share interactive transcript request transcript/captions live captions
download
|
MyPlaylist
PRESENTER: For today, we are privileged to have with us Dr. Charles Brainard. He currently holds a position as professor and department chair in human development. And Dr. Brainard's research and teaching are actually interwoven in the fact that his latest research is included in some of his courses that he teaches.
Today's topic, memory and aging, evolved from his scientific study of the development of adult cognitive process and the effects of normal aging and the diseases of late adulthood. Today's session is going to be taped by CornellCast, and it will be available for viewing in about a week. Because this is being taped, I would like to ask that you hold your questions to the very end of the presentation. With that, I'd like to turn this over to Dr. Brainard.
CHARLES BRAINARD: Anybody here from Kendall or Longview? OK, good. I won't have to apologize for the fact that one or two of the overheads you would have seen before.
Well, here's what we're going to talk about today. The general topic today is the memory declines that we hear a lot of complaints about during healthy aging and also during cognitive impairment late in life. The first point I want to make is that when people talk about or complain about memory, the first reaction of an experimental psychologist like me, who studies memory, is what kind of memory.
In psychology, memory is a many-splendored thing. And there are a number of basic systems of memory. I put three up here, which I'll define very quickly-- episodic memory, semantic memory, and procedural memory. And when people have memory complaints, they're not about most of the different memory systems. They're about only one of them, which is episodic memory.
So let's just quickly say what these three basic systems of memory are. Episodic memory is what most people mean when they say memory, as I said. And it's our memory for events that have specific place and time tags.
What happened to me at a specific place at a specific time? Did I have a hot dog to eat at the hockey game last Saturday? That's episodic memory.
These other two major forms of memory-- semantic memory and procedural memory-- semantic memory is our store of concepts, our understanding that, say, a dog is an animal, maple is a tree, Chevy is a car. That's semantic memory, conceptual memory.
And finally, procedural memory is our memory system that has to do with our knowledge and our prior learning of how to do things-- how to ride a bike, how to swim, and so on, how to deal cards. Do you do them left handed or right handed, from the top of the deck or the bottom of the deck?
That's all procedural memory. You do have to learn how to do those kinds of things. They do improve as a function of learning. So they are all memory. So it's basically episodic memory that we're talking about when we're talking about declines in memory during healthy aging and in cognitive impairment.
The research that I do in this general concept-- and this represents the themes I'll talk about today-- attempts to parse this up. So it's about episodic memory declines. It's attempting to-- within Episodic memory-- there are specific components within episodic memory. So what we're trying to do is identify the specific types of episodic memory that decline during healthy aging and in impairment late in life. And we want to pinpoint their brain regions once we've identified those specific types.
But equally important-- maybe even more important-- is we want to also identify the specific types of episodic memory that are spared during healthy aging and impairment, that don't decline. And we want to pinpoint their brain regions, too. And then last, we want to develop a theory that explains all this and that makes new predictions about memory and aging.
So now I'm going to say something that I'm not going to say any more throughout this talk, but it's actually, to me, the most important thing about work in this area. There's really nothing, to paraphrase Kurt Lewin, in this area that's so practical as a good theory.
When you have a theory that really explains the reasons for what declines and what is spared in memory declines, this is really essential for everything we want to do, like developing new treatments for normal memory declines and for the kinds of declines that are associated with MCI, which I'll talk about later, the dominant form of cognitive impairment and for dementia.
There's many other reasons why you need a good theory, too, but these are a couple that will all resonate, too. But I'm really not going to say anything more about theories and how we explain these things today.
Let's start by talking about some general patterns, empirical patterns that we know about in memory declines in healthy aging. First of all, we know that true recall-- your ability to recall exactly what it was that happened to you at a particular place and time, what you had to eat and what you had to drink at the hockey game-- true recall tends to decline during healthy aging. And false recall, which is recalling things that didn't actually happen to you-- oh, I drank a Coke, when in fact you didn't, you drank a 7-Up-- that tends to increase with age.
On the other hand, the other basic form of memory-- the two basic forms of memory, in terms of how we test them, are recall and recognition-- the accuracy of recognition memory tends to be largely spared during healthy aging. Same kind of episodic memories-- what you had to eat and what you had to drink at the hockey game-- except that instead of having to come up with it on your own, instead of having to respond to what did you have to eat and what did you have to drink at the hockey game, the question is did you drink a Coke.
Did you have a hamburger? Did you eat a slice of pizza? You respond yes, no. Did that happen to you? Didn't that happen? That's recognition. That sort of episodic memory basically tends to be spared during healthy aging.
So what happens in the study of memory decline and healthy aging-- we tend to focus on recall. When we complain about memory loss, we're basically complaining about loss of recall, because recognition is spared. And when we're doing research on what it is that deteriorates, we study recall, because that's what declines. So this is basically all going to be about-- today about memory decline in healthy aging and in cognitive impairment where we're looking at recall memory.
