Altered excitation-inhibition balance in the somatomotor and default mode network in multiple sclerosis

Excitation-Inhibition (E/I) balance is a system-level principle governing healthy neuronal information processing. Its disruption has been implicated in the pathophysiology of multiple sclerosis (MS), yet in vivo characterization of E/I balance in MS remains under debate. In MS, altered neural dynamics are thought to reflect widespread structural damage and functional reorganization. Previous studies have suggested E/I imbalance in MS, but available methods require electrophysiology and do not allow investigation using functional MRI (fMRI). As a result, E/I alterations in MS cannot currently be quantified from the large body of existing MRI data. Here, we show that E/I alterations in MS can be captured with multimodal MRI data. We employ a recent multimodal MRI-based neurocomputational framework to infer E/I balance, integrating structural and functional information. We find that people with MS have significantly lower E/I balance during resting state, in the somatomotor and default mode networks as defined from a functionally-based atlas. Furthermore, we discover a significant negative correlation between the E/I reductions in the somatomotor network and motor fatigue, linking inferred circuit dynamics to clinical parameters. These findings reveal that E/I imbalance in MS can be captured using fMRI-informed modeling, extending previous electrophysiological observations. They provide converging evidence that circuit-level dysregulation underlies clinical symptoms. By building on a framework previously validated in Alzheimer’s disease, our results support the robustness of this approach across disorders. This work establishes a scalable and non-invasive strategy to probe E/I balance in MS. It opens a new avenue to exploit the large body of available fMRI data to advance our understanding of circuit dysfunction in the disease.