 |
||||||
|
|
|
 |
 |
 |
|
2002 LECTURE SERIESEducating the Brain: Lessons from Brain ImagingDr. John Gabrieli Thanks very much. I'd like you think for a moment about everything that you know. The values you hold. The memories of your life. The skills that you have. Languages you might know. And think for a moment where that knowledge comes from. And it can only really come from two places. Everything that you know, you believe you can do, can only come from your genes or from your experience. And in either case, it has to be mediated to the brain. Either the genes speaking through the brain to contribute to who you are and what you do. Or your brain recording, retaining, and retrieving your experiences and letting you do things with that knowledge. And I'd like to share with you today some of the work we've been doing. Thinking about how the brain is educated through its lifetime, and lessons from brain imaging as we use modern technologies to see thoughts and feelings flicker across the neural substrates of our mind. This is an ancient coin of the Roman God Janess, January, looking for, and he had the property that he could look forward and backward. And I bring that up, because, in a way, we often think of memory as reflecting on the past and trying to remember things yesterday that we want to remember from yesterday. But, in fact, memories are of the past, but it's for the future. And it's for the future because what we use memory for, and the knowledge we gain and skills we gain, and values we gain, is to become more competent and efficient in accomplishing what we wish to accomplish. In our jobs, in our relations with people and helping our communities. And so, I want to think about that memory in our brains as a way in which the past constantly guides the future and hopefully towards a happier and more successful future. Now how does the brain do this? We're all stunned by this. Even at the Center for Neurological Learning and Memory, one of the absolute world centers of this, it hasn't been quite solved yet. And many of us feel like children of the seashore, to use a, a phrase, because we know there's something huge and remarkable about the way the brain gets knowledge, keeps knowledge, and uses knowledge. And we just have a few things we know about it compared to many years ago. But, there's a lot we don't understand yet. Now what we would like to do is to look inside the brain in some way and we can't really take this approach, of simply removing the top of the skull and poking around for memory. Nor, can we take this approach because the number of participating for this approach [laughing] are understandably few and are, are is far beyond and below our ethical standards. So, you know, that's of course what we'd like to do. We'd like to somehow look in the brain and see the formation of a memory. See the use of a memory. See the growth of a memory. See the retrieval of a memory. And see how does that work and how does our brains endow with the powers of, of knowledge. And I'm going to talk to you today primarily about one kind of brain imaging and I'd like to make one distinction that, between structure and function. I'll show you a little bit of structure, but most of the images I'm going to show you today are function. Images that reflect something about the brain in action. Thoughts, feelings, actions occurring, and seeing which parts of the brain seem to be supporting our abilities to accomplish those things. And our brain imaging technique is not based on what we wish. We know that the brain is comprised of neurons and that neurons compute our actions, thoughts, feelings, and so on. But, we can't measure those from the outside very well. In animal research, people can do remarkably brilliant measures. In human research we can't go inside the healthy brain except under the rarest of circumstances. And so we take advantage not of talking with the neurons that compute the mind, but with the gossipy neighbors of the neurons, the vasculature that flows around the neurons and supplies it with nutrients. And so it is, it's this vascular flow that mostly we take advantage of, it's in a form of imaging I'm going to describe to you in a moment, functional magnetic resonance imaging. Using magnetic resonance imaging to show mental functions as they occur in the brain. And here's how it works. And they have an excellent title for this called the BOLD response. I'm glad they came up with this title because it's good to work in a field with bold research, right? But, it actually stands for blood, oxygen, level, dependency. And I'm just going to show you, just for a moment, how we get this measure. Because in one sense it's straightforward. In another sense it's kind of a remarkable gift from nature. For those of us who'd want to understand how the brain supports the mind, it's a, it's a phenomenally difficult puzzle and it seems to constantly sort of confuse us with its mystery and power. And so, every once in awhile it's like we get a helping hand from nature to give us one more shot of doing better, for letting our minds understand our brains. And such a BOLD response works like this. If there's a part of the brain where there's an increase in neural activity, because you're using that part of the brain as a specific part of the mind. For example, you're looking at something, you'd use a visual part of the brain. You're feeling an emotion, you'd use emotion parts of the brain. When that's active, because that part of the mind is being recruited, then something happens. Which is this increased blood flow to that part of the brain to supply it with extra nutrients and support because it's a heavily active area. Then the brain does something very intriguing. What it does is it sends a lot of extra blood to that part of the brain that's very active. As if it were concerned that without having a good full supply the mind might not accomplish what it ought to accomplish. And it's this extra blood supply that's shunted towards active parts of the brain, mediating what we'll presume are active parts of the mind, that we're able to measure ultimately as an increase in MR value signals. So the signals that we see are changes in blood surrounding neurons that are active and we try to call those neurons into activity by asking people to do very specific things. And I'm going to talk to you today about two structures at the beginning of the topic we are terribly important for memory: the hippocampus and the amygdala. We have two of each, one on