William H. Calvin and George A. Ojemann's CONVERSATIONS WITH NEIL'S BRAIN (chapter 15)
Home Page || Public Bookmarks || Science Surf magazine || Table of Contents
Conversations with Neil’s Brain
The Neural Nature of Thought & Language
Copyright  1994 by William H. Calvin and George A. Ojemann.

You may download this for personal reading but may not redistribute or archive without permission (exception: teachers should feel free to print out a chapter and photocopy it for students).

William H. Calvin, Ph.D., is a neurophysiologist on the faculty of the Department of Psychiatry and Behavioral Sciences, University of Washington.

George A. Ojemann, M.D., is a neurosurgeon and neurophysiologist on the faculty of the Department of Neurological Surgery, University of Washington.

Why Can We Read So Well?

YOU KNOW,” Neil said, “I’ve traveled all over the world. But I’ve never seen an electrical plug like that one. You must have gotten it in Afghanistan.”
      I was being helpful while George’s technician set up the slide projector alongside Neil’s hospital bed. I’d tried to plug the power cord into the wall outlet, whereupon I realized that it was one of those old, oversized, explosion-proof connectors used in operating rooms back in the days when volatile general anesthetics were used.
      The technician smiled and pointed to the adaptor cord stored on the bottom shelf of the equipment cart. And so I explained it all to Neil while temporarily converting the fancy connector into an everyday three-prong plug. The slide projector lit up quite satisfactorily.
      While the technician sorted the circular trays of slides and checked out the equipment, Neil and I talked some more about reading. Neil was still thinking about possible side effects of the operation.
      “George said the removal might be close to important areas for reading,” he said. “He told me I might find reading a little slow for a while afterwards, but that it ought to return to normal after a while.”
      “Over dinner,” he continued, “I got to thinking about what you’d said about the visual system being mostly evolved so that monkeys could find fruit, high up in the trees amid all the leaves waving in the breeze. How did reading areas get tacked onto that system? Writing was only invented 5,000 years ago, and not that many people have been literate until the last few centuries. That means any hereditary tendencies conserving brain areas useful for reading can’t have operated for more than several hundred generations. Which is nothing. So how did we get reading areas so quickly?”
      Reading most likely is a secondary use of a cortical area with some other primary purpose, I told him, what Darwin liked to call a conversion of function in anatomical continuity.
      No question, though — there are now cortical areas that seem essential for reading, at least in people who have grown up reading a lot. My father once had a bad headache, quite unlike any other he ever had before. But the next morning, he felt somewhat better and fixed himself breakfast, then went out to pick up the newspaper off the sidewalk. Upon sitting down to breakfast and unfolding the newspaper, he discovered to his astonishment that he could not read it. The words weren’t blurred. He could name the letters, but not the words.
      Later in the hospital, he was asked to write out a paragraph in longhand that was read aloud to him. He accurately wrote out this paragraph, though tending to write all the way over to the right edge of the paper, rather than leaving a margin.

bk7p232.jpg 24.9 K
[FIGURE 73 Stroke causing reading problems but without aphasia]

