网络连接靶向的近红外神经反馈对工作记忆重度衰退的干预效果与机制OACHSSCD

Systemic deviation of large-scale brain network coordination patterns is the core neural mechanism of working memory aging,and the severe decline in working memory significantly impairs older individuals'quality of life while imposing heavy burdens on society.Existing neuromodulation techniques are either limited to single-point interventions or simplify network feedback signals,making it difficult to reflect the holistic profile of multi-node dynamic coordination.Furthermore,constrained by factors such as equipment safety and portability,current methods fail to meet the long-term intervention needs of the elderly.This study aims to construct a connectome-targeted fNIRS neurofeedback intervention framework:First,multivariate pattern analysis(MVPA)is employed to identify specific network connectivity patterns that significantly distinguish severely-declined groups from function-maintained groups for network target setting.Second,using the spatial similarity between an individual's real-time network connectivity pattern(retaining topological properties)and the target pattern as the feedback index,a closed-loop neurofeedback system is constructed to guide autonomous modulation and systematically evaluate its intervention efficacy.Finally,the performance characteristics of cognitive transfer effects are comprehensively examined across multiple dimensions,and the potential dual-pathway transfer mechanism—comprising"network direct-driven"and"working memory-mediated"pathways—is verified through mediation analysis.This research is expected to provide a novel intervention scheme for improving severe working memory decline and offer a theoretical framework and methodological reference for network-targeted modulation strategies.