An introduction to the work of David Hubel and Torsten Wiesel
2009; Wiley; Volume: 587; Issue: 12 Linguagem: Inglês
10.1113/jphysiol.2009.170688
ISSN1469-7793
Autores Tópico(s)Functional Brain Connectivity Studies
ResumoIt is with enormous pleasure that I add my voice to that of others of my generation in celebrating the semicentenary of the 1959 publication of Hubel and Wiesel's first paper in The Journal of Physiology entitled: 'Receptive fields of single neurons in the cat's striate cortex' (Hubel & Wiesel, 1959). This paper set the stage for the continuous flow of outstanding papers that emerged over the next twenty-odd years from the Hubel and Wiesel collaboration. Their work and that of Vernon Mountcastle opened up the modern study of the cerebral cortex. As a result of their extraordinary accomplishments, Hubel and Wiesel received the Gross Horwitz Prize together with Vernon Mountcastle in 1975, and the Nobel Prize in Physiology or Medicine in 1981 together with Roger Sperry. It was on the occasion of the Gross Horwitz Prize, on whose committee I served, that I was invited to introduce Mountcastle, Hubel and Wiesel. My initial comments in that introduction were in fact directed toward an excellent scientist, a member of our Prize committee, who commented during our deliberations that Mountcastle, Hubel and Wiesel seemed to represent superb science, but their work had limited biological generality. To which I replied: 'You are right, it does not apply to the kidney or the spleen. It is much more restricted. It only helps to explain the workings of the mind.' Hubel and Wiesel's names are enshrined together in the Pantheon of Creative Collaborations in Biological Sciences, much like Hodgkin and Huxley, Watson and Crick, and Brown and Goldstein. In each case, equal partners joined forces bringing unique skills to their collaboration to produce a new level of science and a new family of insights. I first met Torsten and David in 1957, and we became friends in the period 1960–1965 when I overlapped with them at the Harvard Medical School. That friendship continues to this day, as does my admiration for the work, scientific and administrative, that they have accomplished since going their own directions in the early 1980s. In 1983 Torsten moved to New York, first as Professor and then in 1999 as President of Rockefeller University. I served on the Board of Trustees of the Rockefeller during much of Torsten's tenure and this provided me with the additional opportunity to toast his 80th birthday. What follows in the ensuing set of papers in this issue is an outpouring of affection, respect, and gratitude for Torsten and David, for who they are, for what they have given us, and for setting the tone of our science for my generation in the United States. President McGill, Vice President Marks, colleagues of Columbia University, honoured guests. Even the most important contributions, such as the work by Vernon Mountcastle, David Hubel and Torsten Wiesel that we honour tonight, have in addition to their obvious strengths, certain clear weaknesses. I will discuss the strengths in a moment, but I think it important at the outset to consider its weaknesses. Many contributions in biology inform us about general principles – principles that are important for understanding all the cells of the body. The contribution of Mountcastle, Hubel and Wiesel is concerned with only one class of cells – the cells of the brain. Their findings are therefore somewhat parochial. They are only important for understanding the mind. Despite that limitation, the contributions of Mountcastle, Hubel and Wiesel are nonetheless rich in meaning and have significance for many levels of thought. Their work will, I would suggest, be discussed by historians of science from three quite different vantage points. First, it will be discussed from a purely scientific point of view, as a central contribution to neurobiology; second, from a broader philosophical point of view, as an enhancement of our understanding of mental processes; and third, from a sociological point of view, as an example of the importance of scientific lineage and of small group interactions at large universities. First and foremost, as a scientific contribution to neurobiology, the work of Mountcastle, Hubel and Wiesel stands as the most fundamental advance in our understanding of the organization of the brain since the work of Ramón Y Cajal at the turn of the century. By applying morphological techniques to the cerebral cortex – the highest and most elaborate part of the brain – Cajal revealed a hitherto unanticipated precision of the interconnections between populations of individual nerve cells. Using modern cell physiological techniques, Mountcastle, Hubel and Wiesel have revealed aspects of the functional significance for perception of these patterns of interconnections between nerve cells. They have shown us that the connections filter and transform sensory information on the way to and within the cortex, that