CRS Guest Talks, Best Poster Prizes & Travel Awards > CRS Guest Lecturers > Larry Abbott
Larry Abbott is a physicist-turned biologist who uses mathematical modeling to study the neural networks that are responsible for our actions and behaviors. Abbott’s thesis work at Brandeis University was in the area of theoretical elementary particle physics, ulminating in a PhD in 1977. He then worked in theoretical particle physics at the Stanford Linear Accelerator Center and, later, at CERN, the European center for particle physics research. He became an assistant professor in the physics department at Brandeis in 1979, received tenure in 1982, and became a full professor of physics in 1988. His best-known achievements in particle physics include work on the cosmological constant, development of the background field method, calculations of the microwave background anisotropy, and work in gravity and gauge field theories.
Abbott began his transition to neuroscience research in 1989 and moved to the Biology Department at Brandeis in 1993. From 1994-2005, he was the co-director of the Sloan-Swartz Center for Theoretical Neuroscience at Brandeis, and from 1997-2002 was the director of the Volen Center. He held both the Nancy Lurie Marks and Zalman Abraham Kekst chairs in neuroscience. While at Brandeis, Abbott in collaboration with Eve Marder developed the dynamic clamp, a technique that has now become a standard tool of experimental electrophysiology. In 2005, Abbott joined the faculty of Columbia University where he is now the William Bloor Professor of Theoretical Neuroscience and co-director of the Center for Theoretical Neuroscience.
Vision Sciences Society Keynote Address 2007: The Interaction of Evoked and Spontaneous Activity in Visual Processing
In vivo recordings from primary visual cortex reveal that spontaneous background activity can be as complex as activity evoked by visual stimuli. Embedding visually evoked responses in such a strong and complex background seems like a confusing way to represent information about the visual world. However, modeling studies indicate that, contrary to intuition, information about visual stimuli may be better conveyed by a network displaying chaotic background activity than by a network without spontaneous activity.