Now, it turns out-- I said earlier in the talk, we want to identify specific things that deteriorate and specific things that don't-- it turns out that recall memory is a many-splendored thing, too. Not too many splendored. It involves, basically, three component processes.
So when you recall something-- when you succeed in remembering I had a hot dog at the hockey game-- there are three processes that are involved. One way that you can remember that is what we call direct access. It's the recovery of vivid, verbatim traces of individual events. Mostly in the work we're going to be talking about, there are going to be words on word lists is going to be the experimental paradigm that we use to conduct very controlled experiments on this.
This is the kind of memory where, when you remember something, it rings in your mind's ear. It flashes in your mind's eye. You smell it with your mind's nose. You relive the event in a sensory kind of way. That's vivid verbatim memory.
The other kind of way that you recall past events is what's called reconstruction. You basically reconstruct the events that happened to you from gist memories. In other words, there are certain characteristic events. Think of your life. Our lives are a series of tableau. Each day in our lives, we pass through a series of typical situations.
That is getting up in the morning, having breakfast, getting dressed, getting ready to go to work, driving in a car, and so on. These are tableau that have a connected meaning and certain prototypical events. So if we can place ourselves in one of those prototypical situations, we can reconstruct what happened to us.
So that's the second way that we reconstruct. And I've given an example here. If you were studying a word list in an experiment, and you had some birds on the list and you had some household furniture items on the list, you could reconstruct the word "bird," or the word "oriole" from knowledge that there were birds on the list. And you could reconstruct the word "chair" from knowledge that there were items of household furniture on the list. So that's reconstruction. It's very simple.
The final process is what we call a slave process that operates on reconstruction. When you reconstruct something, notice that it doesn't ring in your mind's ear and flash in your mind's eye. The subjective phenomenology that's associated with that reconstruction is what you might call inference, figuring out.
Now, to us, subjectively, that figuring out thing is nowhere near as trustworthy as something flashing in our mind's eye and ringing in our mind's ear. So what we have to do when we reconstruct a piece of information is we then have to assess it. How good is it? Do I really trust that enough to put it out there? That's the judgment process. So if you reconstruct the word "oriole" from knowledge that there were some birds on a word list, now do you trust that enough to put it out? That's the judgment process.
Now, I'm going to be talking a lot about verbatim and gist memory in the course of this talk. So let's just give a concrete example of what we mean by that since we've already been talking about word lists. So let's take a prototypical task and think about it through the rest of this talk.
Imagine-- and this is the kind of task that's used in psychological testing, in the study of dementia and cognitive impairment all the time. So this is a typical task. So imagine you're confronted with the task of studying and recalling a list of 20 words. And the words on the list consist of things like robin, oriole, and then there's some more. And then you have Ford and Chevy, and then there's some more. And then you have table and chair, and then there's some more. And then finally, California and Michigan. That's the list.
Now, verbatim memory is simply recalling that some of those words were on the list, like robin, Ford, table, and California. Gist memory, on the other hand, is recalling the concepts that were on the list-- that there were some birds, cars, furniture, and states, independent of what they were. So this will be our point of reference.
So warning, warning, warning-- the next two overheads are the boring part of the talk. And if you persist through these next two overheads, there will be payoff after that. So the reason why it's crucial to keep these three components separated that I mentioned is it turns out that memory declines don't involve all of them.
In particular in healthy aging, it tends to evolve only one of them. So you need to separate them. That's the motivation. And basically, the way we do that involves mathematical models, so that's the boring part.
Basically, what happens is people perform the kind of task-- tasks like the one that I just described. So they study and recall lists of words or lists of other kinds of things, like sentences and pictures. And then that generates sequences of responses, errors, and successes. And then we have a mathematical model that you can apply to that. There's lots of different kinds of models that do the same thing. But there's a family of mathematical models you can apply to those data that will extract out each of those three components.
There's nothing mysterious about these models. If you think back to high school algebra or college algebra or college calculus, all it is is you have a series of equations, and these memory processes figure as-- and the equations express performance on these recall tasks as a function of these processes. And once the data are available, you can solve the equations for the processes.
So measuring these kinds of things really is just like high school algebra problems in which we solve equations for a series of unknown variables. It just happens that there are components of recall in this case. So it's not at all unlike acceleration equals force divided by mass.
If you have any two of those values, you can solve for the third. That's really all that's involved, except the known values that you're plugging in are from data in recall experiments. That's the boring part of the talk.
So using these techniques, what have we found? Well, I'll say first what we found in basic research on young adults. And then I'll say what we found by applying these same techniques to healthy aging. One basic thing that we found is that it's a whole heck of a lot harder to base recall on this direct, vivid, verbatim access thing. That's a lot tougher for people to do.
It's much easier to base recall on reconstruction, it turns out. It's much easier from the start. It hangs around longer. And reconstruction becomes easier to use with practice, whereas direct verbatim vivid recall doesn't.
What we've found by measuring these different component processes during healthy aging is that the bottom two processes that I mentioned-- reconstructing the events of your life from just memories of what happened to you, and then the quality of the judgments that you make about the accuracy of your reconstructions-- these kinds of things are really spared during healthy aging.
