January 17, 2008
Arriving at neuroscience via physics
Dr. Adrienne Fairhall, assistant professor of physiology and biophysics, is a computational neuroscientist studying how neurons transform input from other neurons and how this input is refined through neuron adaptation.
Fairhall recently received a McKnight Scholar Award, one of the top grant awards for neurosciences researchers in the early stages of their careers, and the Society for Neuroscience Career Development Award, an early- career award recognizing outstanding researchers.
She sat down with University Week recently to talk about her academic background and current research.
Q. Your background is in theoretical physics. How did you wind up going into neuroscience?
A. I was at the Weizmann Institute working on my Ph.D. in theoretical physics, and I met some friends who had physics training but who had gone into neuroscience. I liked the environment. It was very interdisciplinary, the way people from different fields communicated was very respectful, and there were a lot more women. The physics attitude at the Weizmann was very combative. I didn’t respond well to that, and I got tired of it.
Neuroscience was much more inclusive, and much more collaborative. It seemed so full of opportunity to do new work and good work. Understanding how the brain works is so clearly important and exciting.
Then it was time do post-doctoral work to prepare myself; I wanted to make a real move. Institutions were starting to offer fellowships to bring in people from the quantitative sciences to the neurosciences.
Q. And where did you wind up?
A. I went to the NEC Research Institute in Princeton, N.J. (now known as NEC Laboratories America, Inc.). It was great — focused and close-knit, but I was also able to go to seminars at the university. My adviser was Bill Bialek, who also advised the UW’s Fred Rieke. We were working on the fly visual system, trying to quantify the neural code, and what spikes in the electrical signals mean.
Physicists are very dismissive of the neurosciences — they call it “mushy.” But one can do quantitative work: You can take a neuronal spike in a brain and analyze it and determine what it represents about the outside world.
Q. Did you keep up this work on fly vision?
A. I moved to the retina. There we were trying to interpret what different signals from the retina meant, and I’m also interested in adaptation. A particular neuron gets many different inputs from other neurons through synapses, and it’s a complicated signal that varies in time. The neuron is like a mini-computer — it’s looking for a pattern it likes in the input, and when it sees it, it fires.
Where we’re trying to go with that is find out what single-neuron computations allow the brain to do.
The way people have previously thought of neural networks is that they’re a collection of cells where each neuron has some simple basic function. Our analysis shows that we had a simplified view of neurons — single neurons can do so much more. For example, they can look for changes in input, rather than just firing when the signal stays “ON” for a while.
I’m also interested in the issue of adaptation to the statistics of a stimulus. This is what you see in vision, for example, when you go from bright light to a dark room, or if you’re in a dark interior but looking out a window to the bright outside.
This is very complex. As you look around, your ability to see detail stays about the same. We have been studying whether the way adaptation happens is in some sense optimal.
We have found so far that when the environment changes, neurons are able to change their response curve to fit the new stimulus very quickly — but that there are also processes that seem to keep track of how rapidly the stimulus changes, perhaps to match the rate of adaptation to the typical timescales of change of the environment.
Q. You were recently asked to speak at an invitation-only event for TED [Technology, Entertainment, Design], the organization that bills itself as bringing together the world’s most fascinating thinkers and doers. What was that like?
A. I spoke as part of TED University, which was a side event, not one of the main hall speeches. It’s pretty intimidating — a lot of famous people were there. Many people there were really interested in the brain. I talked about sensory coding. The other talks were all over the map — there were people talking about how to bring technology to rural areas of the developing world, for instance.
TED is an amazing conference; it’s not the usual thing for academics. Seeing what people are trying to do to change the world was very exciting and inspiring.
The guys from Google were there, and Jeff Bezos from Amazon.com. They have a lot of resources and they want to do a lot of good with it.
