Longitudinal functional brain network reconfiguration in healthy aging

• Published in: Human brain mapping Vol. 41; no. 17; pp. 4829 - 4845
• Main Authors: Malagurski, Brigitta, Liem, Franziskus, Oschwald, Jessica, Mérillat, Susan, Jäncke, Lutz
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.12.2020
• Subjects: Aging; Brain; Brain architecture; Brain mapping; brain networks; Cognitive ability; Cognitive tasks; Cross-sectional studies; Flexibility; Functional magnetic resonance imaging; Functional morphology; Geriatrics; healthy aging; Heterogeneity; Modular structures; Modularity; multilayer modularity; Multilayers; network flexibility; Older people; Reconfiguration; resting‐state fMRI; Variability; network flexibility; brain networks; healthy aging; resting-state fMRI; multilayer modularity
• DOI: 10.1002/hbm.25161
• ISSN: 1065-9471; 1097-0193

Healthy aging is associated with changes in cognitive performance and functional brain organization. In fact, cross‐sectional studies imply lower modularity and significant heterogeneity in modular architecture across older subjects. Here, we used a longitudinal dataset consisting of four occasions of resting‐state‐fMRI and cognitive testing (spanning 4 years) in 150 healthy older adults. We applied a graph‐theoretic analysis to investigate the time‐evolving modular structure of the whole‐brain network, by maximizing the multilayer modularity across four time points. Global flexibility, which reflects the tendency of brain nodes to switch between modules across time, was significantly higher in healthy elderly than in a temporal null model. Further, global flexibility, as well as network‐specific flexibility of the default mode, frontoparietal control, and somatomotor networks, were significantly associated with age at baseline. These results indicate that older age is related to higher variability in modular organization. The temporal metrics were not associated with simultaneous changes in processing speed or learning performance in the context of memory encoding. Finally, this approach provides global indices for longitudinal change across a given time span and it may contribute to uncovering patterns of modular variability in healthy and clinical aging populations. In this research paper, we used a longitudinal dataset consisting of four occasions of resting‐state‐fMRI and cognitive testing (spanning 4 years) in 150 healthy older adults. We applied a graph‐theoretic analysis to investigate the time‐evolving modular structure of the whole‐brain network, by maximizing the multilayer modularity across four time points. Network flexibility, which reflects the tendency of brain nodes to switch between modules across time, was significantly higher in older adults than in a null model and with increasing age, indicating that older age is related to increased variability in modular organization.