Causal inference and functional dynamics of a visuomotor network demonstrate excitatory/inhibitory alterations in Multiple Sclerosis

Balanced excitation and inhibition are fundamental for stable brain dynamics. In multiple sclerosis (MS), this balance is thought to be altered, yet its impact on the visuomotor network function and disease progression remains unclear. This study uses information from Dynamic Causal Modelling (DCM) and The Virtual Brain (TVB) alongside clinical measures to provide a causal and network-level perspective on MS pathophysiology. We defined a visuomotor network based on previous work, including bilateral primary visual cortex (V1), left primary motor cortex (M1), left supplementary motor and premotor cortex (SMAPMC), left cingulate cortex (CC), left superior parietal lobule (SPL), and right cerebellar lobule VI (CR). We analysed data from 9 healthy volunteers (HV) and 17 people with MS (pwMS) who underwent multimodal MRI including a visually-guided variable grip-force event-related task and resting-state functional MRI, as well as diffusion-weighted imaging. Our DCM analysis indicated that the best-fitting model was the same for both groups, suggesting preserved visuomotor network architecture. However, pwMS showed altered effective connectivity during task (from SPL to M1: excitatory in HV, inhibitory in pwMS) and rest (from V1 to M1: excitatory in HV, inhibitory in pwMS; from SPL to M1, from M1 to SMAPMC, and from CR to V1: inhibitory in HV, excitatory in pwMS). During task execution, grip-force modulations revealed altered feedback connectivity patterns in pwMS as motor demand increased, indicating a shift from the positive coupling seen in HV to negative coupling across most connections. Effective connectivity alterations in task and resting-state fMRI were associated with clinical measures, including disability and motor reaction time. Connectivity changes involved visuomotor and cerebellar circuits and differed between task execution and rest, indicating context-specific network dynamics. Network-level simulations further linked reduced excitatory gain to slower responses, showing the role of excitation/inhibition balance in performing a visuomotor task. Together, these findings identify excitatory and inhibitory alterations as key mechanisms of network functional disruption in MS and demonstrate the potential of multiscale modelling to bridge neurophysiological alterations with clinical performance.