the left and one on the right. And they're on the insides of the temporal lobes. The temporal lobes are the parts of the brain that come down near your ears that are on the insides on both the left and the right. And this is a view of a hippocampus top of the brain of a post-mortem view. And here's the hippocampus itself on the insides here a person's ear would be sitting something like here. And we know that these structures are terribly important for memory. Most of all from arguably the most neurological case of the last century--I'll describe him to you for a moment and I'd be glad to talk in the question period about him as well. When I was a graduate student most of my research was with this man. Up until the early 1950s people had big debates about how memory is organized in the brain. One view was that memory was very distributed. Imagine you had a glass of water and you put in a few drops of milk and really swirled it. That's how memory was viewed. Distributed widely over the entire central nervous system. And one clinical case kind of reversed the view that all of the brain does all of memory. And that's this clinical case of H.M. So, he's a man who began having epileptic seizures at age 16, and by age 27 he was having many seizures a day. In fact, hundreds of seizures a day. The medications at that time were not really effective for controlling epilepsy. And in order to treat his epilepsy they performed neurosurgery. They couldn't find where his seizures were starting from. This is a kind of a surgery which is still performed very effectively for selected cases of epilepsy. There are patients who have terribly frequent seizures, are really captured by the frequency and severity of their seizures. But, with this surgery, removing the part of the brain where the seizures start, they can leave the hospital never having a seizure again. It can be remarkable for the right kind of patient. And in 1953 they saw this man having as many seizures, but they could not localize where the seizures were starting from in the brain. And in order to be safe, they removed bilaterally, on both sides of the brain. Something that you see here in a cartoon, namely, the hippocampus, the amygdala and surrounding tissue. In terms of treatment for epilepsy this surgery was quite successful. Oh, the reason they picked that area, let me say, is that the majority of cases of epilepsy, not all of them, have seizures that start in this part of the brain. As if this part of the brain were doing such a dangerous high wire metabolic act, something such a [inaudible] of what neural systems can do for us that they're very prone to disease. Not only epilepsy, but anoxia, if you don't have oxygen. This is part of the brain that's affected very early. Alzheimer's disease. In the vase majority of cases, it starts in this brain region. This is a terribly important brain region, but a very vulnerable one in our brains. And what they discovered after they removed these structures bilaterally, was that he retained his intelligence, his perceptual abilities, his motor abilities. His knowledge of the past was largely normal, but from that day forward to the present day, this man has been unable to remember an event in his life, or a fact that he learns, for more than a few seconds. He has no idea that the years have passed. If you ask him the year, he'll say 1955. If you ask him his age, he'll say something like 28. If you ask him--he doesn't know for example, that his parents have passed away. He has no idea that we went from 1999 to 2000 in an exciting change of the millennium. He has no idea about what September 11th means to us in this nation and around the world. Every fact and event, no matter how often repeated, no matter how personally salient, no matter how often he watches it on T.V. he forgets seconds after that event has passed. Or, if he hears a fact, reads about a fact, he forgets that in seconds as well. So, here's a man of normal intelligence and normal every kind of ability, but his ability to learn, remember any new event or learn any new fact since 1953 is essentially zero. And this demonstrated very clearly that this part of the brain is essential for us to remember the events and facts of our lives. Now those are very huge kinds of memories. All of the events we have, every experience we have and every fact that we know and learn about are unavailable to him since 1953. And that was a cartoon, but the many years later they took an MR scan and you can barely see it almost, because it's such a small structure. Here's the top of the brain, left and right. Here would be the ear in a healthy person. Here would be this hippocampus that we require to remember events and facts. And that structure was removed in him from the surgery. Now, many of us remember terribly important things all the time. But a huge element of memory is there's a lot of things we don't remember, day to day. And we know that's frustrating when we try to think of something, where the car keys are, where we parked our car, or, you know, the shopping list if we didn't bring it with us. But, I just, I want to demonstrate to you how selective memory is, even in the healthy mind and brain. So, you can make an estimate of how many times in your life you've dealt with a Lincoln penny and depending on how old you are, it's a huge number. Even if you're a high school student, it's a huge number. Every time you've had that penny in your hand, in your pocket, in your pocket book, on your bureau, right? How many times you get that four cents. You know, that alone would account for the irritation where you're jangling change in your pocket. But now, ask yourself whether you'd be willing to bet something big on the following information. Which way does a Lincoln penny face? Left or right? Okay. [laughing]. Is anything above his head? Okay. Is anything below the head? All right. Uh, is anything to left or to the right? And you might think that for the thousands of times that you've handled a penny and seen a penny, thousands and maybe tens of thousands, if you're a little bit older, how absolutely certain are you about these answers? If somebody were to tell you, I'm going to show you something very simple ten thousand times, you'd say I've a pretty good shot, right? I mean, and yet, most people are not really confident they know every aspect of this answer. And in fact, if you get multiple choice [laughing], you're not, you know, even sure. And here's the answer, just so nobody leaves here tortured