Altered brain-ventricle coupling modes over Alzheimer's disease progression detected with fMRI

Despite showing significant impact in cognitive preservation, the relationship between brain activity captured with functional Magnetic Resonance Imaging (fMRI) in gray matter and ventricular cerebrospinal fluid dynamics remains poorly understood. We analyzed 599 fMRI scans from 163 elderly participants at rest with varying degrees of cognitive impairment employing a unified phase coupling analysis that breaks from convention by incorporating both tissue and ventricular signal fluctuations. This whole-brain approach identified distinct brain-ventricle coupling modes that differentiate between cognitive status groups and correlate with specific cognitive abilities. Beyond the previously reported anti-phase coupling between global brain signals and ventricles--which we confirm occurs more frequently in cognitively normal controls--our analysis method uncovered additional coupling modes where signals in specific brain networks temporarily align with ventricle signals. At the cortical level, these modes reveal patterns corresponding to known resting-state networks: one overlapping with the Default Mode Network occurs significantly less frequently in Alzheimers Disease patients, while another revealing the Frontoparietal Network correlates positively with memory scores. Our findings demonstrate that different brain-ventricle coupling modes correlate with specific cognitive domains, with particular modes predicting memory, executive function, and visuospatial abilities. The coupling between signals in brain ventricles and established resting-state networks challenges our current understanding of functional network formation, suggesting an integral link with brain fluid motion. This reconceptualization of brain dynamics through the lens of fluid-tissue interactions establishes a fundamental physical basis for cognitive preservation, suggesting that therapeutic interventions targeting these interactions may prove more effective than approaches focused solely on cellular or molecular mechanisms.