      But asked to read aloud from his own handwriting, he couldn’t, except for the shortest two- and three-letter words. He would try to piece together longer words but often made errors. His spoken language was normal. He understood everything that was said to him. He didn’t have any abnormal blind spots that might interfere with reading. His color vision was normal. He just couldn’t read anymore.
      A year later, he had recovered his abilities to read the newspaper, but he still tired easily and wouldn’t read for more than twenty minutes at a time. He watched the television news more often.
      “Do people often recover like that?” Neil asked.
      Recovery of some function is typical of mild strokes. Whatever function is lost — whether reading or muscle control — some lost abilities return in the weeks and months following the stroke. In part, this recovery occurs because, while some brain cells are gone forever, the merely injured cells tend to recover.
      Furthermore, the brain’s plasticity may allow other areas to do the job instead. This on-the-job retraining depends quite a lot on where the stroke is located, and on the patient’s age. For example, soldiers receiving brain injuries at age 18 recover more function than soldiers who are 30 years old and suffer similar injuries.
      But, to come back to reading, there are many brain sites that have been identified over the years with alexia — that’s just the name for the loss of previously normal reading abilities, without losing much else. Following the disconnection models for the Broca-Wernicke variants, alexia was usually ascribed to disconnection of language areas from visual centers. Indeed, the brain imagery showed that my father’s stroke had destroyed a half-dollar-sized area just above and behind his left ear. It was to the rear of the posterior language (“Wernicke’s”) area, but in front of the traditional visual areas of the brain, perfect for an interrupted pathway between the visual cortex and the language cortex.
      “I have a nephew who’s dyslexic,” Neil said, “smart but he can’t read or spell. Is that problem tied in with those same pieces of the brain?”
      Dyslexic kids are considered relatively nonverbal and quiet, with their development of reading and spelling considerably delayed, compared to their other skills. Their spoken language may be a little slow too, but they are often smart otherwise. The result, of course, is a major mismatch between the child’s delayed reading skills, his other normal skills, and a lockstep school system — one that increasingly relies on reading textbooks in the higher grades to get the lessons across. And so these kids experience a bottleneck, just not being able to absorb information as rapidly as everyone else — even if smart.
      “That’s my nephew, all right.”
      Actually, the long-term outlook for his learning to read isn’t too bad. Few dyslexics get to the point of enjoying reading, but most can master enough reading to get by. And there’s the story of John Hunter who, George tells me, founded surgical anatomy back about the time of the American Revolution. Hunter was born into a wealthy British family that tried to educate him with tutors. They despaired of ever teaching him to read. He didn’t learn until he was nearly twenty. But that didn’t prevent a later academic career.
      “So what’s the problem in the brain?”
      Many cases of dyslexia are associated with abnormalities in brain development. Some dyslexics seem to be have problems in the way the structure of the language areas developed. In a few dyslexics who have died from other causes, the brain shows mistakes in cortical development. The appropriate layering is lost and neurons are in misplaced bundles called “heterotopias.”
      Dyslexics seem to have less difference in the size of that temporal lobe language region — the planum temporale — that is usually bigger on the left. However, most dyslexics have left-brain dominance for language, just like most of us. There is a suggestion that some male dyslexics may have less strongly lateralized visual-spatial functions than other males. That raises the possibility that visual-spatial functions compete with language for the same left-brain territories. And so the dyslexic kids can’t do either job well, neither language nor visual-spatial.
      Dyslexics also have problems in processing rapidly changing sensations. The magnocellular pathways that transmit rapidly changing visual sensations are slowed in some dyslexics.
      “Is that a consequence of those two magnocellular layers of the geniculate? They got scrambled? No longer the nice layer cake that you told me about earlier?”
      Right, the two bottom layers are rather jumbled together. And there are 27 percent fewer neurons in the magnocellular layers than is normal, as if they’d lost a lot of them.
      What causes these abnormalities is much less clear. In some cases, the brain changes associated with dyslexia may have been acquired from an event that occurred during development or early life. But more commonly, it seems to be genetic. Dyslexia is more common in males than in females. It frequently runs in families. It’s apparently not a matter of just one gene, for dyslexia has been tightly linked to chromosome 15 in one family — but not to that chromosome in other dyslexics. Like reading in general, as we discussed before, there can’t have been much natural selection yet to eliminate such variants, given the short time that reading has been important for many people.