the cortex is organized into functional compartments or modules, and that this organization can be altered by experience. By any scientific criteria, these contributions are of the highest rank. But on a second level, the work of Mountcastle, Hubel and Wiesel takes on greater significance because it contributes to our understanding of mental processes, a contribution with profound physiological implications. We appreciate important science because it tells us something new and exciting about the world around us. What is at once so special and so parochial about the work that we honour tonight is that it tells us something new and exciting about the world within us, about ourselves. Let me give you an example. We have the feeling that when we interact with each other – when I speak to you and you listen to me – that we are directly experiencing one another. Hubel, Wiesel and Mountcastle have made us realize that this is an illusion, a perceptual illusion. The brain does not simply take the raw data that it receives through the senses and reproduce it faithfully in the brain. Rather, each sensory system first analyses and decomposes, and then restructures the incoming raw sensory information according to its own built-in connections and rules. These insights are not only remarkable; they are also timely. Hints that similar processes may be involved in the development of language and thought are now emerging from the studies of structural psychologists such as Chomsky and Piaget. On still a third level, the work of Mountcastle, Hubel and Wiesel is interesting because it illustrates in a unique manner the role of social context upon discovery and how small groups in a university can shape social contexts so as to make them conducive to creativity. Mountcastle, Hubel and Wiesel are exceptionally creative, bright and energetic. Each would have made an important mark on science no matter where he worked. But I think it fair to say that the special nature of their contribution was fostered by the particular collegial environment to which they belonged. They, in turn, have now restructured their environment anew so as to foster creative activity in their younger colleagues. Thus, in a certain sense, their work illustrates the role of intellectual continuity and intellectual renewal in the achievement of excellence by segments of two American universities, the Department of Physiology at Johns Hopkins and the Department of Neurobiology at Harvard. Academic life is now often challenged, beleaguered, and fragmented. It is therefore inspiring – and more important, it is instructive – to learn which aspects of academic life are most conducive for the establishment of a powerful intellectual environment that is resistant both to external and administrative pressures – an environment that is at once sensitive to historical perspective while at the same time it encourages the emergence of novel and important ideas. I would like to consider these three implications of Mountcastle's contributions and those of Hubel and Wiesel – the scientific, the philosophical, and the sociological – by tracing the development of only one aspect of their contribution: the discovery of one of the central ideas in the functioning of the brain – the fact that the cerebral cortex is organized into computational modules consisting of vertical columns of nerve cells. That particular strand of research had its origins at Johns Hopkins Medical School in the mid-1930s. Now, as you know, we experience the outside world through our five senses: touch-pressure (and the related skin or somatic sensation), sight, hearing, taste, and smell. Each sensation is first analysed by appropriate receptors and coded in lower relay stages. Most sensations are then elaborated in the cerebral cortex. Modern research on the role of the cerebral cortex in somatic sensation began in the Department of Physiology at Johns Hopkins Medical School in about 1936, with the work of Philip Bard and Clinton Woolsey. Philip Bard was 34 years old and an Assistant Professor in Walter B. Cannon's Department of Physiology at the Harvard Medical School when he was called to chair the Department of Physiology at Hopkins. Not only was he extraordinarily young at the time of this appointment but he had published only three original papers. He was the sort of person whose future one would worry about nowadays. Dean Tapley assures me that he would never pass our Appointments and Promotions Committee. In fact, Dean Tapley was surprised that he slipped by even at Hopkins. Soon after coming to Hopkins, Bard teamed up with two even younger colleagues, Clinton Woolsey and Wade Marshall. Using gross electrophysiological recording techniques developed by Marshall and a strategy developed by Woolsey, these three young men discovered that the body surface of monkeys was systematically represented on the surface of the brain. This was soon confirmed in humans by the Canadian neurosurgeon Wilder Penfield and established the fact that not only