Longing for Faster Results, Theoretical Physicist Turns to Neuroscience

This summer’s announcement that the Large Hadron Supercollider in Geneva had detected likely proof of the Higgs boson, an elusive and long-sought particle, brought back memories for Columbia neuroscientist Larry Abbott.

Abbott, the William Bloor Professor in Neuroscience, Physiology and Cellular Biophysics, was a theoretical physicist until the late 1980s, when it struck him that it might take decades to see the fruits of his research.

He hungered for a field of study where results and insights could come more quickly. A chance visit to a Brandeis neuroscience laboratory where lobster neurons were being studied in a petri dish introduced Abbott to what would become his new passion.

Abbott, co-director with Ken Miller of Columbia’s Center for Theoretical Neuroscience, was initially intimidated by the dramatic shift to a new career. But he found that his colleagues were extraordinarily welcoming and credits Brandeis Biology Professor Eve Marder for guiding him through the process of learning his new field.

When Abbott joined Columbia in 2005, he had completed the transition.

Today, Abbott’s work involves creating mathematical models that join two areas of research—how external stimuli drive perception and how internal processing influences behavior. By allowing scientists to make predictions, the models open the door to future research on the science of decision-making and other aspects of human perception and behavior.

Describing his work, Abbott said, “We’re a loop. We take data from a lower level, build a model and work out what higher-level implications that might have. Then we go to the lab and test it. You can try a lot of things on a model more efficiently than if you were only in the lab. The model tells you what’s consistent, and then you can test it and see if it’s true.”

It’s already known that neuron responses get more complicated as more factors are involved. Previous research on single neurons has revealed a linear story about what neurons do—each neuron with a single task. By modeling larger groups of neurons, Abbott and his colleagues are showing that neurons work on the same task in different ways.

Abbott’s work at Columbia focuses on two main areas. The first, in which Abbott works with University Professor Richard Axel, who shared the 2004 Nobel Prize in physiology or medicine, aims to better understand the process of olfactory sensation—for example, how do you go from a sensory experience such as smelling coffee or sewage to the determination that one smell is good and the other foul?

Understanding such links could potentially help people whose sensory experiences are associated with inappropriate responses, such as those with post-traumatic stress disorder, and potentially reveal how such individuals could learn to change those pathways. Axel has joint appointments in the departments of biochemistry and molecular biophysics as well as pathology, and is an investigator of the Howard Hughes Medical Institute.

Abbott’s second area focuses on the body’s motor system—for example, an activity like stretching your arm across your desk to pick up a cup of coffee. He examines the sequence of muscle activities involved, how to make such circuits more flexible, and how new skills or movements are learned. In this area Abbott works with Thomas Jessell, Claire Tow Professor in Neuroscience and Biochemistry and Molecular Biophysics; Mark Churchland, assistant professor in neuroscience; and others.

A complement to Abbott’s work is the establishment of the Grossman Center for Statistics of Mind through a substantial gift from the Sanford J. Grossman Charitable Trust. Co-directed by Churchland and Liam Paninski, associate professor in statistics, the center will support new quantitative methods in neuroscience, emphasizing an interdisciplinary approach to engage Columbia faculty from across the University, and enabling researchers to work with larger and more complex data sets.

Grossman, who established the center in his name, is an accomplished academic and quantitative economist turned hedge fund manager. “It’s fun to have a donor that you can speak to as a fellow scientist,” Abbott said of Grossman, “and he has lots of good ideas.”

“You don’t know what’s in the big data unless you synthesize the real nuggets of information that tell us what’s happening,” he added. “If you get that wrong, you can misrepresent all the data. If you get it right, you can capture what’s going on in a very useful way. That’s exciting.”

Abbott is enthusiastic about this expanded opportunity to “build mathematical techniques that are as good as the experimental techniques,” describing constant interaction with the experimentalists.

“Columbia’s environment is very co-operative and collegial,” Abbott said. “Everyone wants to know the answers, so the focus is really on the science.”