Science

When I was about three years old I wanted to be a plumber. In primary school my aspirations turned towards astrophysics. Then when I was thirteen my parents got me my first computer and I started writing programs for it. All these occupations involve predicting or controlling well-defined processes or systems, and that, I suppose, really is my fundamental fascination: knowing how the world works relieves my anxieties. By the time I was finishing high school, I was already involved in work in artificial intelligence (AI), writing a program for the US Navy that helped engineers to understand shipboard equipment failures.

At university, I got involved in computer science research and was surrounded by people who were so brilliant that at the end of four years I came out feeling really stupid. An old friend of mine has described this phenomenon as being "the scum at the bottom of the cream," which I think is a funny and apt metaphor. I learnt a great deal at university, but for about two years afterwards I felt too worthless even to touch a computer. I drifted, underemployed, back to London.

In London I applied for a job as a conductor on the Tube. Ever since I was a small child I've enjoyed the comforting determinism of train systems, and had I been able to get a job working with trains I may very well never have done anything else. I was passed over by the Tube, though, and ended up working in a computer shop in Kensington, where I sold discs and laser printer cards, changed printer ribbons, and moved boxes to and from the stock room. The work was dark and rather solitary, and paid hardly enough to eat. When my money and spirit ran out I scurried back to my old university town, got a job in a kitchen, and spent a year and a half working on theatre and arts projects.

Though it isn't science, kitchen work is very much a scientific endeavour, and in retrospect it perhaps is the job that I was best at. There were algorithms to be followed, and a concrete product by the end of each day: so many meals served, so many pots and pans washed up. And unlike what I've been doing since, it wasn't the sort of work that commanded much thought outside of working hours. On the whole it interacted very healthily, therefore, with my obsessionality, and afer I few months at it I was promoted to shift supervisor. As my undergraduate friends left town one by one, though, it began to become clear that I too had to move on, before I was the only one left. My foray into the world of real jobs having turned out so dismally, I began to consider graduate school as an honourable way to get out of town.

I had always been interested in artificial intelligence, but it seemed to me that traditional AI was nowhere near as exciting as the architectures and algorithms that are implemented in neural systems. At the same time, I was becoming interested in autism as a research topic: it seemed that analytical and scientific styles of thought had a lot in common with the ways in which my autistic elder brother seemed to think, and that an understanding of autism could tell us a lot about how and why the human mind is so compelled to impose an order on the world's cacophany. (I've written in the fall 2004 issue of Advances, the newsletter of the Cure Autism Now foundation, about why it is that I find autism so interesting.)

I knew nothing of graduate school, and next to nothing about neuroscience, but I sat down with a Peterson's Guide and wrote away for details on a few programmes. I was rejected by all of them except the University of California San Diego — which happened to be the top-ranked neuroscience research institution! So in June 1992 I packed all my possessions into a van and headed for southern California, despite the US National Science Foundation's official opinion that I was not one-sided enough to be a scientist.

I turned up in California without a clue, and found that the faculty, for the most part, hadn't much time to give me one. The programme had little money of its own; most of the support came from research grants to individual members of the faculty. As a result, teaching and other programme responsibilities always took a back seat to time spent on research. With very little advice, and without having taken a biology course since I was in junior high school, I ended up proposing a second-year project that was entirely naïve. Miraculously, though, the examiners at the viva passed me, perhaps only because it was clear that I'd done a very thorough job with an indefensible topic. (Or perhaps it was the sympathy factor: after weeks of being too anxious to sleep more than a couple of hours a night and too nauseous to keep food down, I must have looked like hell.) With the second-year project out of the way, I embarked on my doctoral research on the neurobiology of autism.

When I was in my first year in California, a senior graduate student gave me some advice. "Choosing a research advisor," he said, "is like getting married. You're going to be depending on each other for a long time." It's only in hindsight that I understand what this person was trying to tell me. My supervisor and I were very similar; that's what drew us together and, four years later and only one year before I would have finished, that's what broke us apart. I came back to the east coast to pick up the pieces, and spent the next two years in New York City working on an MFA in fiction. Whilst there I eventually found a job in software development for visual neuroscience with the Howard Hughes Medical Institute at the New York University Center for Neural Science.

NYU gave me a second chance to get into the neuroscience community, and I appreciate that very much. After a year and a half there, I left in order to enter another graduate programme, again the only one that would admit me. I came to Boston and convinced my new supervisor to apply for a small grant to study autism. We got the grant, and in 2001 — after nine years of graduate school in three different cities — I finally finished what I'd started in California.

