Exploring Neural Function: From Retina to Visual Perception in the Cat's Brain, Exams of Neuroscience

Download Exploring Neural Function: From Retina to Visual Perception in the Cat's Brain and more Exams Neuroscience in PDF only on Docsity!

Born: Philadelphia, Pennsylvania March 11, 1938

Education : Yale University, BA (1960) Harvard Medical School, MD (1964)

appointmEnts: Intern in Surgery, Massachusetts General Hospital (1964–1965) Peace Corps Physician, Venezuela (USPHS) (1965–1967) Assistant Resident in Surgery, Massachusetts General Hospital (1967–1968) Resident in Neurosurgery, Massachusetts General Hospital (1968–1969) Postdoctoral Fellow, Department of Neurobiology, Harvard Medical School (1968–1971) Postdoctoral Fellow, Department of Biophysics, University College London (1971–1973) Assistant Professor–Professor, Department of Physiology and Biophysics, Washington University School of Medicine (1973–1985) Professor of Neurobiology, Department of Anatomy and Neurobiology, Washington University School of Medicine (1985–1990) Chair, Department of Neurobiology, Duke University (1990–2002) George Barth Geller Professor, Department of Neurobiology, Duke University (1990–2013) Director, Center for Cognitive Neuroscience, Duke University (2003–2009) Director, Neuroscience and Behavioral Disorders Program, Duke-NUS Graduate Medical School (2009–2013) Executive Director, A*STAR Neuroscience Research Partnership (2009–2012) George B. Geller Professor of Neurobiology, Emeritus and Research Professor, Duke Institute for Brain Sciences (2013–present)

Honors and awards (sElEctEd): Mathilde Solowey Award in Neuroscience (1979) Alexander Forbes Lectures, Grass Foundation, MBL (1983) Camillo Golgi Award Lecture (FIDIA) (1988–1989) Election to National Academy of Sciences (1989) Grass Lecture, Society for Neuroscience (1990) Lezioni Lincee Award Lectures, Accademia Nazionale dei Lincei, Italy (1992) Election to the Institute of Medicine (1996) Election to the American Academy of Arts and Sciences (1999) Although Dale Purves is known for his earlier work on neural development and synaptic plasticity, his research during the last 20 years has sought to explain why we see and hear what we do, focusing on the visual perception of lightness, color, form, and motion, and the auditory perception of music and speech. The goal of this work has been to establish an operating principle for nervous systems based on the fact that sensory systems cannot measure physical reality directly and therefore must use accumulated experience to generate the basic qualities we perceive. This understanding of neural function runs counter to the widely held view that sensory information is used to compute models of reality.

occasional autopsy, the ambience of medicine was appealing. The obvious respect of patients and nurses for the doctors encouraged the sense that medicine was a good bet. And my high school courses in chemistry and phys- ics under an excellent teacher began to restore my academic credibility. It was soon clear that I would head to college as a premed student.

College

When it came to choosing a college, I gave the issue very little thought. My father, presumably having seen too many Princeton products in Philadelphia society, indicated that if this was my choice, I would have to pay my way. That left Harvard and Yale. Harvard had a reputation for being effete and Yale appealed to me as more “manly.” My ignorance about what would be facing me and my unsuitability for playing the proverbial Jack Armstrong All American Boy role was profound. When I matriculated at Yale in the fall of 1956, I quickly discovered some unhappy consequences of my choice. My class of about 1,000 was frontloaded with young men (it was a decade before women were admit- ted) from New England prep schools who knew the ropes and came with a cohort of friends (or at least knew who their enemies were). I knew no one. In my zeal to “belong” at Yale, I had indicated on my application I would just as soon have WASP roommates. As a reward, I was assigned two lily white Protestants with whom I had nothing in common and one of whom was patently weird. Things got worse when my high school girlfriend at the time, a first-year student at Mount Holyoke, unceremoniously dumped me. I was also brought up short by a failing grade on my first chemistry test. None of this helped my mental status. I should have sought counseling but did not want to be “stigmatized.” The final straw in my comeuppance at Yale was “rushing” a fraternity only be turned down despite family connec- tions. Having failed miserably in my misguided effort to become a proper Yale man, I told my father that I was quitting college to join the Army. He convinced me over a somber lunch at the Century Club in New York (to which he had been recently been elected) that this was not a good idea. Although his new club was not the best locus for the meeting, perhaps the subtext was that if he could succeed after being fired in Mexico, I should not quit so easily. Thoroughly beaten, I returned to New Haven. The aspect of Yale that had little of my attention in high school—its academic excellence—eventually began to sink in. Although I had failed my first test in freshman chemistry, I had buckled down and ended the semester with the top grade. Deciding to work hard (I was a premed after all) was the first sensible choice I made at Yale. The second was to major in philosophy. I had a knack for thinking more or less logically and my philosophy professors seemed to recognize that they’d hooked a live one. By the time I was a rising

