Our Research Project

Flexible encoding of multiple task dimensions in human cerebral cortex

Introduction: Cognitive models have proposed that behavioral tasks can be categorized along at least three dimensions: the sensorymotor modality of the information, its representational format (e.g., location vs. identity), and the cognitive processes that transform it (e.g., response selection). Moreover, we can quickly and flexibly encode, represent, or manipulate information along any of these dimensions. How is this flexibility in encoding such information implemented in the cerebral cortex?
Methods: To address this question, we devised a series of functional magnetic resonance imaging (fMRI) experiments in each of which participants performed two distinct tasks that di ered along one of the three dimensions.
Results: Using multivariate pattern analysis of the fMRI data, we were able to decode between tasks along at least one task dimension within each of the cortical regions activated by these tasks. Moreover, the multiple demand network, a system of brain regions previously associated with flexible task encoding, was largely composed of closely juxtaposed sets of voxels that were specialized along each of the three tested task dimensions.
Discussion: These results suggest that flexible task encoding is primarily achieved by the juxtaposition of specialized representations processing each task dimension in the multiple demand network.

  • Tamber-Rosenau, B. J., Newton, A. T., & Marois, R. (2024). Flexible encoding of multiple task dimensions in human cerebral cortex. Frontiers in Cognition, 3, 1438390. doi: 10.3389/fcogn.2024.1438390. [PDF]

 

fig4
FIGURE 4. Overlap of searchlight decoding of process, format, and modality. Results projected onto the inflated surface of the Talairach transformed COLIN brain. Each experiment is thresholded at false discovery rate q < . . White outlines represent the MD network. We emphasize that the brown color represents areas of overlap of all dimensions of decoding; meaning that these locations had distinct decoding patterns (as detected via searchlight MVPA) for the di erent conditions of each task dimension: process, format, or modality. In other words, these regions contain the greatest heterogeneity of information decoding across the tested dimensions. These maps are best understood as complementary to those in Figure , which depict decoding for each individual dimension (Tamber-Rosenau et al., 2024).