Send to

Choose Destination
Front Aging Neurosci. 2018 Jan 4;9:419. doi: 10.3389/fnagi.2017.00419. eCollection 2017.

Aging and Network Properties: Stability Over Time and Links with Learning during Working Memory Training.

Author information

Department of Psychology, University of Michigan, Ann Arbor, MI, United States.
Department of Mental Health, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, United States.
Department of Human Development and Family Science, Virginia Tech, Blacksburg, VA, United States.
MIND Research Institute, Irvine, CA, United States.
School of Education, University of California, Irvine, Irvine, CA, United States.
Functional MRI Laboratory, Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, United States.


Growing evidence suggests that healthy aging affects the configuration of large-scale functional brain networks. This includes reducing network modularity and local efficiency. However, the stability of these effects over time and their potential role in learning remain poorly understood. The goal of the present study was to further clarify previously reported age effects on "resting-state" networks, to test their reliability over time, and to assess their relation to subsequent learning during training. Resting-state fMRI data from 23 young (YA) and 20 older adults (OA) were acquired in 2 sessions 2 weeks apart. Graph-theoretic analyses identified both consistencies in network structure and differences in module composition between YA and OA, suggesting topological changes and less stability of functional network configuration with aging. Brain-wide, OA showed lower modularity and local efficiency compared to YA, consistent with the idea of age-related functional dedifferentiation, and these effects were replicable over time. At the level of individual networks, OA consistently showed greater participation and lower local efficiency and within-network connectivity in the cingulo-opercular network, as well as lower intra-network connectivity in the default-mode network and greater participation of the somato-sensorimotor network, suggesting age-related differential effects at the level of specialized brain modules. Finally, brain-wide network properties showed associations, albeit limited, with learning rates, as assessed with 10 days of computerized working memory training administered after the resting-state sessions, suggesting that baseline network configuration may influence subsequent learning outcomes. Identification of neural mechanisms associated with learning-induced plasticity is important for further clarifying whether and how such changes predict the magnitude and maintenance of training gains, as well as the extent and limits of cognitive transfer in both younger and older adults.


functional connectivity; graph theory; intraclass correlation; intrinsic activity; reliability analysis

Supplemental Content

Full text links

Icon for Frontiers Media SA Icon for PubMed Central
Loading ...
Support Center