junior, I had become a committed intellectual and by the end of my junior year, my early failures had dimmed, although they certainly were not forgot- ten. In the end, I graduated 1 of 15 students awarded a degree summa cum laude. The most important thing I had learned at Yale, however, was that I was capable of incredible stupidity at an age when I should have understood myself and the world better. When it came time to apply to medical school, I gave the matter more thought than I had given what college to attend, but it didn’t take much investigation to recognize that Harvard topped the list. I applied for early admission and was accepted with the idea that, given my interest in philosophy, psychiatry might be a good specialty for me.

Medical School

Although “enjoyable” is not the right adjective to describe medical school, I felt for the first time that I was in the right place with the right colleagues and the right teachers doing something serious. Given my potential interest in psychia- try, I was especially attuned to what I might get out of our first-year course on the nervous system. I assumed it would be the beginning of a new effort to learn about brain biology in a more serious way than I had managed as an undergraduate, and so it was. The senior member of the neuroscience teaching faculty was Stephen Kuffler, then in his early 50s and already a central figure in neuroscience. Having recently arrived at Harvard from Johns Hopkins, he had presciently promoted to faculty status two postdoctoral fellows who had been working with him, David Hubel and Torsten Wiesel. They were then 34 and 36, respectively. He had also hired David Potter and Ed Furshpan, two even younger neuroscientists who had recently finished fellowships in Bernard Katz’s lab at University College London. The last of his initial recruits was Ed Kravitz, who at 31 had just gotten his doctorate in biochemistry from the University of Michigan. Furshpan and Potter taught us how nerve cell signaling worked, Kravitz taught us neurochemistry, and Hubel and Wiesel taught us about the organization of the brain (or at least the visual part of it, which was their baili- wick). Kuffler gave a pro forma lecture or two, but teaching was not his strong suit and he had the good sense to let his new young faculty carry the load. Unlike teaching us about nerve cells and the mechanisms of the action potential and synaptic transmission, conveying some idea of what the brain is actually doing was a difficult task in 1961. Hubel and Wiesel dealt with this challenge by simply telling us about their work on the organization and function of the visual part of brain. It was not unusual for professors to cop out by telling us what they had been doing in their labs rather going to the effort of putting together a broader and more useful introduction to some subject. But in this case, it was obvious that what Hubel and Wiesel were trying to do was extraordinary. Based on what Kuffler had established in the retina, they were exploring the properties of visual neurons in progressively higher stations in the visual system of the cat, and this was the work that we students heard about as an introduction to the brain.

