Age-related differences in test-retest reliability in resting-state brain functional connectivity
Jie Song1, Alok S Desphande, Timothy B Meier
1Department of Biomedical Engineering, University of Wisconsin-Madison, Madison, Wisconsin, United States of America.
Plos One
|December 11, 2012
Summary
Older adults exhibit reduced resting-state functional connectivity reliability compared to younger adults. Global signal regression impacts reliability and stability differently across age groups.
Area of Science:
- Neuroscience
- Neuroimaging
- Brain Connectivity
Background:
- Resting-state functional MRI (rs-fMRI) is crucial for studying brain functional connectivity (FC).
- Previous research focused on FC reliability in younger adults, leaving older adults understudied.
- Test-retest reliability of resting-state functional connectivity (RSFC) in older adults remains unexamined.
Purpose of the Study:
- To investigate age-related differences in RSFC reliability and stability.
- To examine the impact of global signal regression (GSR) on RSFC reliability and stability.
- To compare RSFC reliability between younger and older adults.
Main Methods:
- Utilized rs-fMRI data from 29 younger and 26 older adults from the ICBM dataset.
- Defined 92 regions of interest (ROIs) across key brain networks.
- Calculated RSFC using z-transformed correlation coefficients from three scans per subject, with and without GSR.
Main Results:
- Younger adults demonstrated significantly higher RSFC reliability than older adults, both with and without GSR.
- Both age groups exhibited high multi-scan stability of RSFC, with no significant age-related differences.
- GSR reduced overall RSFC reliability but increased stability in both groups, potentially affecting group comparisons.
Conclusions:
- Aging is associated with reduced RSFC reliability, though within-subject stability remains high across scans.
- GSR influences RSFC reliability and stability differently across age groups.
- Findings highlight the importance of considering age and preprocessing methods when interpreting RSFC findings.


