Abstract
Background/Objectives: Amnestic mild cognitive impairment (aMCI) represents a transitional stage between normal aging and dementia, constituting a critical intervention window for Alzheimer's disease (AD). As a non-invasive intervention, neurofeedback training (NFT) has demonstrated potential in ameliorating cognitive deficits and clinical symptoms in aMCI patients; however, its mechanistic effects on functional brain connectivity remain inadequately elucidated. Methods: This study employed low- and high-order functional analytical approaches to comprehensively investigate the effects of NFT on dynamic brain functional networks in aMCI. Results: Our findings revealed that following NFT, aMCI patients exhibited enhanced connectivity strength, global efficiency, and nodal characteristics within the delta band, whereas connectivity was generally attenuated in the theta, alpha, and beta bands. Dynamic network analysis indicated increased entropy in short-time windows. Cognitive assessments showed a significant short-term improvement in MoCA scores among 92.9% of participants. Conclusions: These results suggest that NFT effectively remodels brain network activity patterns in aMCI patients, thereby facilitating cognitive improvement. These findings provide preliminary insights into the brain network mechanisms underlying NFT-mediated cognitive enhancement in aMCI.