or uncertain. Okay. Here's the point. We experience so many things every second of our waking days that we remember a very limited number. But, what we remember is terribly important, because what we remember in a sense is who we are. You, what we learned about life, the skills we have, who we value, what we do, you know, our political values, our religious values. We learn about those things so what we remember in many ways becomes who we are. And yet, there's lots of things we experience that we don't remember. So, we want to perform an experiment, but we want to look at something about the brain basis of the hippocampus and, and nearby structures. Giving birth to new memories. Selecting what experiences are going to be remembered and what experiences will pass through our minds and brains, but not be remembered, but be forgotten very soon. And so we have people see while we were imaging their brains, pictures like these. And they had to simply judge whether they were indoors or outdoors. So, that would be indoors. And that would be indoors. And that would outdoors. And that would be outdoors. And that's all they were doing. Then we did something like this. We took pictures of their brains. They later got a memory test. And then we went back to the original pictures, each picture and I'll show you in a moment, for each individual seen and asked what's happening in the brain as you see something like that's determining what will be remembered and what will be forgotten. And so, the idea again, is you're seeing a number of these pictures and for each of these pictures an individual brain response is collected. And now in particular we're focusing on the medial temporal lobe that we know is essential for memory in a patient like H.M. And they would see 96 of these scenes, come out of the scanner and get a surprise memory test. We did not tell them there would be a memory test. And they would see then those pictures they saw before and 32 ones they did not see before and their job was to say for each picture, did you see it before? And if you think you saw it before--if you say yes, I saw it--do you remember it very clearly and vividly and powerfully? Or, do you feel like you saw it, but you can't say much more than that? Now, we know for the 96 pictures they saw them, they classified them as indoor or outdoor, so there can be three mnemonic destinies or fates for experience. It can be powerfully remembered, modestly remembered, or doomed to be forgotten moments after it has passed through your mind. And we went for parts of the brain that had the property that they were most active for things that were well remembered, intermediately for things that were modestly remembered and least active for things that were doomed to be forgotten minutes later. And what we found was this, so this is six, average of six subjects, top of the brain, left and right of the brain. Here's the activity in one, two, three, four, five, six spots in the parahippocampal area, the spot very near the hippocampus. All have the property that the more active they were, the more certain you were to remember something, the less active they were, as you saw it, the more certain you were to forget it. So, we can see in this part of the brain already what we might call the making of memories or component of that, or the birth of memories. Processes, as they're experiencing something start to build memories that will become lasting memories or if they're not engaged powerfully then experience will float away as a ghost into the mists. And in fact, here's an example of a sort of biological squiggly signal. You know, here's the response for things that will be powerfully remembered and in blue the things that are going to be forgotten within minutes. And so we know this part of the brain that is keyed or tuned to selecting certain things that seem for some reason to be memorable and for other items deciding it's, it's not important and not going to be remembered. And so this a mnemonic slide, which is that if you have a beautiful memorable sunset in your life sometime you will remember it to the extent that the parahippocampal cortex is active. Now, many people think that takes the romance out of a, almost everything. And it's true that this is a, this story, you know, warms a neurobiologist's heart. It's not a full story of human experience and why we remember things. But, it's an apparatus without which we know you cannot remember anything. In the case of H.M. even that your parents have passed away or the most public and tragic ill events pass through your mind without any chance to record them. So, it's a remarkably powerful apparatus for selecting experiences from the present to be kept for the future and to, to sort of guide our future behavior and values. So, other kind of memory deals with traumatic events and this is a shared public traumatic events from last September. And we know that a part of the brain is terribly important for the formation of memories that are tinged with memory is the amygdala and Professor McGaugh is one of the absolute world leaders in many of the critical discoveries in how the amygdala enhances memory on the basis of arousal and emotion. The amygdala also removed in the patient H.M. so it's a little bit anterior or in front of the hippocampus. Now, we had participants view pictures again. And as they viewed them, they ranged from neutral to highly negative and their job was to tell us, as they viewed each picture, how emotionally intense they found them. Now, this is a alerted audience so when you see a book like this your heart is pounding, your palms are sweating. [laughing]. But, undergraduates at Stanford, for reasons that none of can understand, rate this picture as kind of ho-hum neutral in terms of emotional intensity. I mean, all right. An important thing of this is, of course, different pictures do evoke different feelings in different people depending on your prior experience and your current feelings. Here's a picture that most people rate as more intense. And I'm going to show you now a picture that people rate as very unpleasantly intense. If you don't want to see it, don't look for just a moment. Okay? It's very unpleasant. I'll tell you that. So, don't look for a moment. For those of you who're looking, everybody rates this as terribly unpleasant to look at. Okay. All right. Everybody can look again. So what we did now is we had people look at this range of neutral to highly negative scenes and again, we recorded a response