monkeys but each of us has within our brain a naked representation of our own body, the closest thing to our true self-image. This remarkable discovery that animals and man have a representation of their body on the surface of their brain raised a number of conceptual problems. Somatic sensation is not unitary but a composite of several distinct sensations called submodalities. We can, for example, readily distinguish the pressure on deep tissue from the light touch on the skin. However, it appeared that the maps for these deep and superficial submodalities were completely congruent. Clearly with the relatively gross techniques used by Marshall, Woolsey and Bard – techniques that averaged the responses from thousands of nerve cells – some critical dimension in the map was overlooked. After making this major contribution, Bard withdrew from the study of somatic sensation, leaving it to Woolsey and the younger members of his department who came along later. The particular question of submodality perception was picked up in 1948 by Vernon Mountcastle, Bard's most gifted student. Born in Virginia and educated at Johns Hopkins Medical School, Vernon Mountcastle was dissuaded from a career in Neurosurgery by Bard, who enticed him into Physiology. I have always considered Mountcastle's decision a great gain for Physiology, but having recently become familiar with the economics of neurosurgery, I am only now beginning to appreciate what a loss this decision has meant for Vernon Mountcastle. Over the years Mountcastle not only took over Bard's fascination with skin sensation but also other aspects of Bard's mantle. In 1946 when Bard retired Mountcastle assumed the directorship of the Department of Physiology. He also took on the editorship of Bard's distinguished Textbook of Physiology. Mountcastle realized early on that by using the cellular techniques that became available in the late 1940s he might be able to detect new dimensions in the map of the somatic sensory system that eluded the gross recording techniques used by Bard and his colleagues. This task required a number of major technical innovations including new microelectrodes and precise quantifiable natural stimuli – innovations to which Mountcastle contributed importantly. With these tools in hand – tools that formed the basis of modern cortical physiology – Mountcastle addressed the question of submodality specificity. He found that at the cellular level, there is within all areas of the somatic sensory system a segregation of submodalities that was not resolved with gross recordings. First, he found that single nerve cells respond specifically either to superficial touch stimuli or to deep pressure stimuli, almost never to both. Second, he found that cells responding to one submodality were located together and were segregated from cells responding to other submodalities. The most fascinating example of segregation is found in the cortex. In a classical paper published in 1957, Mountcastle described his remarkable discovery that submodalities were distributed in the cortex as vertical columns running from the surface of the brain to the white matter below it. Each column is submodality-specific. All the cells in a column receive information from a particular point on the skin and from a particular class of receptors, either superficial or deep. Thus, each region of the skin projects to a particular area of the cortex, and the separate receptor classes are distributed in adjacent columns. The distribution of neurons in columns is therefore the mechanism whereby the depth of the cortex is used to handle different functions for the same small region of the bodily map. Each column is an integrating unit, or logical module, comprising thousands of neurons that form the initial stage in the cortex for elaborating sensory experience into consciousness. In order to follow the history of the discovery of columnar organization, I will now describe the numerous other contributions that have subsequently come from Vernon Mountcastle. These include the analysis of the flow of information from skin to the cerebral cortex, a correlation between cellular responses and perception, and recently a study of the mechanisms underlying attention and the control of purposeful movements. In the 40 years that have passed since Bard, Woolsey and Marshall first mapped the representation of the body surface onto the brain, the Department of Physiology at Johns Hopkins, first under Bard and subsequently under Mountcastle, has been preeminent for research training in skin senses. As a result of Mountcastle's recent work, this preeminence has now been extended to the study of attention and behaviour. Indeed as a result of the leadership of Bard, Woolsey and Mountcastle, Hopkins was for many years so outstanding in the study of sensation that it also dominated research in hearing and in vision. For example, while he was still at Hopkins, Wade Marshall, who had already contributed importantly to the early study of skin