactly the same every way. The core thing that they had learned was when they look at a pair of objections, they had a positive feeling about one of them and a less positive feeling about the other one. So when we asked them, for example, to verbalize their answer rather than reach for the object to make their choice, they had no difficulty doing that because they were essentially expressing the same choice. And so there are some minor things that one could do like that, but when one changes the whole way the information is being accessed by putting it in a pile, for example, that's when they really did have trouble.

PALCA: OK. Let's take a call now from June in Tulsa, Oklahoma. June, welcome to the program.

JANE (Caller): Thank you. It's Jane rather than...

PALCA: Oh, Jane, I'm sorry. It's my eyesight. I apologize.

JANE: I'm concerned about Alzheimer's. It seems like you'll have an Alzheimer's patient that will be able to do something for years and years and years and all of a sudden they forget how to do it. How does this study relate to that issue?

PALCA: Interesting question.

Prof. SQUIRE: Yeah.

PALCA: Thanks, Jane.

Prof. SQUIRE: Well, Alzheimer's disease is a progressive condition that begins with memory problems of the conscious type. The disease begins in the part of the brain that is damaged in these patients we've been studying. So the first signs of Alzheimer's disease are forgetfulness and losses of conscious memory. But as the disease progresses, the disease progresses into the cerebral cortex, more widely into the association cortexes, and ultimately both the habit system and the conscious memory system are going to be affected. So in the end, you know, one doesn't see the distinctions in patients like that because all of the systems are taken over by the disease.

PALCA: We're talking with Professor Larry Squire of the University of California School of Medicine in San Diego. And we're talking about a very interesting paper in this current issue of Nature magazine about different kinds of memory systems that we may share with monkeys.

You're listening to TALK OF THE NATION from NPR News.

And let's take another one of your calls. How about if we go to Matthew in Miami Beach? Matthew, welcome to the program.

MATTHEW (Caller): Hello, hi.


MATTHEW: My question is there was a guy named H.M. you're probably familiar with...

PALCA: Sure.

MATTHEW: ...who...

Prof. SQUIRE: Yes.

MATTHEW: ...learned that his father had passed away. And over a period of time, he began to understand that there was something wrong with his father, but he didn't know exactly what it was. And the person who was conducting the study--or I guess several people were conducting the study on H.M.--they thought that he was gaining an emotional attachment or he had an emotional understanding for memory that there was something wrong with his father. Is it possible that these individuals that were in this experiment also are gaining an emotional attachment or an emotional memory of this particular task that they are trying to complete?

PALCA: Interesting. Larry Squire?

Prof. SQUIRE: Yeah. Yeah. That's an interesting question. I think that, you know, the way we understand it is that primarily the basis for the learning was an attachment of mild positive and negative feeling, if you will, about the two objects. But emotion might be too strong of a word to describe it because in these patients, the area of the brain important for emotional learning, the amygdala, is also damaged. So one doesn't, I think, think of this past performance in terms of having strong emotional attachments or negative repulsions from one object to another, but rather something more subtle if you will: just the changing of response strength and the changing of the valance, the changing of disposition towards one object and another through trial and error and through feedback over hundreds of trials.

PALCA: Matthew, thanks very much for that call.

MATTHEW: Thank you very much.

PALCA: So let me just ask you as we--we're coming to the end of this segment, Larry Squire, but I'm just wondering: How far along are we in understanding human memory? How far does this particular insight take that?

Prof. SQUIRE: Well, memory is a large, large topic in neuroscience, and in neuroscience as a whole we're in the early days of the discipline, of course. It's only the middle of the last century that we really began to study the brain in earnest with modern biological techniques. But memory is an especially promising avenue for discovery because not only are we studying at the level that we're talking about today--at the level of brain systems, at the level of taxonomy, at the level of understanding how different kinds of memory systems operate and what brain systems support them--but we're also learning a lot at the molecular and genetic level about what happens in the brain exactly between neurons when memories take place. What does experience do to nerve cells? It strengthens them. And this can be studied in very simple animals, in vertebrate animals because of the universality of molecular biology. So we're beginning to learn a lot both at the systems level and we're learning a lot about how synapses can change, which is the fundamental piece of machinery that happens, that changes when we acquire new experiences.

PALCA: Exciting times. Well, Larry Squire, nice to talk to you again. Thanks for coming on the program.

Prof. SQUIRE: Thank you.

PALCA: Larry Squire is a professor of psychiatry and neuroscience at the University of California School of Medicine in San Diego and a research career scientist at the Veterans Affairs Medical Center in San Diego.

When we come back, something perhaps even more mysterious than the workings of the brain--religion and science. We'll talk about the Roman Catholic Church and some questions about evolution. So stay with us.

This is TALK OF THE NATION from NPR News.

Copyright © 2005 NPR. All rights reserved. Visit our website terms of use and permissions pages at www.npr.org for further information.

NPR transcripts are created on a rush deadline by Verb8tm, Inc., an NPR contractor, and produced using a proprietary transcription process developed with NPR. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.