      And natural selection often can’t operate against traits that are linked to other inherited traits. The neurologist Norman Geschwind noted an association between dyslexia and an odd constellation of other abilities. Extraordinarily good mathematical abilities. Left-handedness. Allergies. Not all those have been confirmed by all the later studies, but you can see why things get complicated when there are links.
      Several possible explanations for Geschwind’s constellation have been proposed. Perhaps there were unusually high levels of testosterone in the mother during the later parts of the pregnancy. Or perhaps it is a manifestation of an autoimmune disease, when the body’s natural defenses are turned on its own tissues. In either case, you could get defects in neuron organization in cortex and sensory system relay nuclei. The autoimmune disease suggestion has had a particularly promising spin-off. There is a strain of mice that have an autoimmune disorder. These mice also show disorganized cortex and mouse-learning deficits, providing what may be an animal model for the study of dyslexia.
      “Why so much trouble with reading, though,” Neil asked. “Especially if the abnormalities are distributed throughout language areas? Why not other aspects of language as well?”
      Well, most dyslexic children also had some trouble with speech earlier. It just wasn’t as bad as that with reading. George has found that many more neurons change activity with word reading than with naming objects or repeating words. So reading may just be intrinsically harder, requiring more neurons. If there’s a shortage of well-organized neurons, it would then be more apparent in reading.
      Also, written speech lacks many of the redundant clues present in real speech. When I’m talking, my voice rises and falls, my facial expressions change, I wave my hands and shrug my shoulders. That’s additional information. When you’re reading, the written word is all you’ve got.
      It seems possible that the major fault was originally not with reading but with listening to speech — detecting the fine timing differences in the sound waves that make a “pa” different from a “ba” (there’s a little z-like buzz at the beginning of “ba” because the vocal cords vibrate then, and they don’t at the beginning of “pa”). That’s again a magnocellularlike task. Yet it might have a lot to do with reading — because learning to read is all about matching up speech sounds with letters, at least in a phonetic language like English.
      Since George’s studies showed separation between naming and reading for many of the essential pieces of cortex, I suppose that it’s possible that the specific system for reading is defective. Nobody really has any information on that in dyslexia.
      George also showed a strong association between the location of reading and naming sites and, of all things, that patient’s verbal IQ. Presumably that’s because verbal IQs measure both reading and other parts of language. So some patterns of location of these functions in the brain are more favorable than others. He found that in patients whose sites for reading were in the superior temporal gyrus, with naming in the middle temporal gyrus, had high verbal IQs. And he found the reverse in the patients with low verbal IQs.
      Neil seemed puzzled. “I can’t see why reading above and naming below should be such a good thing,” he said.
      We haven’t thought of a really good explanation, although people who study how we read emphasize that good reading requires the ability to readily convert from speech sounds to the visual representation of those sounds. And since auditory processing is mostly buried in the great infolding of the sylvian fissure, that might be the reason why having reading nearby, in the superior temporal gyrus, is associated with better function.
      “But how does brain development organize all this, if reading isn’t specified by the genes?”
      Well, the ability to name is learned first. Maybe if the language cortex is innately less efficient in the below-average folks, more patches are needed to do the naming job. When reading is learned a few years after naming, the superior temporal gyrus is — in this theory of George’s — overcommitted. And so reading must “make do” with sites in the middle temporal gyrus — a potentially less favorable location, since it is farther away from the cortical area receiving sounds.
      So a superior temporal gyrus that can make room for reading at age 5 — because naming can be done efficiently elsewhere by then — will be a better home for making the connections between the sounds of words and their written phonetic equivalents. And so a higher verbal IQ results.
      Speculations such as these are useful because they suggest new questions: What is the relation between vocabulary acquisition between ages 2 and 4, and the child’s reading speed upon reaching age 6, and the verbal IQ even later in life? Can the lower IQ patterns be spotted using functional MRIs early enough in life to do something about them, through intensive reading therapy? And whatever the explanation for the mapping findings, it’s apparent that where you have naming or reading located has important implications beyond those involved in planning epilepsy operations.

