Image of NLM Family Foundation banner
Image of hands, bodies, and instruments around a face: Autism is a complex developmental disability.


December 2, 2005


NLMFF interviewed Albert Galaburda, MD, Chief of Behavioral Neurology at Beth Israel Deaconess Medical Center and Emily Fisher Landau Professor of Neurology and Neuroscience at Harvard Medical School, about dyslexia research, its relation to autism and language, and the future of autism science.
Image of Al Galaburda, MD

NLMFF: In the 1980s, you wrote a series of classic papers on cerebral lateralization with a pioneer in behavioral neurology, the late Norman Geschwind. Could you please comment on your recollections of your work with Geschwind, on your experiences and thoughts as you wrote these papers, and on how many of these ideas have stood the test of time?

Dr. Galaburda:
As you may know, Norman Geschwind was my mentor. He was the head of the department where I did my neurology training, at the then Boston City Hospital (now Boston Medical Center). In the mid 70s the then Mayor of Boston, Kevin White, decided that three medical schools at BCH was administratively unwieldy, so Tufts and Harvard left with their training programs, leaving Boston University the sole medical school affiliated with the municipal hospital for the City of Boston. Norman Geschwind and the rest of us moved to the then Beth Israel Hospital (now the Beth Israel Deaconess Medical Center) and most of my research collaborations with NG took place while we were at the BIH. In fact, while my clinical work was at the BIH, my lab remained at BCH until 1980, at which time it also moved to BIH.

Norman was very interested in the brain substrates for the phenomenon of cerebral lateralization (having to do with the notion that one side of the brain seems to have a dominant role for some cognitive functions, whereas the other side leads for others). In the 1960s he had rediscovered and extended the observation that the planum temporale, a region of the brain that participates in language, was larger on the left side in the majority of people. My first involvement in this subject with Norman had to do with his asking me to look for more microscopic bases for cerebral dominance. Thus, I used methods of cytoarchitectonics, which I learned from the German anatomists Friedrich Sanides and later Heiko Braak, to show that there existed in the human and animal brains asymmetries in the sizes of architectonic areas. There were several papers published on this subject and my career was launched.

In the 3 years prior to Norman's death, we worked together trying to compile all the available data from various areas of neuroscience to attempt to discover all possible biological agents that could affect brain asymmetry. By doing this we became aware of hormonal and immunological influences, mainly, and some genetic influences, then and still not well understood, and we published 3 serial papers in the Archives of Neurology presenting and interpreting our findings.
In short, hormones, particularly male hormones, we thought, were involved in the diminishing of brain asymmetry during development, and the same hormone modulated the development of the immunological system, such that disorders of the immune system and disorders of brain lateralization, both modified by male hormones, traveled together. This research was highly controversial, but so exciting that many investigators felt compelled to prove it right or wrong.
There were many papers in either direction until the interest of the scientific community gradually shifted to other questions that could be answered with emerging methodologies that appeared only in the 80s and 90s, after Norman's death.

I myself did not pursue many of the areas of research that were set up by the three papers, mainly because I felt that we didn't have the tools to answer them adequately. Instead I chose to study specific aspects relating to brain asymmetry and abnormal brain development with small experiments using contemporary tools. This led to my work in dyslexia. There was also another issue: Norman was much more phrenological in his thinking than I was. He was still highly influenced by (and highly interested in) the work that had begun at the start of the 19c, known as the phrenological school (c.f., Franz Josef Gall, Broca, others) that made two specific claims. The first was that bigger was better, that is, that a bigger planum meant that it was better (this collapsed when we started to see that dyslexics, whose main problem was language, had two big plana, not two small ones, as predicted by the phrenological model). The second claim was that functions were localized to specific nodes in the brain. Well, work before Norman died, and work that came out of the introduction of functional brain imaging, were clearly showing that complex brain functions are instantiated in distributed networks, rather than localized nodes. It was not the case that every part of the brain participated in every function, but it did appear that the circuits were quite large and non-focal. Thus, I abandoned attempts to show the speciality of small nodes of brain and focused instead on explaining skills and loss of skills in disease by the disruption of large networks.

NLMFF: Could you comment on the current state of knowledge about dyslexia, and on your views regarding the future of dyslexia research?

Dr. Galaburda: Dyslexia research is at an exciting juncture today because a tentative pathway is in its rough outline now available between a gene mutation and an alteration of a complex phenotype, in this case phonological knowledge (a fundamental issue in most dyslexia). There have been 4 dyslexia susceptibility genes discovered in the past 2 years, all of which are involved in brain development. Three are involved in neuronal migration, which my lab showed to be abnormal in dyslexia, and one is involved in the establishment of neural connections across the hemispheres. One of these neuronal migration genes, DCDC2, is part of a developmental pathway that is already rather well understood. Except for the possible exception of DCDC2, none of these genes are actually markers for dyslexia (some people who are not dyslexic have the gene mutation), but this issue will undoubtedly be resolved. The future of dyslexia research, therefore, seems bright to me. We are in a position to clear up a lot of details in the pathway (or set of pathways) by which a gene mutation changes brain development in specific ways that can explain difficulties in phonological processing. Perhaps working out these details is not as exciting as having gotten this far, but the work will undoubtedly continue and good answers that have the potential to be applied to the diagnosis and treatment of dyslexic individuals will be forthcoming. If one gets bored doing this, there is always autism to challenge us.

NLMFF:  In your opinion, what are the lessons to be learned from dyslexia research about autism and language?

