|
A common model of representational spaces in human cortex| old_uid | 13300 |
|---|
| title | A common model of representational spaces in human cortex |
|---|
| start_date | 2017/02/27 |
|---|
| schedule | 11h |
|---|
| online | no |
|---|
| summary | Multivariate pattern analysis affords investigation of fine-grained patterns of neural activity that carry fine-grained distinctions in the information they represent. These patterns of brain activity in different brains can be recast as vectors in a common high-dimensional representational space with basis functions that have tuning profiles and patterns of connectivity that are common across brains. We derive transformation matrices that rotate individual anatomical spaces into the common model space with searchlight-based, whole cortex hyperalignment. Transformation matrices can be derived based on patterns of response to a rich, naturalistic stimulus, such as a movie, or on patterns of functional connectivity. Basing hyperalignment on functional connectivity makes it possible to hyperalign brains based on fMRI data obtained in the resting state as well as during movie viewing. The common model provides a common structure that captures fine-grained distinctions among cortical patterns of response that are not modeled well by current brain atlases. The model also captures coarse-scale features of cortical topography, such as retinotopy and category-selectivity, and provides a computational account for both coarse-scale and fine-scale topographies with multiplexed topographic basis functions. |
|---|
| responsibles | Blancho |
|---|
| |
|