for each scene and in collaboration with Larry Cahill, a member of the faculty here, we did a study with [inaudible] we wanted to see what's happening in the amygdala as we see these things and how's it contributing to memory formation? And we, we looked sort of watching people's brains as they were looking at these experiences, and feeling them. And here's the first thing we found. That in the amygdala, in this case on the left, we could track brain activity that went with how you felt. So, for pictures that were rated as most emotionally intense you got the biggest amygdala response. Less intense, less intense, neutral. So, on average, what we're, what we're watching in this sort of fascinating way is a person's feelings in part, you know, a part of their feelings in their brains. And the more intense they feel about something the more the amygdala is activated or turned on. And then it goes down as they feel less active. Now, when we test their memory for this material three weeks later, it turns out that the only memory, the material that they remember quite well is the material they rated as most intense. Everything else is pretty similar. But, this category they said, wow, that's super intense, like the picture I showed you just a moment ago. That's the category they remember pretty well three weeks later when we tested them. And what we found is that, for that intense material the greater the activation in the amygdala, the more likely you were, you were to remember it. So unlike the hippocampal area, which seems to apply for all kinds of knowledge, the amygdala seems to be specifically tuned to the way that memory uses emotion or emotional intensity to heighten the probability you will, you will remember something. And so now we know something about H.M.'s problems because we can see that these structures have a tremendous effect in making memories or giving birth to memories as we experience things and in determining what in our present will be with us in the future. Now, the kind of memory I have spoken to you about so far is what people call declarative memory. Memory for events or facts. A huge aspect of, of memory and without that ability, like with H.M., you can't hold a job, you can't form human relations, you can't really do almost anything that we consider worthwhile in everyday activities. But, there's another kind of memory altogether that does not depend upon these brain regions that people have called procedural or sometimes implicit memory. And we can see that particularly in skills. And skills are like a wonderful thing. You know, if we practice something hard because we want to get good at it, our brains let us get better at it. And it turns out that many skills don't depend on simple memorization. I mean, it's frustrating for us. When we want to get good at something, we can't just sit down and memorize a manual. We have to learn by doing. We have to practice, practice, practice to get better at those things. And so, one example that you can do in a laboratory is have people read these mirror reversed texts. So everybody has the first word down, maybe. Elephant. No. [laughing]. It's, it's kind of hard, you know, especially when a P looks like it could be a G or a Q or a--the first one's platinum. The second one's charity. The last one's houseboat. Now, at first it's pretty hard to do this because this is weird way of looking at letters. And in fact, some of them actually go against the way in which we normally look at letters. But if you practiced this for awhile, you would get considerably better at it, if you did this day in and day out, for whatever reason you might do this, whatever psychologist draws you into doing it. And we know the following remarkable thing and I'll just describe this to you from the work in San Diego of, from Larry Squire and Neil Cohen, which is that if you practice this day in and day out and they measure how quickly you do this, how well you do this, what kind of skill you get, people get faster and faster. And the remarkable thing is that patient's with a kind of terrible memory problem like H.M., with this kind of global amnesia, an inability to remember any new fact or event, they learn this skill perfectly normally. They come in the next day, they have no idea they did it the day before. You ask them what words they look at, you probably could remember the word platinum. They couldn't tell you any of the words they saw before. But, when they sit down and start doing it from day one to day two, their performance picks up right where it left off. So, here's an unconscious form of learning in the sense that the, amnesic patients don't know they've gotten good at this. I mean, if you talk to a healthy normal person they tell you a lot. At first it wasn't, at first it was hard, but then I got better and I got better because my family has raised me the right way and I have the right kind of work ethic. And I got really good flipping letters, and you know, and the secret to my success well, you know. I mean people tell you all kinds of things about why they got better and how proud they are they got better and you know. Like, it's less sense of accomplishment in a limited way. The amnesic patient comes in and has no sense that he or she has ever done this task. Can't tell you any of the words. You have to give them the instructions. They have no idea what to do. But, then, obviously somewhere in the brain they recorded, retained and now retrieve that kind of knowledge perfectly normally. It's a different part of the brain and this comes, contributes to our view that the brain is sort of like a symphony orchestra comprised of many instruments. The different instruments make different sounds. And within the human brain we have different kinds of memory instruments that are specialized for different forms of knowledge and expertise. Now, we can do one more thing. We can say, well, the amnesic patients learned that. But, how do we learn a skill like that? What goes on in our brain as we get skills through practice and effort? How does our brain let us get good at things? And what we did is a sort of before and after brain imaging study. Where we had people, healthy people come in and we looked in their brain as they were looking at mirror reversed text. They weren't doing very well. They were making mistakes. It was pretty slow. Then they came into our laboratory and practiced, practiced, practiced, for several days until they got pretty good. And then we put them back in the scanner and