This reconstruction thing, there's little or no decline in it as long as people remain cognitively healthy into the 90s. And the judgment process-- your ability to make an accurate assessment of the quality of your reconstruction-- it doesn't seem to decline, either.
What does decline is recall that's based on this vivid, direct access-- ringing in the mind's ear, flashing in the mind's eye. That does decline steadily from midlife through late adulthood.
So what this leads to, by way of summary, is what you could call a verbatim gist rule of healthy aging. When you're complaining about memory declines, basically, the complaint turns out to be about declines in recall that are based on declines in the ability to base recall on the verbatim details of experience, memory to be able to recall the verbatim details of experience.
And what you have when you're recalling the details of experience is a sparing of recall that's based on reconstruction from the gist of your experience. And you also have a sparing, as I said, of judgments and inferences about the contents of memory. So memory that's based on the gist of experience-- reconstruction, judgment-- basically spared. Memory that's based on recall-- a precise, verbatim details of experience-- not spared, declines steadily.
So I said we want to pinpoint the brain regions of these things, too. So here's a toy brain for you to look at. Well, it's less of a toy brain than it's a pretty real brain, only it's been colored with the major lobes-- frontal, parietal, occipital, and temporal lobes. And we'll have more to say, particularly about what's going on in the frontal and temporal lobes in a minute.
These three processes that change and don't change-- here's basically how they map with those areas of the brain. Gist memory, as I defined it earlier, the activation areas-- frontal lobes, prefrontal. Judgment-- frontal cortical areas.
And the verbatim memory, which is the one that shows these declines, is behind the ears here, behind in this part of the brain. And we're going to have a picture of that, and in particular, in an area of the medial temporal lobes that's known as the hippocampus and the parahippocampal region, particularly what's called the parahippocampal gyrus.
Now, I'm going to show you a slide now that's going to have a toy brain on the left, and it's going to have a sliced real brain on the right. And in the toy brain on the left, the hippocampal region will be designated with a little arrow. And it's going to jump out at you so you can see a little bit more of it. So see it's lighting up?
So here's the hippocampal region here, and here is the parahippocampal gyrus here. And so these are-- there are certainly other areas of the brain that are involved in this verbatim memory stuff, vivid verbatim recollection. But these two areas are the ones that have been very strongly implicated in the most research that's going on in this area, all different kinds of research.
In the slides at the bottom, since we're about to transition now from healthy aging to cognitive impairment late in life, I have put some information here about the volume in these two brain regions as a function of whether or not you are testing older adults that are diagnosed as normal cognitively or are diagnosed as being cognitively impaired, either mildly cognitively impaired or with Alzheimer's dementia. So MCI is a non-dementia form of cognitive impairment that's known as Mild Cognitive Impairment that tends to be the prodromal, the precondition for Alzheimer's dementia. And then Alzheimer's dementia.
And basically, what you see is a pretty complete overlap in the spread of tissue volume in these areas for controls in mild cognitive impaired. But the mean, which is this line here, is clearly lower for the mild cognitively impaired. And then when you get to Alzheimer's dementia-- and this is hippocampus. This is parahippocampal gyrus over here. You'll see clearly a dramatic difference for Alzheimer's dementia. And then over here for the parahippocampal gyrus, you see a bigger drop from here to here.
So now let's talk about memory declines late in life that are not healthy, that are not normal, that represent disease, so pathological declines in memory during aging. We're going to talk about two broad categories, which I've already mentioned. One is what we will call-- or what is called cognitive impairment without dementia-- and I'll say very precisely what that means-- and then dementia itself.
Now, in the next couple of slides after this, I will define these two things in detail. But now I just want to give you the general topography of the two conditions. In cognitive impairment without dementia, there are a couple of categories that are important, because they predict what's going to go on in the future.
One is called mild cognitive impairment, and this is a memory disease. So this is about memory. Then there's another form-- or various other forms-- of cognitive impairment without dementia that are not memory diseases. In most instances, they are forms of non-demented cognitive impairment that come from other medical conditions-- stroke, cardiovascular disease, secondary diabetes. All of these produce stresses on the brain that result in cognitive impairment. Now, we're going to see that in terms of their consequences for dementia, these are very different conditions.
Then there's dementia proper. And we're talking about two basic forms of dementia-- Alzheimer's dementia and then all the other kinds of dementias. And why do we do it like that, although in the category of other dementias, there are a whole lot of things? The reason why we split it up like that is that Alzheimer's dementia is most of the ballgame.
So people who receive dementia diagnoses late in life, 2/3 to 70% are diagnosed as either probable Alzheimer's disease or possible Alzheimer's disease. So this is a natural way to split things up. Alzheimer's disease is a memory disease. The other dementias are not memory diseases.
And once again, you have things like dementia that's resulting from some of the same conditions that produce cognitive impairment without dementia-- things like stroke, cardiovascular disease. Alcoholism is another one. [INAUDIBLE] is another one, and so on.
