Distinct patterns of cortical manifold expansion and contraction underlie human sensorimotor adaptation

Abstract

Sensorimotor learning is a dynamic, systems-level process that involves the combined action of multiple neural systems distributed across the brain. Although we understand a great deal about the specialized cortical systems that support specific components of action (such as reaching), we know less about how cortical systems function in a coordinated manner to facilitate adaptive behaviour. To address this gap in knowledge, our study measured human brain activity using functional MRI (fMRI) while participants performed a classic sensorimotor adaptation task, and used a manifold learning approach to describe how behavioural changes during adaptation relate to changes in the landscape of cortical activity. During early adaptation, we found that areas in parietal and premotor cortex exhibited significant contraction along the cortical manifold, which was associated with their increased covariance with regions in higher-order association cortex, including both the default mode and fronto-parietal networks. By contrast, during late adaptation, when visuomotor errors had been largely reduced, we observed a significant expansion of visual cortex along the cortical manifold, which was associated with its reduced covariance with association cortex and its increased intraconnectivity. Lastly, we found that individuals who learned more rapidly exhibited greater covariance between regions in the sensorimotor and association cortices during early adaptation. Together, these findings are consistent with a view that sensorimotor adaptation depends on changes in the integration and segregation of neural activity across more specialized regions of unimodal cortex with regions in association cortex implicated in higher-order processes. More generally, they lend support to an emerging line of evidence implicating regions of the default mode network in task-based performance.

Individual word representations dissociate from linguistic context along a cortical unimodal to heteromodal gradient

“…Our findings extend this evidence by demonstrating that brain responses along the gradient capture a wide range of linguistic features, and that the gradient explicitly dissociates individual word representations from contextual information during sentence reading. More generally, the finding that both contextual parameters were associated with the heteromodal end of the gradient in MEG is in line with previous fMRI studies beyond the language domain that implicated this organizational pattern in memory-guided cognition (Smallwood et al, 2021) across a broad range of functions (e.g., motor control, Gale et al, 2022;automated rulebased behaviour, Vatansever et al, 2017;working memory, Murphy et al, 2018;2019). The present MEG findings complement the previous fMRI evidence by showing that this organizational motif can impact on rapid features of information processing.…”

Whole-brain modular dynamics at rest predict sensorimotor learning performance

“…This approach contrasts with the more common one of measuring some aspect of pre-identified resting-state networks during task performance [e.g. Vatansever et al (43), Gale et al (44)]. Our approach is both philosophical and practical.…”