asked how are they doing now. And here's what we found. This is the practice, you know where imaged them before and after. This is them getting better and better and doing the task. And here's what we see in the brain. And I'll come back to this in a moment. This is the back of the brain. And these arrows point to some big changes that occurred from when people were doing it not so well, compared to when they were doing it very well. And the big changes were that in the right parietal cortex--and I'll explain a little bit why, what we think is going on--there was activity that decreased as you got better and in the left temporal cortex here in blue, there is activity that increased as you got better. And in fact, this kind of pattern, a choreography of different parts of the brain first trying to do their best and then with skill and experience other parts of the brain taking over and doing a better job, turns out imaging has showed us that we never imagined. I mean, intuitively we might have thought something, we could have made up a story. But to literally see parts of the brain initially be very important for doing something and then trade off with other parts of the brain to get good at it. A sort of, you know, choreography, which parts of the brain are mediating performance, is something we can see dynamically changing by brain imaging. So, what do we think is going on? Well, we know a couple of things about the, brain organization in, in humans in big terms. One thing we know is the left hemisphere is specialized for language in the vast majority of people and the right hemisphere for visual/spatial things in the vast majority of people. Another thing we know in humans and, and from animal research as well, is that visual information first enters the cerebral cortex here, and then goes on two different pathways. Visual information that's important for what something is, identifying objects, words, and things like that, tends to go down the temporal lobe. Information for spatial thinking and physical action tends to be extracted and go up in a [inaudible] system through the parietal cortex. So knowing these two things about which there's wide agreement in neuro-sciences, we can now make a story about what we think is happening as the brain is changed by experience and we get to watch that change unfold like a flower. Here's an individual human being, top of the brain and bottom of the brain. His visual cortex. His parietal cortex. These are two different slices from that person. The first time when they were not good at doing a task. And here, and here they turn on right parietal cortex. You can't be in a more spatial part of the brain. The right spatial hemisphere the superior pathway for spatial cognition. So, what we think is happening in the beginning is, it's a spatial problem. What is that letter? How will I turn it around? Let me double check that letter. It's a spatial problem to get the letter turned around in the correct way in your mind's eye. After you get really good at that, it's not a spatial problem anymore. Here's the same person, same brain, when they're really good at it a second time during the scanner. Now, they've gotten quite comfortable with what those letters really are and they can map them into the left verbal hemisphere, into the what pathway that identifies words. They can map them onto their normal use of words in language in the left hemisphere. So a shift from a, you know, desperate spatial strategy, right? To a really competent efficient left hemisphere temporal pathway. And so we can begin to understand something about how humans gain abilities through practice. As we can see different parts of the brain come on, to do their best at the beginning. Leave this stage and other parts come on, so to speak, to take over for competent skilled sophisticated performance. Now, what underlies this brain activities must be changes in the way that neurons are hooked up with one another. There is a physical change that must be occurring in the brain. Some physical change at least that's occurring here, that now allows the information to be funneled to that part of the brain for thought and action. So, even though we're seeing this blood change, it must be reflecting physical changes induced by this experience. And we think this is a model really for practically every kind of skill and ability that we gain through, uh, education or practice or work related things. Depending on the kind of skill, it will involve different parts of the brain. Now, you know, in a way what we've spoken about here is, a, a toy version, if you will, a limited version, a laboratory version, of learning how to look at text. What I want to do is play you some tones for a moment and have you think about what they're related to. What the next experiment is going to be about and you'll guess very easily. Maybe. [tones] Okay. So, what's the next experiment going to be about. [laughing]. First, of all you think, man, is this going to be boring. I mean, how interesting can this possibly turn out to be? But, I'm going to try to make the case in the next 15 minutes that the difference between that sound and that slightly faster sound is fundamental to how humans have language and why other species on this planet don't have books. And you know, I'll have to convince you of that. Right? I've set a high threshold. Okay. The difference between that [sound] and the [sound] is the difference between books and libraries and spoken language and all the other species on the planet that don't do that. So, let me try to convince you. We know that visual language is only about 4,000 years old and visual language is a very funny thing, if you think about it. There's no way that our brain evolved to be readers. Because auditory language has been around for a long time. But, visual language is something that came onto the scene relatively recently, in terms of evolution. So, our brains were made to hear language and to speak. Our brains were not designed to read. It's a brilliant, amazing thing that we can read and the power of text. And of course, once we had moveable type and the Guttenberg Bible and things, and all of sudden, this became a flooded society and we now know that reading is a path to empowerment in our society. A poor reader will struggle. A good reader will have many educational opportunities. Now, how do we do reading in our society? Think why, think about school for a moment in the following sense. For all practical purposes, when we send