Representations and decodability of diverse cognitive functions are preserved across the human cortex, cerebellum, and subcortex

Abstract
Which part of the brain contributes to our complex cognitive processes? Studies have revealed contributions of the cerebellum and subcortex to higher-order cognitive functions; however, it has been unclear whether such functional representations are preserved across the cortex, cerebellum, and subcortex. In this study, we use functional magnetic resonance imaging data with 103 cognitive tasks and construct three voxel-wise encoding and decoding models independently using cortical, cerebellar, and subcortical voxels. Representational similarity analysis reveals that the structure of task representations is preserved across the three brain parts. Principal component analysis visualizes distinct organizations of abstract cognitive functions in each part of the cerebellum and subcortex. More than 90% of the cognitive tasks are decodable from the cerebellum and subcortical activities, even for the novel tasks not included in model training. Furthermore, we show that the cerebellum and subcortex have sufficient information to reconstruct activity in the cerebral cortex.
...
Results
...
Metadata-based interpretation of the task organizations in the cerebellum and subcortex
Although the similarity-based analyses in the previous section showed different representation patterns of 103 tasks in the cortex, cerebellum, and subcortex, it was unclear what cognitive factors contributed to these organizations. To interpret cognitive factors related to those tasks, we first performed principal component analysis (PCA) on the averaged weight matrix of the task-type model concatenated across six subjects. The resultant PCs were then associated with independent cognitive factors using a metadata-based reverse inference analysis. For each of the PC score maps, we calculated Pearson’s correlation coefficients with the 715 reverse inference maps taken from the Neurosynth database46. The top and bottom 10 terms of each PC provided their objective interpretations (Supplementary Tables 2, 3). For the sake of intelligibility, we only presented the results of the top five PCs.

For the cerebellum, PC1 was associated with introspection and emotion terms on the positive side (top 10 terms; “theory mind,” “disgust”) and executive function and motor terms on the negative side (bottom 10 terms; “working memory,” “motor”). Contrarily, PC2 was associated with executive function terms on the positive side (“working memory,” “execution”) and introspection terms on the negative side (“autobiographical,” “theory mind”). PC3 was associated with language terms on the positive side (“sentence,” “comprehension”) and motor terms on the negative side (“motor,” “finger”). PC4 was associated with motor terms on the positive side (“finger,” “sensorimotor”) and language terms on the negative side (“reading,” “linguistic”). PC5 was associated with executive function terms on the positive side (“cognitive task,” “execution”) and motor terms on the negative side (“finger,” “motor”).

For the subcortex, PC1 was associated with emotion terms on the positive side (“emotion,” “valence”), whereas it was associated with motor terms on the negative side (“finger,” “movement”). PC2 was associated with memory terms on the positive side (“memory,” “retrieval”) and somatosensory terms on the negative side (“pain,” “somatosensory”). PC3 was associated with motor terms on the positive side (“muscle,” “finger”) and emotion terms on the negative side (“pain,” “emotion”). PC4 was also associated with motor terms on the positive side (“preparation,” “motor”) and somatosensory terms on the negative side (“pain,” “somatosensory”). PC5 was associated with memory terms on the positive side (“retrieval,” “memory”) and motor terms on the negative side (“finger,” “sensorimotor”).
...





Link zur Studie bei nature.com