Reduced dynamic functional connectivity in older ages: are older brains less adaptable?

Abstract

Understanding how brain connectivity reorganizes with age is essential for characterizing healthy aging. While static functional connectivity (sFC) has revealed broad age-related shifts in network segregation and integration, recent work underscores the need to examine how these patterns dynamically fluctuate over time to better understand cognitive decline and resilience in aging. The present resting-state fMRI study was conducted to enhance comprehension of the temporal variability and dynamics of the brain's functional architecture in the context of the aging process. To this end, dynamic functional connectivity (dFC) was extracted in 817 older adults between 55 and 85 years from the 1000BRAINS study. Using a sliding window and clustering approach, we identified four recurring dFC states and quantified temporal metrics, e.g., mean dwell time, or number of transitions. Overall, aging was associated with slower and less flexible network dynamics: integrative states became less frequent, whereas segregated states dominated, reflecting reduced inter-network communication, although the effect size was relatively small. Age-stratified analyses yielded two novel insights, thereby suggesting a refinement of the so far established theories of aging and giving rise to a new model of age-related differences in functional connectivity: First, trajectories of dedifferentiation and segregation shown from younger to older adults seem to stabilize during the transition from mid-to-old age and turn into a process of re-segregation and overcompensation during older-old ages, challenging the assumption of a monotonic increase in integration across the lifespan. Secondly, opposing links between dFC and cognitive performance have been identified, with greater network integration supporting better cognitive performance in mid-to-old adults but poorer cognitive performance in older-old adults. This suggests a shift from dedifferentiated to re-segregated connectivity with advancing age. Sex-stratified modeling further demonstrated stronger age-related reductions in flexibility among females, indicating that pooled analyses might obscure systematic sex-specific dynamics and highlighting divergent adaptive and vulnerability profiles across aging. The findings of this study emphasize the necessity of accounting for temporal variability in age-related changes of functional brain connectivity when studying aging, specifically during advanced age, with the aim of enhancing comprehension of cognitive variability.