Brigham Hospital, where I did my third-year rotation in surgery. Moore was then in his early fifties and already legendary. As an undergraduate at Harvard, he had been president of both the Harvard Lampoon and the Hasty Pudding Club, and was appointed surgical chief at the Brigham at the age of 32. He was a member of the team that had performed the first successful organ transplant in 1954 and had written a widely respected book in 1959 called The Metabolic Care of the Surgical Patient that underscored his status as a physician-scientist. Holding retractors as a student assistant while Moore exercised his jocular authority and surgical skills turned me in a new direction. I decided then and there that I would train in general surgery. After some further electives in surgery during my last year in medical school, I applied for a residency in general surgery at the Massachusetts General Hospital, then considered the top program despite Moore’s preemi- nence at the Brigham. I was accepted and started my surgical internship in the summer of 1964. It is hard to believe in today’s atmosphere of managed care, oversight by insurance companies, and litigation that we “doctors in training” were largely unsupervised. The wards were always filled with patients who could not afford a private doctor, and for better or worse, they were entirely our responsibility (“better” was around-the-clock care by dedi- cated young doctors; “worse” was our lack of experience). The chief resident I worked under during the first part of my intern- ship year was Judah Folkman. Like Moore at the same age, Folkman was thought to be destined for great things, as turned out to be the case. At 34, Folkman was named surgeon-in-chief at Boston’s Children’s Hospital, becoming along with Moore one of the youngest professors ever appointed at Harvard Medical School. He became justly famous in the early 1970s for pioneering a novel way of treating cancer by inhibiting blood vessel growth, and at 46, he gave up this appointment as a professor of surgery to pursue basic research on angiogenesis. Folkman was perhaps the most impressive individual I had ever met. Physically slight, already balding, with a great beak of a nose, he dominated in any setting by intelligence, wit, and force of character. The residents universally looked up to him not simply because he was surgically skilled and supremely smart, but because he radiated confi- dence and integrity that made everyone under him deal better with the daily strife and our inevitable mistakes. For me, however, working under Folkman that year had another effect. Any competitive individual continually measures himself or herself against the qualities and talents of peers. Although as chief resident Folkman was far advanced from an intern struggling to learn the rudiments of the trade, he was only five years ahead of me, and I had to envision myself in his role in the near future. The comparison was discouraging. I didn’t see him as necessarily smarter or feel that I could never reach his level of technical skill (although I had serious doubts on both counts). What dismayed me was his

obvious passion for the craft, which I had already begun to realize I lacked. The recognition that, like psychiatry, a life in general surgery was probably not my calling came at 2 or 3 in the morning when Folkman was trying to make an apparatus to dialyze a patient dying from kidney failure using an old washing machine and other odds and ends he had collected from a hospi- tal storeroom. Although the effort failed, I realized that I would not have made it and that this disqualified me from trying to follow the footsteps of figures like Folkman and Moore.

Venezuela

Thus, I needed to invent another path to the future. This time I was given some breathing room by the war in Vietnam. By 1965, virtually all physi- cians in training were drafted, and my notice arrived about halfway through my internship. Given my concern over the prospects of a career as a general surgeon, being drafted was not entirely unwelcome. There was another reason as well. Since childhood I had suffered periods of depression. A year of psychoanalysis during my last year of med school had not helped, although it confirmed my conviction that psychiatry was not something I wanted to pursue. I ended my internship year without a clear plan and clinically depressed. The bright side was that I would now have two years of enforced service to pull myself together and sort things out. Despite the rapidly escalating war in Vietnam, the options for physicians drafted in 1965 were still broad. I could join one of the armed services; seek deferment for further surgical training; apply for a research position at the National Institutes of Health; join the Indian Health Service; or apply to become a Peace Corps physician. This last option meant serving two years in the Public Health Service taking care of Peace Corps volunteers in one of many countries around the world. Given my uncertain frame of mind about what to do next, my total lack of interest in research, and my opposition (along with almost everyone else I knew) to the war, the Peace Corps seemed the best bet. And so, after a few weeks of remedial training in tropical medicine at the Centers for Disease Control in Atlanta, I arrived in Venezuela in July 1965. As it turned out, the Peace Corps was a good choice. The volunteers that I and another young doctor had to look after were an inspiring bunch who demonstrated all the good things about Americans and American democracy of that era. Since I spoke Spanish as the result of living in Mexico as a kid, I could travel easily and interact well with local doctors. And Venezuela in the 1960s was a beautiful, prosperous, and relatively progressive country. From the medical perspective, the job was easy, serving as general practitioner to about 400 sometimes difficult to reach but generally interesting and healthy young adults. My life in South America was in every way a radical change, and for the first time since college, I had time to think rather than simply meet the demands of medical training.