sensation, teamed up with Samuel Talbot to demonstrate that the cortex also contains detailed maps of the retina. About 10 years later, in 1948, a young man, Stephen Kuffler – whom I am delighted to see here tonight – was recruited to the Wilmer Eye Institute at Hopkins. Upon arriving, Kuffler turned his attention from synaptic transmission to cellular studies of the retina, an area to which he immediately made fundamental contributions. In 1955 Kuffler was joined by Torsten Wiesel, a young postdoctoral fellow from Sweden. Wiesel had experience in child psychiatry and a particular interest in vision. Three years later, David Hubel joined Torsten Wiesel in Kuffler's laboratory. Born in Canada, Hubel took his residency in Neurology at Hopkins where he met Vernon Mountcastle. He then spent two years at Walter Reed working on vision when Mountcastle recruited him back to his laboratory at Hopkins. However, when Hubel arrived Mountcastle's laboratory was in the process of being renovated. Hubel therefore accepted a temporary invitation to work in Kuffler's lab. He turned out to be a guest who stayed for more than dinner. Confronted with two gifted young investigators each interested in vision, Kuffler set Wiesel and Hubel to work together in vision and went off on his own in a new direction. The largely accidental meeting of Wiesel and Hubel in Kuffler's laboratory in 1958 gave rise to what has been one of the most remarkable, sustained and productive collaborations in contemporary science. Although each has occasionally worked with another collaborator, almost all of their fundamental contributions have involved simply the two of them. Soon after the beginning of their collaboration, Kuffler was invited to join the Department of Pharmacology at the Harvard Medical School to head a small laboratory of Neurophysiology. A true pater familias, Kuffler took with him the four young faculty people then working independently in his laboratory. Hubel and Wiesel, the two 'brain boys' as they were called, and Furshpan and Potter, the two 'membrane boys.' They were soon joined by a fifth colleague, an enzymologist, Ed Kravitz, whose function it presumably was to explain it all in the universal language of biochemistry. Harvard responded to Kuffler's brilliant recruitment effort in a typical manner – it immediately demoted Hubel and Wiesel from the Assistant Professorships they held at Johns Hopkins to a non-professional rank. This was of course only to be expected from a university that had within recent years successfully denied tenure to two Nobel laureates, Georg Von Bekesey and Fritz Lipmann. But Harvard found its match in Kuffler and his boys. Unlike Bard who devoted much time to administration in his later career, Kuffler was and still is the bench scientist's bench scientist. Nonetheless over a 10 year period, in his own quiet way, Kuffler turned the laboratory of Neurophysiology at Harvard, consisting of one professor and five postdoctoral fellows, into a Department of Neurobiology, the first in the country – 100 persons strong. A veritable intellectual dynasty, the department now occupies, as far as I can tell, about half of the available space at the Harvard Medical School. And you cannot open a door in all that space without some energetic, enthusiastic and aggressive young student coming out to criticize your every idea and replacing it with one of his own. Much of what we now consider modern neurobiology – the fusion of disparate scientific strands into one – was born out of Hopkins's gift to Harvard. You only have to visit the Harvard department to see why. As with Bard's department at Hopkins, so in Harvard's Department of Neurobiology, the world belongs to the young. While still at Hopkins, Hubel and Wiesel began to apply cellular techniques to the visual cortex. Kuffler had earlier recorded from single cells in the retina and made the surprising discovery that the cells do not simply signal absolute levels of light; rather, they signal contrast between light and dark. The most effective stimulus for exciting these cells was not diffuse light but small spots of light. Hubel and Wiesel found a similar principle operating in the next relay stage, the lateral geniculate nucleus. However, at the level of the cortex, Hubel and Wiesel found that most cells no longer responded to small spots of light. To be effective, a stimulus had to be a line, a square, or a rectangle. Thus Hubel and Wiesel found that these cortical cells did not simply and faithfully reproduce the input from the lateral geniculate nucleus but, by virtue of their connections, the cortical cells were able to abstract linear aspects of the stimulus. The stimulus requirements of the cortical cells are amazingly precise. In addition to requiring linearity, each cell is coded to respond to a specific axis of orientation; some cells respond best when the axis of the line stimulus is running vertically, others when the axis is horizontal, still other cells respond only to various oblique