Time-resolved resting-state brain networks

Abstract

Neuronal dynamics display a complex spatiotemporal structure involving the precise, context-dependent coordination of activation patterns across a large number of spatially distributed regions. Functional magnetic resonance imaging (fMRI) has played a central role in demonstrating the nontrivial spatial and topological structure of these interactions, but thus far has been limited in its capacity to study their temporal evolution. Here, using high-resolution resting-state fMRI data obtained from the Human Connectome Project, we mapped time-resolved functional connectivity across the entire brain at a subsecond resolution with the aim of understanding how nonstationary fluctuations in pairwise interactions between regions relate to large-scale topological properties of the human brain. We report evidence for a consistent set of functional connections that show pronounced fluctuations in their strength over time. The most dynamic connections are intermodular, linking elements from topologically separable subsystems, and localize to known hubs of default mode and fronto-parietal systems. We found that spatially distributed regions spontaneously increased, for brief intervals, the efficiency with which they can transfer information, producing temporary, globally efficient network states. Our findings suggest that brain dynamics give rise to variations in complex network properties over time, possibly achieving a balance between efficient information-processing and metabolic expenditure.

Keywords: dynamic connectivity; network efficiency; time-dependent network.

Fig. 1.

Regions most consistently forming dynamic functional connections in the resting state. (A) Index of consistency for the actual data (blue line) and 250 null data sets (black lines). The 19 regions residing to the right of the P = 0.01 cutoff value (vertical red line) were consistently associated with dynamic behavior across 10 healthy, young adults. (B) Index of consistency rendered onto the cortical surface.

Fig. 2.

Dynamic fluctuations are synchronized across resting-state functional brain networks, occurring at distinct moments in time, where multiple connections transition en masse between high and low levels of connectivity. (A) Time series of correlation coefficients pertaining to the top-100 most dynamic functional connections for two healthy, young adults and a sample null data set. Percentages indicate the amount of variance explained by the first principal components (thick black lines) accounting for at least 20% of the variance. (B) The transition count (blue lines) enumerates for each time point the number of connections that cross their median correlation value. The transition count for a sample null data set is also shown (black lines). The null hypothesis of uniformly distributed transitions across time was rejected at time points when the 0.01 FWER cutoff value (horizontal red lines) was exceeded. Individuals are labeled according to six-digit HCP subject identifiers.

Fig. 3.

Time-resolved analysis of regional network efficiency shows that resting-state functional brain networks spontaneously reconfigure in such a way that multiple regions synchronously transition to high-efficiency states. (A) Regional network efficiency for two healthy, young adults; simulated rsfMRI data and a sample null data set. Matrix rows/columns represent regions/time. Efficiency range for simulated data is 0–0.25. (B) Regional efficiencies rendered onto the cortical surface for representative high- and low-efficiency states.

Fig. 4.

Dynamic functional connections are more likely to interconnect two distinct modules. (A) Time-averaged decomposition comprising four modules rendered on the cortical surface. (B) Test statistic values averaged over 10 healthy, young adults shown in matrix form. High test statistic values (yellow shades) provide greater evidence for dynamic (nonstationary) fluctuations. Low values indicate static connections. Row/columns are ordered such that regions composing the same module occupy consecutive rows/columns. Modules are delineated with thin black lines.