Functional mapping of the human brain: promise, problems and prospects
Jack Gallant
(Berkeley)
Date: May 8, 2017
Description:
When functional MRI was developed it was hoped that this new tool could help validate and refine existing psychological models. However, linking psychological theories to maps produced by functional imaging has been a challenging problem, for several reasons. First, the experimental and data analysis paradigms developed for use in experimental psychology are not optimal for obtaining data in functional imaging experiments. Second, while psychological theories are usually pitched at the computational level, functional imaging produces data at the algorithmic level. Finally, both experimental psychology and functional imaging have been limited by various data quality and quantity issues that have caused a loss of confidence amongst scientists and the public. In this talk I will review some of the efforts undertaken in my lab to try to address these difficult problems. Our innovations include a naturalistic approach to experimental design, a voxel-wise modeling framework that provides quantitative predictions of brain activity in every individual subject; new tools for data visualization, and new methods for aggregating data across individuals while minimizing information loss. Although our approach provides detailed functional maps of the cerebral cortex, these maps do not provide a clear bridge to the computational models proposed in experimental psychology. Bridging this gap is a fundamental challenge for the cognitive neuroscience community.
Created: Tuesday, May 9th, 2017