Dr. Galaburda: We can all agree that however complex phonological development is in children, its genetic background, its brain developmental steps, and its interactions with the environment and culture are ultimately tractable. The dyslexia situation is likely to be a great deal simpler than the case of autism. However, it is reassuring that science is progressing so that the biology of complex traits, including complex cognitive traits, is advancing rapidly. It is possible to think of a gene mutation that changes the developing brain; it is also possible to think of specific changes in the brain to explain specific behaviors. It is more difficult, but possible to see how a gene mutation affects brain development, which in turn affects a behavior. This is what we need to do for autism. It is the same type of challenge as that which has been partly met for dyslexia, albeit quantitatively more difficult. I believe we collectively have the brain power and methodologies, together with sound theoretical notions, to extend knowledge of brain development coming from the fields of genetics and developmental neurobiology to make it relevant to autism.

NLMFF: What are your thoughts on the recent genetic discoveries in autism?

Dr. Galaburda: Difficult to say at this time, because as usual when things are getting started the genetic mutations that are discovered account for very small proportions of the known affected population. Nonetheless, even though mutations linked to autism are at present explaining only a few autistic persons, it will be important to study these in great detail. By this I mean to discover all the downstream effects of the mutation to get us to specific brain regions the interference with which can explain the observed symptoms and signs of autism. Once we are sure of the brain regions involved by converging evidence (functional and structural imaging and molecular biology both pointing to the same brain regions) we can turn our attention to other developmental mechanisms acting on those regions (caused by other genes known to be expressed there during development, or epigenetic effects known to affect the same regions during development).

NLMFF: Is it possible to understand how the brain creates language and how genes influence this?

Dr. Galaburda: In principle, yes. However, in practice, we are far from being able to answer these questions. As you know, even genes currently linked to disorders of language (and presumably related to the building and/ or function of language circuits), are genes that also express in other species, which do not have anything resembling human language.

This makes for a much more challenging research program. The first burning question is whether language emerges in the human brain because large numbers of genes work together to cause the brain to be larger, to have therefore more computing capacity, which permits the emergence of new functions not pre-adapted in other species, including language, musical ability, math, etc; or, is there a single gene or a relatively small set of genes that is not present in non- language species that first appear in humans and builds special modules that support language functions. If we know even this, we may have a chance to discover how one or the other scenario can explain the building of a special brain, the human brain. After that we still have a problem because we do not know the nature of the neural codes that explain language functions. In other words we need to understand the brain (and the genes that build it), the behavior (the stuff cognitive psychologists study) and the neural codes that join the two. I don't see this happening in my lifetime.

NLMFF:  You recently wrote a manuscript entitled, "Unlocking the Secrets of Autism". Could you please summarize the main points of this paper?

Dr. Galaburda: The main points of the paper are the following. One, that the core systems of autism are confounded by the fact that they may be causally related to one another. Thus, social brain problems and communication problems may be caused by one another or may be manifestations of the same underlying problem, and stereotypic thoughts and actions may also relate to the development of social interactions and communication. One way of interpreting all the data is to state that first and foremost elements of the social brain do not develop normally in autism and other things happen through the process of developmental brain plasticity. Second, that autistic communication problems may represent a form of disconnexion, a form of apraxia, rather than a fundamental disruption of knowledge representation in the brain. This means that the normal ways for learning and expressing communication skills are blocked by aberrant brain development, and in some autistic individuals one can see that language and thought is capable of boring through and coming out in alternative ways, such as writing and reading, for instance. Third, that a valid interpretation of the available data suggests that the initial problem affects parts of the brain that are ancient, quite vertically organized (meaning that damage thereof has a hard time recruiting other systems for help) in the mesial, inferior frontal cortex and its connections with other parts of the frontal lobe, the parietal lobe, the temporal lobe, and the occipital lobe. Finally, the location of this vulnerable zone suggests that it can be hit by genetic or epigenetic factors early in development leading to comparable symptoms, the severity and variety of which depend on timing and extent of early involvement. Among the genetic factors are genes specifically involved in the development of this brain region and viral, toxic, or immunological agents that can damage it.

NLMFF:  What is your present view regarding the central issues of autism research?

Dr. Galaburda:  The central issues in autism research are clearly to get at causes, to understand them at multiple levels (molecular--to get diagnostic markers and pharmacologicals; systems--to get at diagnostic markers and at rehabilitation approaches), and to quickly translate lab results into clinical applications. An important early way to get quickly moving on this agenda is to do away with too much discussion about variability of clinical manifestations in autism and the fact that it has multiple causes, and to extract a kernel that unifies all autistic persons. One loses useful information by doing this, to be sure, but this information can be retrieved later when it is possible to deal with it instead of being distracted by it.

NLMFF:  Given your present view of the central issues of autism research, what do you feel would be the most promising avenues of research in autism?

Dr. Galaburda: The success of plans for the future is best predicted by the success of recent projects. Thus, the powerful tools of molecular biology and cognitive neuroscience are in the best position to get us where we want to be. We need to get from gene to brain (even if the gene accounts for a small number of autistic persons) and we need to get from behavior to brain (using imaging in living autistic and control participants). Even at the risk of criticism by those who think autism is about human beings only (they had the same criticisms about our animal models of dyslexia), there needs to be more research on animal models, at least regarding preadapted behaviors in these animals the understanding of which will shed light on behaviors that suffer in autism.

NLMFF: Thank you so much Al for your reflections, insights, and efforts to advance dyslexia and autism research.

Copyright © 2011 Nancy Lurie Marks Family Foundation
Link to homepage Link to links Link to contact page Link to search page