It was shortly after I turned up in Boston that I started spending time at MIT. Ever since primary school I had dreamt of being part of MIT; it's a place that understands the way geeks think, where one doesn't have to translate to the way the rest of the world thinks. Alas, though I was keen on MIT, MIT never felt the same about me! (My loss, or theirs?) Eventually I found my way into the MIT community through the back door, by teaching there, becoming a member of the MIT Student Information Processing Board, and working with some brilliant and creative people at an MIT startup company. I like the fact that MIT is such a meritocracy: its culture recognises smart people and affords them the resources to accomplish what they want to do.

I was always bothered, though, by the degree to which the mainstream MIT student culture marginalised intellectual activities other than science and engineering. Many of the undergraduates at MIT had grown up during the 1990s technology boom, surrounded by people telling them how smart they were, asking them to build web sites and so forth, and this cultural background seemed to have bred an attitude that if someone weren't working in technology, it must be because they weren't smart enough to work in technology. Along with this low valuation of non-scientific endeavours came a tendency to approach complex social problems as though they were engineering problems, by isolating the fault and replacing the bad component. I am glad that MIT exists; there's no other place like it. I'm also glad, though, that the world isn't run by engineers.

Shortly after I finished my degree work in Boston, the government of the United States asserted that everyone must be either with it or against it. Faced with this choice, I decided that I must be against it, for I certainly was against its growing curtailment of civil rights and suppression of political dissent domestically, and its foreign policy eschewing negotiation and emphasising the use of force. Accordingly, I decided to return to England for my next appointment.

In June 2002 I moved to the University of Cambridge to continue my research on the neurophysiology of attention in autism. I discovered that Cambridge as an institution was rather inhospitable to postdoctoral researchers, mostly because postdoctoral researchers had existed for about a quarter of a century and Cambridge takes about half a century to adjust to any sort of social change. Trying to insinuate myself into the social fabric of the place, I applied the same strategies that had worked so well at MIT: teaching, and joining societies. These were, of course, miserable failures at Cambridge since the colleges and other institutions are so closed. At MIT, one becomes accepted because of what one knows and what one can do: it's possible to succeed socially by being good with engineering systems; one starts with that and the personal connections come later, scaffolded by the shared language of science. At Cambridge, though, one becomes accepted more because of whom one knows and where one has been, and social success depends on being good with people, straight from the start, without the scaffolding of shared experiences. Too often, then, rather than encouraging individual initiative and creativity, Cambridge trains students to know their place — and this is a great loss of human potential.

Most of my first year at Cambridge had been spent finishing work from Boston, and much of my remaining years there was taken up by large, collaborative review papers, and then in the midst of it all I returned to the United States for a while to volunteer for the Kerry campaign. So it took me four years to find the time to complete the research project that was to have been funded by my two-year postdoctoral grant. Luckily, I was accustomed to living on a student budget, and had been able to save enough money to get through those two years without a pay cheque.

After the election (alas), the political situation in the United States began to improve: all of a sudden, it seemed, the Americans caught up with the rest of the world in viewing the war as unjustified and invasions of privacy and liberty as unwarranted. I began spending less and less time at Cambridge, instead travelling about with a laptop on which I handled all my work, sofa-surfing and pirating network connectivity with friends at Caltech, UCSD, Cornell, MIT, Yale, and Duke. (For the record: I wrote the 2004 Journal of Neuroscience paper in the Green Dragon café at Cornell whilst camping illegally in Fall Creek Gorge, the 2005 Annual Review of Neuroscience paper during one very caffeinated week at the Broad Café at Caltech, and most of the JCPP paper in the Alpine Atrium and Francesca's coffee shop at Duke.) I knew that this Erdös act couldn't go on forever, though, and I knew that I couldn't face the loneliness of Cambridge again, so I began to cast about for alternatives, in a job search that was at least as unconventional as my graduate school search had been.

I spent four years at Cornell (minus a six-month stint at the Broad Institute back at MIT), where I built a state-of-the-art high-density electroencephalography laboratory, received my first large grant from the US National Science Foundation, and accomplished some good science with a team of some very smart undergraduates, an expert postdoctoral researcher and a superb doctoral student. At the game of politics, though, I was failing just as utterly at Cornell as I had done at Cambridge! And Cornell's location in the middle of nowhere presented some unique geographical and social challenges. After those four years of trying to jam a square peg into a round hole, I shifted my research base to a private foundation. Since then, my work has become international, with projects in India and England, and has taken on more of a clinical, applied focus.