Default Mode Network Segregation Decreases in Healthy Brain Aging.

The default mode network (DMN) constitutes a cluster of interconnected brain regions engaged in internal cognitive processes, such as interoception, memory recall, and self-referential thinking. The DMN is a task-negative network, displaying increased activity during rest but reduced activity during cognitive tasks. Resting-state functional magnetic resonance imaging (rs-MRI) examines this task-negative nature by tracking low-frequency fluctuations in the blood oxygenation level-dependent signal as an indirect marker for changes in brain activity. Many cross-sectional studies have focused on understanding DMN organization in healthy brain aging, however, these studies fail to account for intra-individual variance. To address this limitation, we conducted a longitudinal, pilot analysis that investigated changes in DMN segregation index ($SI$), simply defined as the ratio of within-network connectivity relative to between-network connectivity, across the healthy adult lifespan. Specifically, a subset of the Calgary Normative Study data (67 healthy individuals scanned twice over an interval of 3.1 ± 0.3 years, mean ± standard deviation) was analyzed. A linear mixed effects model was used to examine the effects of $Age,Age^{2}$, and Sex, as well as Age $\times$ Sex interactions on DMN SI. We observed a significant decline in DMN $SI$ with $\operatorname{Age}^{2}(t_{128}=-2.79,\ p\leq 0.001)$. Males demonstrated reduced DMN segregation compared to females $(t_{126}=-2.53,p=0.01$). By identifying this decline in DMN $SI$, our study advances the understanding of age-and sex-associated changes in the DMN architecture, laying the groundwork for future work. In conclusion, our study confirms the significance of changes in the DMN in healthy brain aging using longitudinal analysis. Using the $SI$ metric strengthened our investigation by combining within- and between-network connectivity into a more robust metric that reflects complex and dynamic brain changes.