Now, let's talk about Alzheimer's dementia first and exactly what that is. As I said, we're going to all lead back to memory with respect to both of these things. It's, by far, the most common form of dementia. Roughly 20% of Americans over age 75 is estimated with Alzheimer's dementia.
The interesting thing about it is it's a diagnosis by exclusion of other diseases. So it's a dementia for which the symptoms, which are cognitive, which I'll mention in a moment, cannot be produced by other diseases. And there are many other diseases that can produce these cognitive symptoms. So you exclude the diseases. And when you've ruled out all of the disease possibilities, that's when you've got an Alzheimer's classification if the cognitive criteria are met.
Now, what are the cognitive criteria? They come from the DSM IV, the Diagnostic and Statistical Manual of the American Psychiatric Association. There are five cognitive criteria-- episodic memory-- so clinically significant impairments in episodic memory-- what's called executive function, language, motor function, and spatial ability.
Now, what happens is-- I said Alzheimer's dementia is a memory disease. What happens is you must have a clinically significant memory impairment. If you don't have that, you don't have Alzheimer's disease. So you have to meet that one.
Then you have to have a clinically significant impairment in at least one of these other areas. That's why it's a memory disease. Whereas with other forms of dementia, you don't have to have a memory impairment. You can just have clinically significant impairment in any two of these areas, for example.
And so executive function-- language and motor ability and spatial ability, I think, are self-evident. But in the case of memory research, what executive function means is basically things that involve making decisions, holding some information in memory and making a decision about it, judging it. Those are executive functions. There's a series of neuropsychological tests that measure that.
We have some biomarkers of Alzheimer's dementia. The best biomarker is a genetic marker. It's the e-4 allel of the APOE genotype. This genotype is involved in catabolizing triglyceride-rich lipoproteins. And about 50% of people over age 75 who are diagnosed with probable Alzheimer's disease carry the e-4 allele. So that's the level of penetrance. Far from perfect, 50-50, but really significant, particularly by comparison to everything else that's available as a biomarker.
The best behavioral marker of Alzheimer's dementia, once again, is our old friend recall. Of all the neuropsychological tests you can give, the best single one is those simple recall tests that we were talking about.
Now let's talk about mild cognitive impairment, the most important form of cognitive impairment without dementia. This is the most common form of cognitive impairment without dementia. It's also the prodromal condition for Alzherimer's dementia. That is to say, if people have been-- OK. So I'm not going to say that people who receive an Alzheimer's diagnosis have normally received a prior mild cognitive impairment diagnosis.
I'm not saying that. That's because people often don't seek treatment until symptoms have become very severe. But for people who ultimately receive an Alzheimer's diagnosis, particularly a probable Alzheimer's diagnosis, who have been seen very regularly, beginning at a time when they still received normal cognitive classifications, they will have normally received a mild cognitive impairment diagnosis before the Alzheimer's diagnosis.
Most people who-- think of it this way. Mild cognitive impairment is the route that you pass through to get Alzheimer's dementia. There's about a 5% to 10% annual conversion rate to mild cognitive impairment among Americans who are over age 70.
There are subtypes of mild cognitive impairment. And there's one of them that's dominant, and it's that one that's the precursor of Alzheimer's dementia. That's called amnestic mild cognitive impairment, as in memory. So no surprise there.
Basically, to be diagnosed with mild cognitive impairment without dementia-- we saw earlier you had clinically significant decline in two of those areas. In the case of mild cognitive impairment, you have clinically significant decline in just one of those areas, but not two or more.
You could have a clinically significant decline in episodic memory. That's amnestic mild cognitive impairment. Or you can have what's called non-amnestic mild cognitive impairment. That's when you have a clinically significant decline in one of those other four areas. About 2/3 of the people who are cognitively impaired without dementia, it's AMCI. It's the memory problem.
This also is a diagnosis by exclusion of diseases, the other kinds of diseases that can produce these impairments-- cardiovascular disease, also, particularly in people that are not demented, depression. Depression is a commonly mistaken in late life for mild cognitive impairment for the simple reason that depression produces declines in episodic memory. And if you treat depression properly, episodic memory comes back. It doesn't come back in true AMCI.
The people who do research in this area have been somewhat surprised for some years, since the e-4 allele of the APOE genotype has a 50% penetrance in probable AD, that it's not a risk factor for this mild cognitive impairment. You'd think it would be.
It turns out that-- we've shown very recently, just a few months ago in an article that we published, that actually it is. There was a large number-- well, not a large number, but a substantial number of studies going back 10 or 15 years that looked at the question of whether or not this particular allele increased your risk of mild cognitive impairment. And it was a very mixed, inconsistent predictor. We went back and looked at those studies and found out that there was a lot of problems with them that I won't go into methodologically.
And we went to a database that didn't have these problems and analyzed those data, and this allele jumped out as a risk factor for mild cognitive impairment. So the story is now very nice and very consistent.