AFTER REHEARSAL, Neil and I got back into our discussion of brain research, and why there wasn’t more of it — all those unanswered questions. My answer to that was relatively simple — four out of five grant applications for brain research are rejected.
      Approved by the peer-review groups as likely being good science, I told him, but rejected administratively by the National Institutes of Health because of inadequate funds.
      “I suppose Congress can’t fund everything.”
      Yes, but the annual costs, in the United States alone, of the major diseases of the nervous system are estimated at $400 billion — which is more than the U.S. defense budget! And even more of a drag on the national economy. The estimate doesn’t even include the drug-abuse costs in most cases, which for alcohol is quite high. About $107 billion is direct costs such as hospital bills, and the rest is the disability payments, the lost salary, and so on. About a third of the total is for the 4 million persons with Alzheimer’s dementia — and that number has been growing rapidly.
      “How many people does this add up to?”
      In the United States, about 50 million people with the various neurological and psychiatric disorders.
      “So one person in five — one in every family, on average. And the total cost is a few hundred billion. So what’s the research outlay — a tenth of that?”
      If only. The federal dollars for neuroscience research funneled through the National Institutes of Health and the National Science Foundation — the basic research on brains plus the slightly more applied research focused on the psychiatric and the neurological disorders — add up to about $1.2 billion. Drug company research and private foundation money adds some — but our overall commitment to finding out more about the brain and its disorders is probably less than 1 percent of the overall costs. Development costs by manufacturers adds some more. That’s for brains — for health-care expenditures in general, one estimate for 1990 was 3.3 percent for research and development, down from about 5 percent in the 1960s.
      “In other words, a drop in the bucket. Peanuts. That’s an absurdly low research-and-development percentage even for a horse-and-buggy industry like the utility companies, never mind the 10 to 20 percent for a rapidly changing industry like the one I’m in. And I thought that medicine had a real commitment to changing things.”
      Curiously, medicine doesn’t have much control over its research expenditures. The drug companies are an exception, but most of their research is applied, not basic — and most of their expenditures are more properly called product development. The development costs, just to get just one new drug to market in the United States, are about $240 million. They get their ideas for new products from the basic research of the public sector, and that’s the underinvestment problem. Unfortunately, because medical expenditures are so dispersed and so lacking in central authority, there is no way to reinvest a small part of health care expenditures into research for tomorrow.
      And so, quite reasonably, it is funded through federal tax dollars — government has accepted the responsibility but continues to fund it at an absurdly low rate. The basic research that eventually paves the way for new drugs and new treatments has to compete for federal dollars with the highly visible, and more easily understood, needs.
      “Compete, in other words,” Neil interjected, “with the big-ticket items like space stations which provide thousands of jobs in someone’s state.”
      The average brain research grant probably provides two jobs. And, in the United States, over 70 percent of all research and development jobs have been in defense-related industries — compared, say, to Japan where it’s about 5 percent. Compared to other countries, we’ve diverted a lot of our best intellects into arms.
      Basic research isn’t the sort of thing you can just increase to meet a crisis, the way that the shipbuilding industry was built up during the World War II. How well we can meet a problem like AIDS infections of the brain — and if you can’t fix those, there’s not much point in fixing the rest — depends on how much was spent on basic research in prior decades. Over the years, the research outlays certainly haven’t increased in proportion to the health-care costs.
      “What medicine needs,” Neil said, “is the equivalent of the highway trust fund, where construction and repair funds are kept proportional to road use via the gasoline tax. Something like a tax on health-insurance premiums ought to assure an adequate research-and-development base. And a few percent of tax is nothing — certainly not when compared to the absurd costs of running that paperwork mill for medical insurance, which has been driving me crazy. Paperwork is supposed to be 25 percent of health-care costs. It’s really galling to think that a little creep in the medical costs due to paperwork is the same amount of money that could have doubled our research investment.”