my interest in perhaps giving research in neuroscience a try. On the face of it, my arguments were feeble: the sum total of my research experience was the disastrous summer spent in a pharmacology lab, and my desire to try research had been reached largely by excluding other options. Despite the weakness of my case, Potter said he would think about my situation, and we agreed to meet again. In the meantime, I asked Sweet if I might take my first year in the neurosurgery program as a research fellow, and, primarily because of scheduling issues, he agreed. When I returned to Potter’s office a couple of weeks later, he considered my intention to try research plausible enough, and suggested that I contact John Nicholls to see if he could take me on as a fellow. Nicholls was work- ing at Yale as an assistant professor and had just been recruited to Harvard by Kuffler. I was disappointed because I had no idea who Nicholls was and had hoped to work with Potter himself or perhaps Hubel and Wiesel. I was even more dismayed when Potter told me that Nicholls was working on the nervous system of the leech. I had no idea what any of the neurobiology faculty was doing then or why, but it was difficult to imagine how the leech was pertinent to my ill-formed ambition to become a neuroscientist. In fact, the suggestion was a good one. I was somewhat reassured when Potter told me that Nicholls had been a graduate student with Bernard Katz in the late 1950s after he had completed his medical training in London; that he had been a fellow in Kuffler’s lab thereafter; and that his work on the leech was widely regarded as an outstanding example of what was then a new approach to understanding neural function—namely, studying the nervous systems of simple invertebrates. More to the point, Potter went on to say that because Nicholls would be starting up a new lab at Harvard, he would probably welcome a fellow, even one whose experience in neurosci- ence was nil. Thus, I wrote to John who invited me to visit him at Yale. And so on a bleak Saturday in February 1968, Shannon Ravenel, whom I was engaged to marry later that spring, and I drove down to New Haven. I had met Shannon, an up and coming editor at Houghton Mifflin, while still in med school and we had had an on-again off-again relationship that, happily for me, had been on-again since I returned from Venezuela. Yale Medical School was unimpressive (I had never actually been there, even though it was only a few blocks from the residential college where I had lived as an undergraduate). John’s lab was equally nondescript, and it was quickly evident that John himself had a complex personality that might not be a good fit with my own. Driving back to Boston that night, Shannon pointedly asked me if I really wanted to make this change in the light of all the evidence that I would be sailing into uncharted waters. Even though my confidence in answering was minimal, there were several reasons that argued for seeing it through: Potter’s word that Nicholls was a good mentor; the lack of obvious options if I wanted to try research; and a determination on my part to do something that might ignite a passion for

understanding the brain. And I could always go back to the neurosurgery program at Mass General if the research year failed. Thus, a few days after getting back to Boston, I called John to say that I was willing if he was. Shannon and I married in May, I finished my year of residency in June, and after spending that summer in Vietnam under the auspices of a Boston- based antiwar group selecting war-injured children for treatment in the United States, at age 30, I began my life as a neuroscientist.