The best behavioral marker-- once again, it's good old recall. Basically, what you have is that performance on these standard little recall tests-- if you have amnestic mild cognitive impairment, the Alzheimer's precursor, you're about a standard deviation and a half below people who are cognitively normal who are exactly your age. That's the primary dividing factor.
So let's talk about what specific things are lost and what specific things are spared, the way we did with healthy older individuals. When you make a transition from normal cognitive functioning to a mild cognitive impairment diagnosis-- we said earlier that this vivid, verbatim basis for recall, it was declining steadily during healthy aging. Basically, what happens is it goes to zero when you make the transition to mild cognitive impairment. So you've lost that as a basis for recall.
But you continue to find sparing of gist-based reconstruction and the ability to judge the quality of your memories. So all of the memory-- rather than most of it, all of the memory, all of the recall becomes inferential and reconstructed when you make that transition.
When the transition from mild cognitive impairment to an Alzheimer's diagnosis is made, what happens is, first of all, the vivid, verbatim basis for recall, it's already gone. You can't lose that. So what happens is you now get declines in your ability to reconstruct the events of your life from gist and to make a judgment about whether those reconstructions are accurate or not.
Now I'm going to show you some pictures of-- yes, I have enough time, I think-- quickly show you some pictures. Here's some pictures of what I just showed you. Here's a big, big, big, big study of cognitive impairment and dementia late in life that was conducted by the National Institutes of Aging. It's called the ADAMS. It's the Aging Demographics and Memory Study.
It has about 900 people in it who have undergone rigorous neuropsychological testing and psychiatric diagnosis. And they're broken into people who are diagnosed as cognitively normal, people who are diagnosed as cognitively impaired without dementia, and people who are diagnosed as demented.
And then within those categories, you have people in the demented category that are diagnosed as, say, Alzheimer's, probable Alzheimer's dementia, possible Alzheimer's dementia, and other dementias, then within the cognitively impaired category, people that are diagnosed as mild cognitively impaired and other kinds of cognitive impairment without dementia.
And what I've done is just broken out the key groups of interest. The people in this study who are-- HC means healthy control, so healthy older adults, and then people of the same age who have been diagnosed with mild cognitive impairment, with probable Alzheimer's dementia, AD1, and with possible Alzheimer's dementia, AD2.
And what these bars are is how good is this vivid, verbatim memory stuff when you're given a recall test. And what you'll see is it's about 18% in healthy controls, and it's basically nothing in the other two groups.
Now let's jump to reconstruction. What you see is a different picture. Now there just isn't any difference between healthy controls and people with mild cognitive impairment of the same age. But there's a substantial drop off with the two forms of Alzheimer's dementia.
Now let's go to your ability to judge, the quality of your reconstructions. What you see is the same different picture-- no difference between healthy controls and people with mild cognitive impairment, but a substantial decline for the two forms of Alzheimer's dementia. Not as big as the decline in reconstruction, but nevertheless, a very substantial decline.
Here's some brain details. Because Alzheimer's dementia is a disease by exclusion-- remember, we can really only confirm the disease post-mortem when we look at the brain. And so in early phases of the MCI AD sequence, what you see are neuronal lesions. You see such things as neurofibrillary tangles. And these are the neurons, instead of being stretched out nicely, are all tangled up. There's a tangle, and you see these plaques, neuritic plaques. There's a plaque.
And these things develop-- remember the parahippocampal gyrus? They develop in an area of the brain first. It's called the entorhinal cortex. And that's in the front part of that parahippocampal gyrus, and then they spread out with time.
And here is a look at a cross-section of a healthy brain and a cross-section of a confirmed AD brain. And you can see here in the medial temporal part of the brain that we looked at earlier that's involved in episodic memory so much, you can see the obvious pathology in those areas.
So we're just about out of time. So a few conclusions. The first conclusion is that by studying episodic memory with recall tests and by breaking recall down into these distinct components, we've been able to produce a smooth, unified picture of memory declines as they occur in healthy aging, and then in transitions to mild cognitive impairment, and transitions to Alzheimer's dementia.
The story in healthy aging is basically a loss of vivid, verbatim recollection as a basis for recall, and everything else being spared. Mild cognitive impairment-- the story is you lose what's left of your vivid, verbatim recall, but you retain your ability to reconstruct and make judgments.
And then when you get to Alzheimer's dementia, basically, nothing is spared. Vivid, verbatim recall has gone on recall tests, and there's very substantial reductions in your ability to reconstruct and make accurate judgments about the quality of what you recall. Yes?
AUDIENCE: Where does the rate of recall come in? For example, you see somebody, and you can't think of their name. You go on to something else, and boom, it pops into your mind. How does that fit into all this?
CHARLES BRAINARD: That's a different-- OK, rate of recall. What we would call that is-- reminiscence is usually what it's called. So this is-- an extreme example of this phenomenon happens to me all the time where my wife and I will be talking about something in the evening, I won't be able to remember it, and [MAKES POOF NOISE].
