1:54 pm Apr. 3, 2012
You don’t often see people hoping to scalp tickets outside of university lecture halls, but Monday night, as 350 people poured into Havermeyer Hall at Columbia University, one ticketless fan stood out front in a lab coat holding a sign reading “Will Analyze Data For Tickets.”
The crowd was there for “Does the Brain's Wiring Make Us Who We Are?” featuring two leading lights in the field. The debate, the second annual event hosted by the science-writing collective NeuWrite, had sold out weeks ahead of time.
Sebastian Seung, M.I.T. professor and author of Connectome: How the Brain’s Wiring Makes Us Who We Are, has argued that developing a detailed map of neuron connections in the human brain will be the key to understanding how it works. Perhaps “detailed” is too mild: the connectome Seung has in mind is like the Human Genome Project on steroids—the human brain has a million times more connections than the genome has bases. Tony Movshon, director of the Center for Neural Science at New York University, argued the other side—that seeing the structure of the brain at that level of detail won’t actually help us understand how the brain functions, and that “connectomics” would divert funding from more targeted research.
“He’s kind of a rock star!” said Kathryn Devaney, a graduate student in neuroscience who studies vision and drove down from Boston for the event. She had seen Seung speak four years earlier at neuroscience conference and made the trip just to see him.
Inside, the crowd of bespectacled graduate students was punctuated with intellectual celebrities like author Lawrence Weschler and Nobel Prize–winning neuropsychiatrist Eric Kandel. The panelists sat in front of a wall of chalkboards scrawled with equations, beneath a giant periodic table. Seung lived up to his flashy reputation. While the other panelists were dressed in professorial grays that fit the room, Seung was sporting a neon blue anime T-shirt and a pair of glinting gold-sequined sneakers.
Stuart Firestein, the chairman of Columbia’s biology department, introduced Seung, Movshon, and the moderators—Robert Krulwich, host of NPR’s hit science program Radiolab, and award-winning science journalist Carl Zimmer. The audience was ready to learn but also ready to giggle, and they had plenty of opportunities for both. The room burst into laughter when Firestein introduced Seung by saying “The Wall Street Journal hailed his book as the best lay—” and was still quieting down when he turned the page to finish, “… book on science I’ve ever read.” From there, the jokes got more inside.
“The retina: it’s like the invertebrate brain of the vertebrate brain” and “We’re not keeping score—the NIH is keeping score!” got two of the night’s biggest laughs. But even the uninitiated could follow the argument thanks to Krulwich and Zimmer. “Some of this will go right over your head,” Krulwich warned, “and our job is to pluck it down from the air.”
Seung argued that the connectome will give neuroscience something new to work with in three areas where the current theories have been “calculated to death”: in understanding perception, memory, and some psychiatric disorders. His ideas about “connectopathies” were especially interesting. In some disorders, like schizophrenia and autism, scientists have not discovered anything unhealthy about the brain tissue itself. Perhaps neurons in these cases are connected abnormally—a completed connectome could help us see how such diseases work.
Movshon argued that there is a “scale mismatch” in what Seung is proposing, that we simply don’t need a synapse-level map of the brain. We can’t look at the structure of the brain and figure out what it’s doing in the way we can with other organs. You can look at the structure of a kidney and see what it does, but you can look at the structure of computer chip—or a brain—all you want, and you won’t know what information it is processing.
The two went back and forth over whether the connectome would be useful enough to justify the massive resources of time and funding that it would require. Is the connectome simply too grandiose? Movshon said he thinks science functions best as a “cottage industry,” where “small groups of people tackle small problems … through focused, hypothesis-driven experiments.”
“Hey, I just wanna map the connections!” Seung defended himself, saying the connectome isn’t grandiose at all. In its final minutes, Krulwich tried to make the debate more heated, asking Seung and Movshon to identify the worst part of the other’s position. But they weren’t taking the bait. “Ultimately,” Movshon said, “we’re both seekers after a particular kind of truth.”
Was there a winner? Jeannie Vanasco, a poet who teaches creative writing at NYU, said she found Movshon’s arguments more compelling. “When I came in I thought it was definitely going to be Seung—I mean, just look at those shoes!—but by the end Movshon won me over.”
Tim Requarth, co-director of NeuWrite, says simply highlighting the debate was their goal: “Science is often reported in the lay press with a certainty that usually doesn't exist,” and that’s not how science actually works. “Hypotheses have proponents and detractors, and it's often not clear whose viewpoint will prevail. Most of this debate occurs behind closed doors.”