Neurofeedback-Induced Network Plasticity in Motor and Default Mode Networks Correlates with Motor Recovery After Stroke

Stroke often results in long-term motor impairements, necessitating innovative rehabilitation strategies. Neurofeedback (NF) has emerged as a promising tool to promote functional recovery after stroke by enabling patients to modulate targeted brain activity. In this study, we investigated how NF training influences whole-brain functional connectivity in chronic stroke patients using a data-driven, network-level approach. The analysis was conducted on the same dataset as a previously reported clinical study, focusing here on the effects of NF through the lens of functional connectivity. Thirty chronic stroke patients underwent either multimodal NF training (combining EEG-fMRI and EEG-only feedback targeting motor regions) or a matched motor imagery (MI) control condition without feedback. Pre- and post-intervention fMRI data were analyzed using Network-Based Statistics (NBS) to identify distributed changes in connectivity. NF training led to significant reductions in connectivity within motor networks and the default mode network (DMN), while no significant effects were found in the control group. These connectivity reductions, particularly in contralesional motor network, were significantly correlated with motor function improvement. These findings highlight the relevance of network-level mechanisms in NF-induced plasticity and support the development of connectivity-informed NF strategies in stroke rehabilitation.