Better yet, you go to your high school class reunion, and your high school history teacher remembers a question that you missed on a history examination that you got graded down for. And you're at the punch bowl, and your history teacher sneaks up over to you, and he says, Mr Brainard, who was it?
Do you finally remember who it was that said "in the course of a long life, I've often had to eat my words, but I've always found it a wholesome diet." Does anybody know who said that? Who do you think it-- oh, OK, gist. Who do you think it might be?
AUDIENCE: It sounds like Mark Twain.
CHARLES BRAINARD: No, it wasn't Mark Twain.
AUDIENCE: Churchill, maybe?
CHARLES BRAINARD: Bingo. Second one. See? Exactly. It was Churchill. So reminiscence. So what you do is-- oh, my god. Oh, jeez. Who said that? So you say, oh, I remember. I remember. But I've got to do something right now.
And you go over, and you get some punch, and you talk to somebody, and you wait a minute. Boom. It comes to you-- Churchill. You go back and say Churchill. That's reminiscent.
So it turns out that that process is one that's generally spared in aging. Why? Because it's a reconstructive process, just like we saw here. You didn't actually have to know who it was that said it. that sounds like Churchill. So that's a gist-based reconstructive process.
Other questions? I think you were first.
AUDIENCE: You didn't use short term or long-term memory. Could you replace those with episodic, short term?
CHARLES BRAINARD: No. It turns out that the old saw that short term memory is the first to go is-- OK. I'm going to say it's not true, but then I'm going to explain what I mean by that.
When a memory researcher measures short-term memory, they-- a short-term memory task to a memory psychologist is really not what most people mean when they say short-term memory. What most people mean when they say short-term memory and they say short-term memory is the first to go-- they actually mean what, to a memory researcher, is actually long-term memory.
Now, because when you probe examples of what do you mean by that-- you say, oh, my short term memory is going-- what people mean is something like, well, I'm having trouble-- I can't really-- I often can't remember what I watched on TV last night. Or I often can't remember what I had for dinner last night, or where we went to the restaurant a couple of days ago. To a psychologist, that's not short-term memory. That's long-term memory.
When a psychologist measures short-term memory, they use-- how many people know what the digit span task is from the Wexler or the Stanford-Binet? OK. Standard thing on intelligence tests are things that are called digit span tasks. And they're things like-- they say 3, 7, 2, 9, 1. Recall. That's short-term memory.
Another kind of short-term memory test is what's called a Brown Peterson distracter test where we flash triads of symbols at you. It may be sequences of-- well, let's say consonants. We flash them at you. They go away. We give you 10 seconds of fill time. Maybe we play music.
And then we flash one letter on the screen and we say what position was it in. Or we say was it there. Those are short-term memory tests. Those things do not predict, for example, shifts to mild cognitive impairment very well. It's these more longer-term-- so to a psychologist, though not to the lay person, listening to a series of 20 words and then recalling them 30 seconds later-- that's long-term memory. Go ahead.
AUDIENCE: A waitress remembering what somebody ordered is long-term?
CHARLES BRAINARD: Exactly. Exactly. A waitress remembering what somebody ordered-- in other words, not taking full, detailed notes, like they ordered the porterhouse steak, but not putting down I want it cut off the bone, and I don't want it served with the bone, I want just the meat-- that's-- you didn't write that down, and then you get-- that's long-term memory.
You had a question, I think.
AUDIENCE: [INAUDIBLE] All the people today have better memories than back when, for example, [INAUDIBLE].
CHARLES BRAINARD: Yeah. So I can't answer that question scientifically. Because it's like my wife says about-- when they take her medical history and they say-- and they ask about her mom and her dad, and she gives that information. So my wife-- for those of you that know my wife, her grandfather was one of Pancho Villa's 12 lieutenants. And how they wound up here was when they cornered Villa, when the Federales cornered Villa and killed him-- of course, it was kind of hot for the officers, so they got out of Sonora and came to the United States.
So that's a black hole prior to her father. So my wife says things like, they died of a broken heart. That's all we know. So in the case of prior generations, we just don't know. We just don't have good information. So these kinds of very detailed normative studies, like the data I was showing you from the Adams at the end there-- these things just don't exist.
But what you can say is, like we used to say about cigarette smoking and cancer, we don't have good data yet, people would say back in the '50s. But they're all dead, so maybe it's bad. What we can say is that the conditions that experimentally produce better brain health and better memory those conditions are certainly more prevalent today among a larger group of people. And we know more about the things that put off memory decline.
So for example, we know exercise, weight control, don't smoke, don't drink. Why? It all has to do with oxygenating the brain, and in particular, the hippocampal area, which tends to be very sensitive to that. And the hippocampal area tends to be very sensitive to being bathed in glucose for long periods of time.
So we know about that, and we can manipulate those kinds of things. And it should be having effect. But we just don't have the comparisons to say for sure.
Go ahead.
AUDIENCE: A follow-up question, and maybe this is a stupid question.
CHARLES BRAINARD: I have plenty of stupid answers.
