Neural mechanisms underlying loss of functional independence after stroke: EEG evidence of impaired motor-to-sensory directed connectivity on the affected side

Post-stroke reorganization of the prefrontal-sensorimotor network is critical for functional recovery, yet single connectivity metrics fail to capture its multidimensional characteristics. This exploratory study investigates network reorganization patterns and their clinical relevance to motor function and daily independence using multimodal electroencephalography (EEG) connectivity analysis, including Coherence (COH), phase lag index (PLI), and Granger causality (GC). Resting-state EEG from stroke patients ( n = 22) and healthy controls ( n = 22) was analyzed. COH, PLI, and GC metrics were computed for key regional connections involving the prefrontal (PFC), motor (MC), sensory (SC), and parietal (PC) cortices and correlated with Fugl-Meyer Assessment for Upper Extremity (FMA-UE) and Modified Barthel Index(MBI) scores. For stroke patients, hemispheres were classified as ipsilesional (i, affected) and contralesional (c, unaffected) based on the lesion side. Compared to healthy controls, stroke patients exhibited distinct reorganization patterns. Preliminary evidence suggests enhanced parietal-motor coherence (COH_cPC-cMC↑) in the unaffected hemisphere alongside reduced motor-to-prefrontal directed connectivity (GC_cMC→cPFC↓). Interhemispherically, the findings may indicate weakened information flow from the unaffected to affected motor/sensory areas (GC_cMC→iMC↓, GC_cMC→iSC↓), accompanied by reduced bilateral motor synchronization (PLI_iMC-cMC↓). Within the affected hemisphere, there was evidence of enhanced bidirectional parietal-motor phase transfer (PLI_iPC-iMC↑), whereas directed pathways were significantly impaired (GC_iMC→iSC↓, GC_iPC→iSC↓, GC_iPFC→iMC↓). Clinically, the affected motor-to-sensory pathway (GC_iMC→iSC) showed a strong negative correlation with the MBI ( r = -0.61, P = 0.002), and both interhemispheric motor regulation (GC_cMC→iMC) and synchronization (PLI_iMC-cMC) were significantly correlated with functional deficits ( P < 0.05). Post-stroke networks may exhibit a “local compensation-global impairment” reorganization pattern. The affected motor-sensory pathway (GC_iMC→iSC) may be a biomarker for functional independence and thus allow for guided sensory integration-based precision rehabilitation.