Title

Computation through dynamics and its application to the neurobiology of time

Bio

Mehrdad Jazayeri is an Associate Professor at MIT. He is also the Director of Education at the Department of Brain and Cognitive Sciences and an Investigator at the McGovern Institute for Brain Research. He received his Bachelor’s degree in electrical engineering majoring in telecommunications from the Sharif University of Technology in Tehran, Iran. He then completed his Master’s degree in physiology at the University of Toronto, and his Ph.D. in neuroscience at New York University.

Abstract

I will introduce a framework we refer to as “Computation through Dynamics” that has guided our research over the past few years for linking the structure and dynamics of neural activity to behavior. This framework is guided by two key ideas. First, we examine the geometry and structure of neural activity at the appropriate level of abstraction to understand how the brain creates invariant representations needed for flexible cognitive computations. Second, we examine the organization of the latent dynamics (fixed points, vector fields, etc) that inputs to and connections between neurons establish to understand how the brain exerts cognitive control over structured representations. I will highlight the application of this framework in one line of research in my lab focused on the neural basis of timing.