AUDIENCE: I have contact with a lot of my colleagues, and some people I've know-- I know a lot of people because of [INAUDIBLE] Cornell. But we sit down for coffee sometime, and someone will say, do you remember so and so?
What was his name? Remember the guy that did that? And somebody will say Jack, and the other guy will say Jones. My wife and I do that all the time. And sometimes, it takes you a while to remember Jack Jones if no one else can. What kind of memory is that?
CHARLES BRAINARD: That's a good question. This is why you like to give lectures at Cornell. People have great questions.
That is actually a form of reconstruction, but I've got to explain that. This may surprise you, but there's a huge amount of basic memory research that shows that when you remember an event, you can actually remember a part of the event, not the whole thing. An early example of this is-- has anybody ever heard of a form of memory research called the tip of the tongue phenomenon?
AUDIENCE: Yeah.
CHARLES BRAINARD: Yeah, OK. It was extensively studied, originally starting back in the 1960s, where people said, you know, I can sort of feel something on the end of my tongue, but I can't say it. So people got in and studied it.
And what they found out was that the reason for that experience was this thing that we're now talking about-- that people could actually-- what that experience was being caused by was that people were remembering a part of what the word was-- the [MAKES "J" SOUND] in Jack, for example. So you can actually remember, in a holistic experience, the word onomatopoeia if it was on word list. Without remembering that word, you can actually recall a part of it.
And if you can recall a part of it, it helps you retrieve-- it clicks in processes that will help you cover the rest of it. So that's what it is. It's reconstructed memory using this amazing ability that we have to not remember something, but actually remember a physical part of it, and regenerate from that.
AUDIENCE: I use that to go through the alphabet. If I can't remember something, I'll go A, E.
CHARLES BRAINARD: Yeah, OK. How many times have you sat at home with your wife watching Jeopardy, and-- I know that. It begins with a K. I can't tell you. That's an example of this.
AUDIENCE: Where does and how do brain exercises, such as chess or [INAUDIBLE] fit into--
CHARLES BRAINARD: OK. So let me cut right to the chase and answer this question with are brain exercise programs any good for forestalling mild cognitive impairment or forestalling dementia transitions in people who are already mild cognitively impaired.
Now, there are-- if these areas of the brain, for example, are really important in episodic memory, and if it's episodic memory that's the problem that we're losing, then in principle-- now, this doesn't always work in medicine. As my wife likes to say, if you have a headache, you'd think you just bore a hole in your head, and-- well, we know that doesn't work.
So these physical analogies are not always good. But in principle, if you have people perform tasks that you know, on the basis of imaging research, will fire up these areas-- OK. Maybe it's like this, and it will be helpful. There's actually some positive data on this. This is good news.
So my colleagues at Mayo have developed a-- and I'm so sorry there's nobody here from-- actually, I am glad, now that I think of it, that there's nobody here from Kendall. Because they'll tell you what a torture this is. But there are these kinds of programs.
And there's actually one, for sure, that's called Posit Science. I'm not plugging it. I have nothing to do with it. But in the scientific literature, it's the one that I know of that there's actually positive evidence on it.
Now, if you ever do this thing, you'd think, oh, god, I hope it has a positive effect, because this is a lot of work. You do it literally for an hour or an hour and a half a day for six weeks. And it involves taking a CD, and putting it in your PC, and doing all kinds of complicated search things. Something pops up over here, and you press a button down there.
And basically, what it does is we know that the kinds of tasks that are up there on the screens, or ones when we image the brain when people are doing them, they're heating up those areas. And it actually does have a-- there's randomized controlled trials with it that actually shows that it does have a significant effect in the sense that it reduces normal memory declines.
So you do a randomized control comparison, and people that have done this six-week program that are normal and healthy, for example, will show less memory decline a year later than people that haven't done the program, and so on. So there is some evidence.
But, but, but, don't-- just like everything else, don't go for something for which the evidence is, well, they've done it in Germany for 1,000 years. So there's a lot of things out there of this ilk for which-- well, most of it for which there's no evidence. So you have to find the stuff for which there is evidence.
NIH website, NIH website, NIH website. Whenever you have a question about is there such program, are there randomized controlled data, does it work, just plug it into the NIH website and see what they have say about it.
Yes, I'm sorry. Go ahead.
AUDIENCE: [INAUDIBLE] trying to remember a word. I couldn't remember what letter it started with, I go through the alphabet. And at some point, I lost that trick. It pretty much barely worked anymore. But I developed another way to search for the word. I would hear how many syllables it had and where the accents were on the syllables. You've heard of that before?
CHARLES BRAINARD: Yeah.
AUDIENCE: Do I have brain damage?
CHARLES BRAINARD: No. Because although you think you're talking about exactly the same thing as he was talking about, you're not. What he was talking about was a case where you're trying to remember something, and Jack comes back to you.
What you're talking about is a systematic retrieval cuing process where you take a letter, and another letter, and another letter, and another letter, and see if you can retrieve the word based on it. That's a different deal. There's a principle in memory that's called the principle of encoding specificity. And it says-- what you're trying to do when you do that is to get something that matches up with the memory that will fire it, make it ring like a tuning fork.