children off to learn to read, it's like sending them to Marine boot camp for like four years, right? [laughing]. And our entire country is hoping that everybody learns how to read. Parents worry a lot--their child succeeds at learning how to read, it's so essential for doing well at so many things at school. It's so unnatural that we have to have a school system organized for the multi-year indoctrination program. For their own good, right? Nobody has a school that says, everybody get together we're all going to learn how to hear words, right? I mean, you know, everybody just talk to your infant, talk to your kid, under most circumstances. Reading, it's like, oh, boy, you know, here comes several years of effort and organized training to hopefully get a child to understand what a printed word is about. And part of the reason I think it's so difficult to get reading in terms of, in terms of education is, it's not what our brains were ever geared to do. It's a remarkable adaptation, but not something that's natural for our brains and our minds. And one surprise I wanted to share with you, those tones and how they relate to reading, visual reading. But another surprise may be for the field, some years ago was this. Oops did I? It's all right. Sorry. That, orthography is a word that we use in psychology for the appearance of words, the look of text. Phonology is the sounds of words. Reading, it turns out, is a process by which we take the phonology that we know, the sounds of language that we learn at home with our families and with our friends in, in early school, and then somehow turn on top of that, build on top of that our knowledge of visual language. We all start with auditory language and our visual language builds on top of that. And people call this phonemic awareness. A sense that the printed text on the page is about the sounds of language that we know and easily pick up under most circumstances in daily life at home. And English is extra tough about the rules of phonology. Many people know this, compared to other languages, but now I want you to tell me how you would read that word. [inaudible] Oh, man, has this been done already here? [laughing]. No, I would say goatee. I don't know what you get. [inaudible] Sorry. [inaudible] All right. If you--has this been done in this lecture series? Or is this just like, is this something that people in this community just like know? [laughing]. I assure this amazes Stanford undergraduates when I--all right, the reason it might be fish, um, is because the "GH" could be pronounced as in tough, the "O" as in women, the "TI" as in nation. All right. Now, English is extra tough also then in the sense that compared to a very regular language, what is a regular relationship between what you see and what they sound like, like Italian--and fairly recently there was an imaging study that compared native English speakers and native Italian speakers reading nonsense words. Now a nonsense word is just like goatee or fish. You just read it as you see it. But it turns out that when you compared the Italians and the English speakers, what they found was more brain activation, this much more brain activation in the English speakers to read the same nonsense words. Because, English is so cluttered with exceptions and irregularities. [laughing]. That even when we see a nonsense word it brings to mind all the possible sort of traps of how it could be pronounced. So, although it is pretty clear that reading difficulties around the world are very similar in nature, some languages bring them out more easily and are more difficult to learn inherently and English is one of the absolute worst actually in that regard. Okay. So, here are children learning how to draw letters and things like that. Developmental Dyslexia refers to an unexplained difficulty in reading. That is, a child who is motivated with the perceptual abilities to do well, with the education opportunity to do well, the home environmental to do well, but struggles at acquiring the ability to read the text. And depending on how you define it, it affects a surprisingly large percentage of our population, from five to ten percent is an estimate. And again, I think it's so common because our minds were not made to be, to be readers at all. It's a remarkable, cultural, you know, tour de force, but one that's tough for brains. And in the study with Elise Temple what we did is we looked at children now who are age 10-years-old who are either normal reading--average age of 10-years-old or who were having troubles in learning how to read, who were dyslexic by clinical criterion. And so there, there are in the 8, 10-year-old age range. And one of the exciting things about imaging of the kind I'm telling you about tonight, I think, is that because it looks at naturally occurring changes in blood activity in the brain, we can apply it to a tremendous age range. The youngest we've tested so far is seven. We also do studies of aging and Alzheimer's Disease, tested people up to 90. And these children have good, non-verbal IQs, but when it comes to the reading scores these dyslexic children are struggling. And we had them do a very simple task in the scanner. We had them look at pairs of letters. I'll just say about these two things. In this case, they just decided whether the letters were the same or different. They didn't have to sound them out. So a "d" and "m" are different. They look different. You don't even have to know English letters or what we know to know that. And a "p" and "p" are the same. We compared that task to another one where children saw pairs of letters like these and have to decide whether they rhymed. So "t" and "d" share the same ending sound. Those would rhyme. "G" and "K" don't have the same ending sound. They would not rhyme. So we were forcing them or, or asking them to sound out the letters, this phonemic awareness. What do these letters sound like to make these judgments? Still, a very simple task, even for a 10-year-old, a very simple task. And here's what we found. These are 10-year-olds. Now, I'm showing the front of the head, the back of the head, and here's the left hemisphere. And in the normal reading of 10-year-olds, what we see is activity in the anterior area near what's called Broca's area and posteriorily in what people call Wernicke's area. And these are different slices moving up in the brain. So, an extensive activation, specifically in frontal and posterior language areas as these children sounded out the letters. What