Brain-Network Temporal Variability in Resting-State fMRI: A Test-Retest Reliability Study

Brain-network temporal variability is a widely used metric for characterizing temporal fluctuations in the dynamic brain connectome. Despite its wide application, however, the reliability of this metric in estimating brain dynamics remains inadequately investigated. To fill such a gap, this study evaluated the test-retest reliability of temporal variability across four separate resting-state functional magnetic resonance imaging (fMRI) scans, as measured by the intraclass correlation coefficient (ICC), and explored potential influencing factors using a test-retest fMRI dataset from 337 healthy adults. The main findings are: (1) Using a step length of 40 seconds and a window width of 100 seconds for dynamic network construction with the sliding-window approach, temporal variability demonstrated at least moderate (ICC > 0.4) test-retest reliability across the whole brain and in most networks. (2) With other parameters fixed, the reliability was not significantly altered by different step lengths but decreased with longer window widths and shorter total fMRI scan durations. These results were generally consistent when using two different atlases across two randomly split subsamples. Our results indicate that temporal variability is a relatively reliable metric for identifying inter-subject differences in brain dynamics. Nevertheless, moderate window widths and sufficient scan durations are necessary for producing reliable results.