The problem with that technique, and the reason why it tends to not-- it becomes progressively less useful in all of us with age is that's a technique that matches up with verbatim memory traces. In this verbatim memory stuff that I was talking about, what we do is we represent our experience in two kinds of ways. We store verbatim traces of the actual physical form of the experience and its contextual details, so the representational equivalent of pictures and sounds. And we also store gist memories of the meaning content, our understanding of the meaning of our experience.
The former stuff tends to go away with age. So it's harder and harder to cue that with something that matches it. So long story short-- to cue a verbatim memory with a surface feature of that memory tends to go away. But to cue a bird-- oriel, robin-- with the word "bird" doesn't go away. So that's why that declines with age. It's something that matches a verbatim representation, which is going away with age. There's nothing there to access anymore. Yes?
AUDIENCE: They'll say sometimes, well, so-and-so can remember everything from the past, but she doesn't know what she had for breakfast this morning. Can you explain that?
CHARLES BRAINARD: Yeah, high school was a lot of fun. The rest was all responsibility.
No, it's a very common-- so it's a very common phenomenon as people talk about it. It's actually been fairly-- I shouldn't say this. It's actually been fairly extensively studied in recent years. But the research isn't particularly good, because you can imagine how hard it is to actually study this. Because first of all, you have to find out, here's the bad part. When people say that, the presumption is that those memories are accurate. And they ain't so accurate.
My first memory in life-- typically, most people's first memory in life-- one single, first memory, not a lot of them-- will be sometime after age 2 and 1/2. So I have this vivid memory of when I was 2 and 1/2 when my baby brother came home from the hospital after my mom had him. And we're all sitting in the den, and there's my father, and there's my maternal grandmother and grandfather.
There's my mother's younger sister and her husband all sitting here. And it's very vivid. And so I would say I can remember this much better than who was there when my daughter was born. The problem is I've got this picture of my uncle, whom I remember sitting on the couch, on that very December the 17th, sitting on a tank in Germany. So I don't think it's accurate.
So the gist of the research in this area is that it is a real phenomenon, but it's nowhere near as big as people think it is. There's a little bit of-- in that direction.
AUDIENCE: Is the memory-- how far back it can go, at 2 and 1/2, can most people go back that far?
CHARLES BRAINARD: People-- I should word it in a little bit different way. There's a phenomenon that's in the memory literature that's called infantile amnesia. And that's the phenomenon of we don't tend to have memories of our very early life. So then the question arises-- how early can we actually have memories that we'll carry around for the rest of our lives?
And for most people, it's somewhere in the third year of life. So 2 and 1/2, say, as a mode. So that's really all I'm saying.
If you-- now, a lot of people-- perhaps the majority of people-- don't have a first memory that's that early. Their first memory may be their third grade teacher. But that seems to be about the cutting off point for the earliest memories for people that have very, very, very early memories.
AUDIENCE: [INAUDIBLE] one more question, and then we probably should wrap it up.
CHARLES BRAINARD: OK.
AUDIENCE: I think you've mentioned that the old tend including someone who has Alzheimer's [INAUDIBLE]--
CHARLES BRAINARD: Post-mortem confirmation. Probable is the strongest diagnosis.
AUDIENCE: So is there any research going on at this juncture, either with the blood test or MRI or other kinds of a search process of the body, that would enable physicians to determine if, in fact, there's ongoing Alzheimer's?
CHARLES BRAINARD: Yes. There's a number of things that people are doing with brain scans, with cerebral spinal fluid. The problem is that they're all not very reliable predictors. So the problem is that you can have-- when you do post-mortem work, and if you have somebody that is diagnosed as probable AD in life, and then you do a post-mortem brain examination, and you find the kind of pathology that I talked about-- then you've got a confirmation. Oh, yes, that was Alzheimer's dementia.
But notice what I'm not saying. What happens when you look at a brain post-mortem of a normal person, and you see that kind of pathology? You do, all the time. So the magic bullet we don't have yet as far as sensitivity and specificity in diagnosis.
We've got it from a neuropsychological point of view. We've got the symptoms. We can measure them. We can quantify them. But getting the biomarkers is another matter.
Thank you all so much. Thank you.
[APPLAUSE]
Charles Brainerd, chair of Cornell's Department of Human Development and an expert in the scientific study of human cognition, discussed the effects of normal aging and diseases of late adulthood on memory, March 1, 2012 as part of the Cornell Faculty Career and Life Speaker Series.
Brainerd contends that declines in memory for life events are a normal part of healthy aging; that memory declines are not across-the-board; that these declines are concentrated within a narrow set of abilities that is often called "verbatim" memory; and that some of the most important memory abilities are spared even for people in their eighties and nineties. More extensive memory declines, he says, are the main diagnostic symptoms of the two most common forms of cognitive impairment in older adults: mild cognitive impairment and Alzheimer's dementia.