Postdoc at Harvard

Although the Department of Neurobiology at Harvard was probably the best place I could have tested the merits of this new choice, the transition was not easy. For the previous four years, I had been a practicing doctor: whether in Boston, Venezuela, or Vietnam, I had had all the responsibilities and respect that being a physician entails. Suddenly, I was a superannuated student on the bottom rung of the ladder; even the two beginning graduate students in the department knew much more science than I did, and they seemed a lot smarter to boot. The stress was of a very different kind than I had experienced during the years I had worked in surgery, but the first year I spent in John’s lab was nearly as trying. There was, however, a fundamen- tal difference: despite my ignorance and well-justified sense of inferiority, I finally loved what I was doing. For the first time in years, I worked hard not because I had to, but because I wanted to. The approaches to the brain and neural function that Kuffler and his young faculty were spearheading when I was a student in 1961 had flow- ered by the time I returned as a fellow in 1968. The question that always confronts the next generation of scientists—Nicholls and the rest of the faculty in this case—was what to go after next. One answer to the question had already been supplied by Kuffler’s study of the retinal cell responses at Hopkins, the impetus for the work on vision being carried on by David Hubel and Torsten Wiesel. By the time I arrived back at Harvard as a fellow, they were already well on the way to the dominant position in brain phys- iology that they would hold for the next several decades. Another aspect of Kuffler’s work had stimulated a quite different direction that seemed equally promising, and this had determined the work John Nicholls was doing when I joined his lab. Following his graduate work with Katz, John had joined Kuffler’s lab as a fellow in 1962, and they had worked together to understand the function of glial cells using the leech nervous system as a model. By the time John left Harvard to join the faculty at Yale, the work on glia had finished, but he continued using the leech as a simple system in which to explore neuronal circuitry in relation to behavior. The opinion held by Nicholls and many other neuroscientists when I joined his lab in the fall of 1968 was that a logical next step in moving beyond the estab- lished understanding of neural signaling would be to focus on invertebrate

one-year-old daughter. What I would pursue in Katz’s small Department of Biophysics at University College had not been specified, but I was by then sure that I had found my professional niche.

Postdoc at University College

Although Bernard Katz’s impact on the course of neuroscience was as at least as great as Kuffler’s, the styles of the two men were entirely different. Whereas Kuffler’s modus operandi was quintessentially eclectic—he would work on a project with a collaborator or two for three years and then move on to an entirely different problem—Katz was a scientific bulldog. He had seized on the fundamental problem of chemical synaptic transmission in the late 1940s and never let it go. And whereas Kuffler was, superficially at least, an extroverted democrat, Katz was reserved and to some degree an autocrat. As a result of these personal contrasts, as well as the cultural distinc- tions between the way science was practiced then in the United States and the United Kingdom, Katz’s Department of Biophysics at University College London (UCL) was about as different from the Department of Neurobiology at Harvard as one could imagine. The labs were on the upper floors of one of the old UCL buildings on Gower Street that ran the length of a long London block and housed most of the basic science departments. The rooms of the five faculty members in the department were comfortable but modest, and the surfeit of furnishings, equipment, and supplies that I had been used to in Boston was nowhere in evidence. Even Katz’s lab was outfitted with equip- ment that would have been consigned to storage at Harvard, and his small office contained the same simple furniture that must have been used by A. V. Hill when he was director in the 1930s. Among other things, all this made clear to me that superb science could be done in modest circumstances. The students and fellows in Katz’s domain (about a half dozen of us) were in labs along a short corridor on the floor that included Katz’s lab at one end, as well as the supply room and a machine shop. For most of the 1960s, Katz had collaborated with Ricardo Miledi, an extraordinarily talented experimentalist who was Katz’s executive lieutenant and the most prominent of the other faculty members. Miledi had a smaller lab adjacent to Katz’s where he pursued his own projects with a couple of fellows in addi- tion to his ongoing work with Katz. The other faculty members were on the floor above and included Paul Fatt, a brilliant but eccentric physiologist who had collaborated with Katz in the early 1950s in discovering the “quantal” nature of chemical synaptic transmission; Sally Page, an electron microsco- pist; and Rolf Niedergerke, another very good biophysicist and a disciple of Andrew Huxley who was working on the properties of heart muscle. Although I expected to work on a project that would explicitly tap into the expertise and interests of Katz, Miledi, and the others in this new

