For decades, brain imaging research has focused almost exclusively on gray matter as the seat of cognition and mental health. A growing body of evidence now suggests that white matter — long treated as mere wiring — carries its own functional signals that independently encode critical information about how the brain ages, thinks, and breaks down. This reframing has real implications for how clinicians and researchers evaluate brain health.
Published in PNAS, this multi-cohort study demonstrates that functional connectivity (FC) derived from white-matter signals predicts brain age, cognitive performance, and psychopathological risk over and above what gray-matter FC alone captures. Crucially, the white-matter contribution was not redundant — even after accounting for gray-matter connectivity, white-matter FC explained unique variance across all three outcomes. The findings held across independent cohorts, strengthening confidence in their generalizability beyond any single dataset or scanner environment.
This work matters because brain age — the estimated biological age of the brain relative to chronological age — has emerged as a powerful biomarker for neurodegenerative risk and overall healthspan. If white-matter FC adds predictive power to brain-age models, it means current models that ignore white matter are systematically underspecified. From a broader research perspective, this is potentially paradigm-shifting: it challenges the longstanding assumption that white-matter fMRI signals are noise or artifact rather than meaningful biological information. However, important caveats apply. The mechanisms linking white-matter FC to cognition remain poorly understood — it is unclear whether these signals reflect myelin integrity, axonal activity, or glymphatic processes. The study is also observational and cross-sectional, precluding causal inference. Whether white-matter FC changes precede cognitive decline or merely co-vary with it is unresolved. Still, as an incremental but meaningful advance in functional neuroimaging methodology, this finding invites a wholesale reassessment of what brain connectivity maps actually capture.