we think is critical for normal learning ability. And this is just to point out this activation back here, also shown on the side from this view they sound out the letters. Now, here are the dyslexic children and you can see no difference there at all. No activation there at all for this task. They get something in the frontal cortex. It's not quite in the same place, but there's absolutely no activation here at all. That was the one where we saw the second large activation in the successfully reading children. It's as if this part of the brain were simply not engaged in this task of sounding out the letter. It's a pathway that's not used by these children that we think is critical for successful learning how to read. One more thing. Why is language tough on many levels. I mean language is amazingly tough. It's amazing that it works. If it didn't work, you know, if--if a theoretical person, scientist got up and explained to you language you'd think nobody can learn it, no chance, right? One of the reasons language is really tough is we make auditory discriminations incredibly fast. So, this is the "ba" and the "da" sound. And this is a, a spectrogram that shows you kind of a representation of the physical energy that hits your ear. And what you can see, is the difference between "ba" and "da," they look incredibly similar. The only small difference is in the very beginning. Everywhere else the rest of the sound is the same. This is 40 milliseconds of informational difference. So, 40 milliseconds. Think, think about a second. Divide that by ten. You're at a hundred milliseconds. Divide that by half and you're around 40 milliseconds. So, incredibly small difference in what's going on makes an incredibly big difference in what you hear. And of course, you don't sit there and go "ba" or "da," I'll think about that for a while. Right? I mean, you hear a speech stream even with a fast, slower speaker than I. You hear a stream of sounds going by and you're constantly, typically easily, making these difficult discriminations. Pretty naturally, even though they're occurring at light speed in many senses and the order is at tens of milliseconds of critical information. So, language is a phenomenal accomplishment in the sense of rapid processing. Now, Paula Telol has shown that if she takes children who have difficulty in language and who go on to be poor readers--she found the following remarkable thing. She had them do a task where they heard tones. Now, this will remind you a little bit of the tones we heard a few moments ago. In this case there was simply a high tone or a low tone. So, a beep or a beep. And she would play them back to back and your job was to say which came first. The high tone or the low tone. And then she would make the time be less and less between the two so it got harder and harder. Now, for children who are good readers and good in language it never got that hard, even when they only had 15 milliseconds between the two, which is the almost identical moment, okay? They could still perform about 90% accuracy in telling you which one came first. The children who had reading difficulties were pretty good up until they get to about one-third of a second and then they fall off. So, there's no language here at all. It's just, can you make a discrimination of very rapid auditory information that's changing really quickly? And these children can't. So, that's raised the idea that one element of normal language acquisition and one element that can make it hard to become a good reader, is this element of extremely rapid auditory processing. And so, the experiment you just heard and I'll tell play you more examples in one second, we did the following thing. We had participants hear these sounds. They weren't even judging whether they were rapid or slow sounds. They were simply judging whether they were high or low in, in, in tone. It was just to give them a thing to do, okay? We didn't want them to worry whether they were fast or slow. And then we asked what happens in the brain. But, let me play you some of this again. [tones]. Okay. So, first of all you might be impressed if we could get a difference in the brain at all. Right? Okay. That's pretty subtle. That's not like hugely emotional scenes versus neutral scenes. And again, the only difference between those two sounds the [inaudible] and the faster sound is about 40 milliseconds of transience at the beginning. So, it's kind of like those letters you heard before, the consonants the ba and the da. But, it's a nonsense sound. Here's what we find. Here are normal reading children hearing these two kinds of sounds. And they turn on a number of areas, but most consistently in the left frontal cortex. If they hear the rapid versus the slow sound. That tiny, tiny difference. One they're not even paying attention to makes a big difference for the brain. It's as if reflexively a part of the brain says, here's rapid information and this is what I'm good at. Okay. Even if I have nothing to do with it because there's no word to decode and no task. It's just, it's my specialty and the left frontal cortex turns on very powerfully. Now look at the children with reading difficulty. Nothing. Okay? Not, and they have some other activations, but nothing in that spot that's so powerfully turned on in the normal reading children. It's as if again, this part of the neocortex was in some sense deaf to this difference. And so, so now we--that seems to be an element to this story. This rapid processing and sensitivity to that. So, one of the things we like to do in cognitive neuroscience, sooner than later, but sometime or another, is think about not only describing the neuromechanisms of normal learning and thinking about neural pitfalls that prevent good learning, like children with reading difficulty, but it's helping people to get good at what they were not good at. And so in collaboration with Paula Telol and Mike Mersnick we looked at these children before and after they went through a particular remediation program. It's a commercial program and I could talk to you a little about it if you like, and other programs as well. Called Fast Forward. It's computer games that play five 20 minute sessions a day, five days a week for six weeks. And the games are all about trying to get better at rapid auditory processing, both linguistic and nonlinguistic. They also get exposed to slowed down speech. The idea is that rapid speed is too tough for them at the moment. But, if it's slowed down they can