environment, I had no idea when I arrived of what the options in London might actually be. The year I had just spent working with Jack McMahan in Kuffler’s lab had been very valuable technically. But my work with Jack had not presented a problem that seemed worth pursuing. Having already soured on the nervous systems of simple invertebrates like the leech, the question for me when I matriculated in Katz’s department was what general direc- tion would make sense in that environment and provide me with a starting point for my own research in the academic job I would have to secure when my two-year fellowship ended. The first day I came to work after getting settled in our flat in Hampstead, Katz invited me into his office to discuss what I might do. Katz, then 60, was austere but certainly not the terrifying figure John Nicholls had described. He listened patiently to my ill-formed ideas and suggested that I should take my time in deciding on a particular project. He mentioned that another postdoc, Bert Sakmann, happened to be at loose ends and was also thinking about what to do next. Accordingly, I met Bert later that day and we chatted about the possibility of working together. Bert had been medically trained in Tübingen and later in Munich, where he claimed he had gone in pursuit of the fellow medical student he eventually married. In the course of his medical education in Munich, Bert had spent three years carrying out research on the visual system with Otto Kreutzfeldt. Like many of us brought up scientifically in that era, Bert thought that working directly on the visual system or some other part of the brain was a rather daunting prospect and, with Kreutzfeldt’s help, had sought out further training with Katz to pursue a future working at the seemingly more tractable level of neurons and their synaptic interac- tions. We hit it off well because of our similar backgrounds, shared fasci- nation with all aspects of neuroscience, and corresponding opinions about the odd cast of characters and their relationships in the Department of Biophysics. Although Bert was four years younger, we were both recently married, ambitious, and faced the need to land academic jobs when we finished our fellowships. Thus, we wanted to do something significant that would get our careers off and running. Our initial conversation made clear, however, that neither one of us had a very good idea about what that might be. The issue that finally captured our attention, as well as that of many other neuroscientists at the time, was how neural activity affects synap- tic interactions and neuronal connectivity. It had long been apparent that understanding how experience is encoded in the nervous system was a major challenge in neuroscience. Successfully addressing this issue would explain the way we and other animals learn, and unlike the mechanisms of neural signaling, this problem was far from being solved. Because the currency of experience in neural terms is the action potential, it had been assumed for decades that learning involves activity-dependent changes at synapses.

allowed to start. Although the results were in the end a modest contribution (another couple of good but rarely cited papers in the Journal of Physiology ), Bert and I had a fine time being on our own and doing what we thought was interesting. Katz gave the two manuscripts that Bert and I wrote up at the end of our time in the department his seal of approval and suggested we also show the papers to Andrew Huxley, who was in the Department of Physiology a few corridors away in the warren of UCL buildings. Huxley had been study- ing muscle contraction since the 1950s in work that was as impressive in its own way as what he had done with Alan Hodgkin on the action potential in the 1940s. Since our work concerned muscle fibers, Katz thought Huxley would be interested and might have useful criticisms. Huxley thought the papers more or less fine, but chastised us for having blacked out some noise around the oscilloscope traces with a marking pen, an innocuous bit of pre- Photoshop improvement of our figures that, for a purist like Huxley, was a cardinal sin. Bert left London in 1973 to take up an assistant professorship in Goettingen, where within the year, he began a collaboration with Erwin Neher that eventually led to a Nobel Prize in Physiology or Medicine in 1991 for having developed patch clamping, a further step in understanding the basis of neural signaling that Hodkgin, Huxley, Katz, and Kuffler had done so much to advance in the preceding 30 years. I also had to worry about getting a permanent job and what I would do when I did. While still at Harvard in 1970, I had met Carlton Hunt when, like Katz, he had come by to visit Kuffler. Cuy, who was then in his early fifties, had been Kuffler’s first fellow at Johns Hopkins and had spent four years collaborating with him. Together they had worked on stretch receptors in muscle fibers, a project typical of Kuffler’s nose for important problems. When I first ran into Cuy at Harvard, he had recently moved to Washington University from Yale and was in the process of building a Department of Physiology and Biophysics in St. Louis, having already put together excellent departments at the University of Utah and then at Yale (where as chair of physiology he had hired John Nicholls). Cuy was then— as always—a distinguished figure, and it was obvious that Steve and the rest of the faculty at Harvard liked him and admired the two departments he had already created. Whatever conversation we had then about future plans must have been quite tentative. I nonetheless took note that—if history and first impressions were any guide—Cuy would be an excellent person to work for. I met Cuy again two years later in the summer of 1972 when he visited Katz at University College. Cuy had done a sabbatical year in Katz’s lab a decade earlier when he had taken a break from muscle spindles to study the effects on neurons of cutting their axons, thus interrupting the connec- tion between the nerve cells and their targets. He was a great admirer of