angles. Every small segment of the retina is represented in the cortex with every angle or orientation. It is attractive to think that these cells are the early building blocks in the perception of form and contour. Hubel and Wiesel next found that cells with similar axes of orientation were grouped together into columns similar to those which Mountcastle had found in the somatosensory system. Now Hubel and Wiesel have profoundly original minds. To confirm someone else's work is anathema to them. They are most happy writing papers where all the references are only to their own work – a happiness they have somehow sustained with surprising regularity. Upon finding columns, they quickly proceeded to extend our insight into the nature of columnar organization. First, they predicted and found another completely independent system of columns in the cortex – the ocular dominance columns – a system concerned with information from the two eyes. These columns serve to elaborate binocular vision necessary for depth perception. Second, they utilized a variety of morphological techniques to visualize the columns in three-dimensional space. The early work of Mountcastle, and of Hubel and Wiesel described columnar organization strictly on the basis of electrical recordings from single cells. Routine histological examination had failed to reveal columnar organization. However, capitalizing on the current revolution in morphological techniques – a revolution that is based upon the use of marker substances that label cells according to one or another aspect of their functional activity – Hubel and Wiesel could independently label both the ocular dominance columns and their orientation columns. The results they have obtained with these methods are not only aesthetically breathtaking but have given us a completely new sense of the organization of the cortex – an insight made possible only by moving beyond the range of traditional anatomical approaches. Thus, they have made us realize that we are just beginning to explore the structural organization of the brain and its possible alterations by disease. No wonder we have so little understanding of the biological basis of most forms of mental illness. Finally, Wiesel and Hubel have used these studies of the normal columnar organization to investigate the effects of sensory deprivation on newborn animals. They found that a fairly subtle procedure such as closing the eyelids of a newborn monkey for just a few days leads to prolonged and sometimes irreversible blindness. Concomitantly, the closed eye loses its ability to control the firing of nerve cells in the cortex. By contrast, similar experiences in an adult animal produce no effect on vision. In a brilliant series of studies, Wiesel and Hubel found that visual deprivation in infant monkeys profoundly alters the organization of their ocular dominance columns. Normally the columns for each eye are equal in size. After deprivation, the columns that receive input for the deprived eye are much narrowed compared to those that receive input from the normal eye. The scientific and philosophical implication of this work is truly enormous. Here is direct evidence that sensory deprivation in early life can alter the structure of the cortex. As Hubel pointed out in his Bowditch Lecture of 1967: Experimental psychologists and psychiatrists both emphasize the importance of early experience on subsequent behaviour patterns – could it be that deprivation of social contacts or the existence of other abnormal emotional situations early in life leads to a deterioration or distortion of connections in some yet unexplored parts of the brain. The columns were first discovered by Mountcastle in the somatic sensory system and their functional properties and their alteration by experience were analysed by Hubel and Wiesel in the visual system. More recently, columns, stripes, sheets and other types of functional modules have been encountered in other areas of the cortex and in yet other regions of the brain. Here these modules are related to other forms of sensation as well as to motor control. It is clear that we are dealing here with one of the key principles in the organization of the cortex and the cornerstone for future work on the brain. In closing, I would like to return to the question of social context. I have already drawn analogies between the social values and attitudes of Bard and Kuffler and the resulting flowering of scientific creativity in Mountcastle, Hubel and Wiesel, who in turn have now created environments where others can be independent and creative. Moreover, much as Mountcastle, Wiesel and Hubel sit here tonight to receive the same honour that this University bestowed on Kuffler six years ago, I am confident that this University will in the future honour the intellectual descendants of Mountcastle, Hubel and Wiesel. By drawing this analogy, I do not mean to imply that these five distinctive and unconventional people – Bard, Kuffler, Mountcastle, Hubel and Wiesel – are