Jack Gallant, UC Berkeley

A functional atlas of the human brain that predicts activity under naturalistic conditions

The structure and function of the human brain are very closely linked. Neuroanatomists have made much progress in obtaining good anatomical parcellations of the human brain and recent technical advances might make it possible to recover detailed structural topography and structural connectivity from living individuals. In contrast, cognitive neuroscientists have made much less progress in obtaining a detailed functional parcellation of the human brain, and most of the studies that have attempted to do this have been based on passive resting state correlations that have no clear functional assignment. My laboratory is engaged in a broad effort to obtain obtain a detailed functional parcellation of the human brain that reveals how sensory and cognitive features are represented across the cortical mantle; to understand how these representations are modulated by top-down processes such as attention, learning and memory; and to determine the correspondence between functional and structural neuroanatomy. In this talk I will discuss our approach to this problem, and I will provide a few clear examples showing of how this approach reveals detailed functional maps in individual brains. Finally, I will show how these results suggest that we should revise several commonly held beliefs about the functional organization and the operation of the human brain.

This talk is part of the BCCN Berlin-organised mini-symposium Artifical Intelligence and the Brain during the Berlin Science Week (Nov 1-10, 2016).

Additional Information

Jack Gallant is Chancellor's Professor of Psychology at the University of California at Berkeley, and he is affiliated with a variety of other departments and graduate programs. He received his Ph.D. from Yale University, and he did post-doctoral work at the California Institute of Technology and Washington University Medical School. He is known for his neurophysiological work on the representation of natural scenes, the function of area V4 and its modulation by attention, and for the development of the voxel-wise modeling approach in human fMRI. His current research program focuses on computational modeling and mapping of human brain activity. These models accurately describe how the brain encodes information during complex, naturalistic tasks, and they show how information about the external and internal worlds are mapped systematically across the surface of the cerebral cortex. These models can also be used to decode information in the brain in order to reconstruct mental experiences. Further information about ongoing work, links to talks and papers and links to an online interactive brain viewer can be found at the lab web page http://gallantlab.org.

Organized by

Raphael Holca / Robert Martin