Katz, and took pains to visit whenever he was in England. Cuy took me to lunch to discuss what I had been doing and the possibility of joining his new department. We agreed over coffee that I would visit St. Louis later that fall to have a look. Although the trip in late October of 1972 included the few other places that had indicated some interest in hiring me, I liked St. Louis, Washington University, and the potential colleagues I met there. Washington University had a rich history of research in neuroscience, but most important, I felt I would be comfortable working in a department run by Cuy and that he would provide guidance to someone still relatively untu- tored in science and the ways of academia. And so, with some difficulty, I convinced Shannon that St. Louis was the right place for us—or for me, she would no doubt wish to add—and we arrived in the Midwest on a sweltering day toward the end of the following summer.

Washington University

Given the need to teach the full range of physiology to the medical students, the Department of Physiology and Biophysics Cuy had put together included people who worked on the lung, kidney, and heart. Nonetheless, Cuy’s enthusiasms clearly favored neuroscience, and 6 or 7 of the approximately 15 faculty members were neuroscientists. In addition to me, the faculty included Carl Rovainen, who had been a graduate student with Ed Kravitz at Harvard and worked on the nervous system of the lamprey; Mordy Blaustein, who had been a postdoctoral fellow with Hodgkin at Cambridge and worked on the role of calcium ions in cell signaling; and Alan Pearlman and Nigel Daw, both of whom had been fellows in Hubel and Wiesel’s lab at Harvard and were continuing to work on the visual system. Although my initial interests in neuroscience had been anything but reductionist, everything I had done over the preceding five years had been at a simple model systems level. Thus, I was ill prepared to launch into a project that focused on the structure and function of the brain. But I was at least determined to work on nerve cells in a mammal as a step in the right direction, and on problems that would have more pertinence to brain func- tion and organization than the projects I had cut my teeth on. While still in London, I had of course given some thought to the possi- bilities. With Jack McMahan, I had worked on autonomic ganglia, a staple of the work that was then going on in Kuffler’s lab. Studying these accessible collections of neurons and their connections with both the central nervous system and peripheral targets seemed a good compromise between plodding onward with some aspect of a model synapse like the neuromuscular junc- tion and a more direct attack on some aspect of the brain, which I knew very little about. Autonomic ganglia had been the focus of many key stud- ies of the nervous system since the middle of the 19th century and had set