cut from one mould. But I do want to indicate that they share two rare qualities that are often overlooked in considering why gifted scientists often train other exceptional scientists. This is a question that has now been examined by a number of sociologists, most prominently by our colleagues at Columbia, Professors Merton and Zuckerman. They have pointed out that good scientists can teach their junior colleagues the importance of working on really significant problems; they also offer resources and visibility, and make available to their students important channels of communication. All of these qualities, however, are in line with the direct expression of a senior person's own ambition, of his own desire for growth. What is evident in the three men whom we honour tonight and in their scientific patrons are two features that are much rarer, much more special. One is the ability to rein in their own ambition and to encourage without reservation the creativity of their gifted young colleagues. The second feature is the ability to build around them an exciting environment made up of gifted peers, an environment where important science is routinely done because the environment is consciously structured around one or more absolutely central ideas. The resultant internal cohesion has the additional consequence that it buffers the environment from disturbing vicissitudes of academic and scientific life. Both of these features derive, I believe, from the ability of these rare scientists to combine a remarkable intuition of what is important in their field with an almost childlike enjoyment of the day-to-day experimental and intellectual activity of science. They enjoy day-to-day science as a satisfying end in itself, an ever-changing intellectual adventure – a repertory theatre of ideas. It is for these reasons that the Columbia community takes such special pleasure in honouring Mountcastle, Hubel and Wiesel. For their contribution to the biology of the brain – the most remarkable and profound of our generation – represents at once scientific and personal qualities at their highest. In honouring them, we are drawing attention to the excitement in learning and the respect of colleagueship that drew so many of us to academic life. Their contribution gives us not only an insight into the brain and into ourselves but it also gives us specific examples of how the best in personal and academic values can be achieved and sustained over generations. (From An Introduction to Torsten Wiesel's 80th Birthday Celebration, Rockefeller University, June 3, 2004) I have the pleasure to serve as the after-dinner toastmaster for Torsten's 80th birthday celebration. Let me introduce things with an overview of Torsten's life and work. I will divide my comments in three parts. First, I will describe Torsten's scientific accomplishments and his style of leadership. Second, I will try to develop a theory of Torsten Wiesel's life and work based on a fundamental typology that distinguishes between different classes of scholars and scientists. Finally, I am going to outline some limitations to Torsten's greatness. I begin with a description of Torsten's life and times. Torsten was born in 1924 in Uppsala, Sweden, the son of a prominent psychiatrist who served consecutively as superintendent of two large psychiatric hospitals. One of the perks of being superintendent was a residence on the property, so Torsten spent most of his early years on the grounds of a mental hospital surrounded by mental patients. In retrospect, it was probably this extensive experience with the institutionalized mentally ill that allowed him to function so effectively as president of Rockefeller University in the aftermath of David Baltimore's resignation. I am here reminded of the story told by Douglas Bond, the great American psychiatrist who rose to become Dean of the Medical School at Western Reserve. A friend once asked him: 'Doug, I don't understand how you could take on an administrative job and give up seeing patients, which you always so enjoy doing.''I haven't given up seeing patients,' insisted Bond. 'I still see patients. It's just that now, my patients have tenure!' In 1945, at age 21, Torsten entered the Karolinska Institute to begin medical training. At the Karolinska he particularly enjoyed the lectures on the brain by Ulf von Euler, the discoverer of noradrenaline, and by Carl Gustaf Bernhard, who worked on epilepsy. Bernhard went on to become Torsten's great mentor and friend. During his last year of medical training, Torsten worked first in adult and later in child psychiatry. But by the end of that year he decided that the available treatments did not satisfy him. So he turned to the biology of the brain and in 1954 joined Bernhard's department of neurophysiology at the Karolinska Institute as an instructor. One year later, in June of 1955, Bernhard asked him whether he might want to go to the United States to work on the retina as a postdoctoral fellow with Steven Ku
Referência(s)