start to pick up some discriminations they had not been able to pick up before. And what's the, what's the behavioral consequence of this training program. Well, here's the control subjects that we ran twice. Of course, they don't get much better from this training. They're reading pretty well. But, here's the children who came in as dyslexic, as having great difficulties in learning to read. Before the training--this broken line is sort of a clinical cut off for reading difficulties and you can see they improve when they get above that line. Here's another task for non-word decoding, pronouncing nonsense words. Here's the improvement. And here's passage comprehension. So, these children do show benefits of all this training that can't simply be explained by retesting because there's no benefit from retesting. So, they seem to benefit from reading… The brains in these areas we said before were not being turned on by auditory sounds or, or thinking about the sounds of letters. So, here to let me remind you, was this left frontal activation in the poor readers and the absence of the posterior activation that was present in the good readers. Now, as their language abilities have improved, here's what we see. This activation gets sharpened and this region that had no activation at all, grows in activation. It's a little bit beneath the area that we saw in the good readers, but this area was silent before and now it's full of activity. So, we're seeing some benefits of educating the brain for somebody's brain which is not well suited for this task. How about the rapid auditory processing? Now, we get not as much as in the good readers, but now we get activation in this left frontal cortex that before showed nothing, as they hear the rapid sounds versus the slow sounds. So, what we're able to do is track now--in a way, when we've looked at the mirror reading we wanted to learn about skill learning, but it's an experiment, you know, in a laboratory. Now, we're trying to see whether we can relate these images of brain learning or difficulties to learn and relate them to real educational challenges that we have. The last thing I want to tell you for about two minutes, is one last measure. That's a very new measure that very exciting, I think. It's called diffusion tenser imaging. And let me tell you what it's about. If you're a piece of water bumping around in the brain with hydrogen, if you're in the central, in the, in the fluid, the cerebral spinal fluid or in gray matter, you can go all over the place pretty well. You can diffuse all over the place randomly. But, if you're sitting in strongly myalinated areas--some myalin is the fatty substance that surrounds large axons and promotes the rapidity with which they can signal, send signals from one part of the brain to the other. They are the this, the information superhighways of the brain. Those are, those are very sort of thick structures and the water has to go along those routes. They can't freely go anywhere. They sort of go along those routes or grooves, if you will, along with the myalin. And through technology that is amazing to me, you're able to measure at about the millimeter level of movement of water in this way. And what this shows is a single brain. Top of the brain, back of the brain. And this is the movements of the water. And water that's moving left to right is shown here. Or up and down in the brain. Inferior, superior or back to front. So, here's a white matter pathway going from the back of the brain to the front of the brain. And with this thing, with this technique, we can look at the conductivity of the individual brain, healthy person or patient and how they're brain is organized in terms of white matter conductivity. What is the networking in their brain? Of a healthy intact brain. And what we found when we compared adults who are poor readers or good readers was that there is significantly less good organization of white matter in this posterior area--you know where we said there was a lack of activation for the sound of language in children. And we know, furthermore, that if we look very carefully in that area, we can look at three overlapping things. What's shown in, okay it was shown in blue is areas that were significantly different between the patients and the controls, good readers and bad readers. What's shown in yellow is areas that correlate with reading--I'll show you that in a moment. And green is the overlap. Let me rephrase that. This part of the brain that showed a difference in myalination between good readers and bad readers, also showed the following property. The more it seemed to be myalinated by this measure, the better a reader you were. Not only for poor reader in these dots, but even for good readers up here. It's as if there's a correlate in some way of what reading is about. Whether it's the basis of who becomes a good reader and not such a good reader or there's a consequence of doing a lot of reading or a little reading, we don't know. So, whether it's cause or consequence. But, one of the striking things is this relationship is about the same in the poor readers down here as it is in the good readers up here. And this suggests that poor reading, in most, in many cases, is on a continuum really. You fall below some threshold and it's a problem. And it makes us think about the role of reading in the following sense. That if we're not too good at music--I can't draw worth beans and so people laugh when I draw. We don't pay tremendous prices for that in our society. We respect great artists. We're amazed at wonderful musicians, right? But nobody says, your future is not too promising because your drawing is not too good. Reading is just a different kind of skill, but it's one that in our society is a, is a root to successful opportunity for many people. And so, this is one of the reasons we like to understand who can't read and then what can we do to help them at the younger age the better. So Kirkagard pointed out something that we all ruefully know, that life can only be understood backwards, but it must be lived forwards. But, what I hope I've shown you is that our memory systems in the hippocampus for events and facts, in the amygdala for emotional information, in cortical regions that let us gain skills, both laboratory skills as well as learning how to read--that in fact, our brains empower us by collecting information from the past and letting us live forward in a pretty smart useful way. Thank you very much. |
Irvine Health Foundation |