effects in other targets, such as constricting the blood vessels of the ear. Langley had assessed these differences in the innervation of the ganglion simply by looking at these peripheral effects while electrically stimulating the outflow to the ganglion from different spinal levels in anesthetized cats, dogs, and rabbits. When he stimulated the outflow from the upper segments of the thoracic spinal cord, the animals’ pupil dilated on the stimulated side without any effect on the blood vessels of the ear, whereas when he stimu- lated the lower thoracic cord segments, the pupil was not affected but the blood vessels in the ear on that side constricted. Moreover, when he cut the sympathetic trunk that carried the axons to the ganglion and waited some weeks for them to grow back, he observed the same pattern of periph- eral responses. Langley thus surmised that the mechanisms underlying the differential innervation of the ganglion cells must occur at the level of synapse formation on the neurons in the ganglion. He further suggested that selective synapse formation is based on differential affinities of the pre- and postsynaptic elements arising from some sort of biochemical markers on their respective surfaces. Given these studies and more modern ones by Roger Sperry at Caltech, it seemed well worthwhile to pursue the issue of neural specificity at the level of electrical recordings from individual neurons in autonomic ganglia. Arild Njå, a postdoctoral fellow from Oslo, who was the first to come my way, and I pursued the merits of this idea in the autonomic system of guinea pigs by dissecting out the whole upper portion of the sympathetic chain, keeping it alive in a chamber, and making intracellular recordings from individual neurons in the superior cervical ganglion while stimulating each of the input levels from the spinal cord. The results showed that the synaptic connec- tions made on ganglion cells by preganglionic neurons of a particular spinal level are indeed preferred, but that contacts from neurons at other levels are not excluded. Furthermore, if the innervation to the superior cervical ganglion was surgically interrupted, recordings made some weeks later indi- cated that the new connections again established a pattern of segmental preferences. Thus, neurons of the spinal cord associate with target neurons in the autonomic ganglia of mammals according to a continuously variable system of preferences during synapse formation that guide the pattern of innervation during development or reinnervation without limiting it in any absolute way. Although this work with Arild resulted in several good papers, it was another project I had begun in parallel that eventually occupied most of my attention over the next decade. The ideas on which this work was based came from another direction altogether. The theme that Bert and I had been working on in London was control of the signaling properties of neurons (although we used muscle cells as a model), and I continued to think—along with many others—that such modulation of signaling and its effects on connectivity over the long haul was especially important. In the first couple

of papers to come out of my lab in St. Louis, I showed by electrophysiological recording that the efficacy of the synapses made by the spinal neurons on the neurons in the superior cervical ganglion declined over the first few days after the axons from the neurons to peripheral targets in the head and neck had been cut, and that this decline occurred in parallel with the loss of a majority of the synapses made on the ganglion cells that could be counted in the electron microscope. Because the loss of synapses from the neurons was reversed when the axons grew back to their peripheral targets, the conclu- sion seemed clear: the synaptic endings made on nerve cells do not just sit there but have to be actively maintained. And whatever the mechanism, this maintenance depended on the normal connections between nerve cells and the targets that they innervated. The clarity of these results in a relatively simple system of mammalian neurons was news, and it encouraged further studies along these lines. This research led to the beginning of a long collaboration with Jeff Lichtman and a deepening friendship with Viktor Hamburger, both criti- cal determinants of how this work progressed. Jeff appeared in my lab one day in 1974 and asked if he could chat about his future. He was then a second-year med student and knew me from the lectures on neural signal- ing I had given to his class some months before. Jeff was in the MD/PhD program, and he was trying to figure out what to do for his doctoral work. He seemed nervous and lacked a good reason for wanting to work with me or ideas about what to do. I think he simply saw me as someone who was young and ambitious and who, based on the lectures he had heard, might be a good mentor. My inclination was not to take him on since my experi- ence at Harvard and UCL had been that the best people populated their labs with postdoctoral fellows rather than graduate students. But before reaching a decision, I thought it would be a good idea to ask Cuy. He pointed out that the MD/PhD students were a highly select group, that Jeff would not cost me anything since the program was fully funded by the National Institutes of Health, and that unless I had a very good reason not to I should certainly take him on. Cuy was indeed right: Jeff was—and remains—one of the smartest and most imaginative people I have known in neuroscience and went on to become a major figure in his own right. Getting to know Viktor Hamburger was equally important. Hamburger was far and away the most notable neuroscientist at Washington University in 1973. Because he was in the Biology Department on the undergraduate campus, I had not met him on my trip to St. Louis as a faculty candidate, and to my great embarrassment, I knew little or nothing about him or his work when I arrived in St. Louis. This woeful ignorance brought home to me the parochial nature of my training up to that point. Viktor was a consum- mate biologist and my conversations with him about his work and neural development led me to think more and more about what nervous systems do for animals and less about the details of neurons.