LATER IN THE EVENING, I got a phone call from Neil in his hospital room. He’d been thinking some more about dyslexic kids and wanted to talk to me about it, before taking the bedtime pills that the nurse had brought.
      “Is there a critical period for preventing dyslexia?” he asked. “I’m thinking of those disrupted layers of the geniculate that you mentioned earlier.”
      The magnocellular layers in the bottom of the lateral geniculate nucleus, that they found were somewhat shrunken and a little disorganized in dyslexics. What about them?
      “Maybe those neurons, being a little disorganized, can’t get their act together in time. And so they lose out in the tuneup period. Can you exercise those fast pathways in infants, keep them competitive with the others?”
      A Head Start program for magnocellulars? There are certainly neurophysiological techniques, not unlike EEGs, which you could use to test infants for tendencies to dyslexia. They’re very similar to the brain-stem evoked potentials that we use to detect deafness in newborns; you could do both tests at the same time. Actually, you don’t even have to wait for that. If the father is dyslexic, just assume that the infant is at risk — inheritance is pretty strong for dyslexia.
      And you can easily imagine daily exercises, giving the destined-for-dyslexia infant some computer displays every day that are entertaining to look at, and which are jittery enough to exercise the magnocellular paths more than the parvocellular paths. Maybe such compensatory exercise would increase the survival chances of those fast-path connections and prevent them from being eliminated during the critical period. Maybe that would keep the child from having trouble with reading a few years later. A lot of maybes, but such are the ways in which basic research insights suggest suitable strategies for applied research and can carry over into solving practical problems.
      “So where do practical ideas like that come from?” Neil asked. “Were people working on the dyslexia problem — or something else — when they discovered this fast-slow distinction?”
      Something else, as usual. Although a more subtle version of the fast-slow specializations had been seen in the spinal cord, its exaggerated visual system version was discovered in the cat retina by the neurophysiologist Christina Enroth-Cugell and her student John Robson in 1966, working in Northwestern University’s electrical engineering department. No one knew what the fast-slow, transient-sustained business was good for, but it was an intriguing puzzle, and basic researchers persisted. Soon it became apparent that the fast-slow distinction was seen in those layers of the geniculate: the bottom two layers were “fast.” And geniculate constituted an even more long-standing unsolved puzzle — those six distinct layers, and no one knew any reason for them.
      Like a partially complete crossword puzzle, it was tantalizing. But it was a quarter century, and hundreds of investigations, later before the relevance to dyslexia became apparent. My guess is that it will also prove relevant to clumsiness and verbal intelligence. It certainly shows you how basic research differs from applied research or product development.
      “Sounds like a 25-year version of my Sunday crossword puzzle! Well, see you tomorrow, bright and early.”
[T]here are three quite different levels of technology in medicine, so unlike each other as to seem altogether different undertakings. . . . Supportive therapy. . . tides patients over through diseases that are not, by and large, understood. . . . [Second are] the kinds of things that must be done after the fact, in efforts to compensate for the incapacitating effects of certain diseases whose course one is unable to do very much about. It is a technology designed to make up for disease, or postpone death. . . .
      It is characteristic of [such halfway medical technology as epilepsy surgery or cardiac-care units] that it costs an enormous amount of money and requires a continuing expansion of hospital facilities. There is no end to the need for new, highly trained people to run the enterprise. And there is really no way out of this, at the present state of knowledge. . . . The only thing that can move medicine away from this level of technology is new information, and the only imaginable source of this information is research.
      The third type of technology is. . . so effective that it seems to attract the least public notice; it has come to be taken for granted. This is the genuinely decisive technology of modern medicine, exemplified best by the modern methods of immunization. . . and the contemporary use of antibiotics and chemotherapy for bacterial infections. . . . The real point to be made about this kind of technology - the real high technology of medicine - is that it comes as the result of a genuine understanding of disease mechanisms, and when it becomes available, it is relatively inexpensive, and relatively easy to deliver. . . .
      It is when physicians are bogged down by their incomplete technologies, by the innumerable things they are obligated to do in medicine when they lack a clear understanding of disease mechanisms, that the deficiencies of the health-care system are most conspicuous. If I were a policy-maker, interested in saving money for health care over the long haul, I would regard it as an act of high prudence to give high priority to a lot more basic research in biologic science. This is the only way to get the full mileage that biology owes to the science of medicine, even though it seems, as used to be said in the days when the phrase had some meaning, like asking for the moon.
Lewis Thomas, The Lives of a Cell, 1974

Conversations with Neil's Brain:
The Neural Nature of Thought and Language
(Addison-Wesley, 1994), co-authored with my neurosurgeon colleague, George Ojemann. It's a tour of the human cerebral cortex, conducted from the operating room, and has been on the New Scientist bestseller list of science books. It is suitable for biology and cognitive neuroscience supplementary reading lists. ISBN 0-201-48337-8.
AVAILABILITY widespread (softcover, US$12).
Home Page || Science Surf magazine || Table of Contents || End Notes for this chapter || Continue to NEXT CHAPTER