Investigating the functional heterogeneity of the default mode network using coordinate-based meta-analytic modeling

Abstract

The default mode network (DMN) comprises a set of regions that exhibit ongoing, intrinsic activity in the resting state and task-related decreases in activity across a range of paradigms. However, DMN regions have also been reported as task-related increases, either independently or coactivated with other regions in the network. Cognitive subtractions and the use of low-level baseline conditions have generally masked the functional nature of these regions. Using a combination of activation likelihood estimation, which assesses statistically significant convergence of neuroimaging results, and tools distributed with the BrainMap database, we identified core regions in the DMN and examined their functional heterogeneity. Meta-analytic coactivation maps of task-related increases were independently generated for each region, which included both within-DMN and non-DMN connections. Their functional properties were assessed using behavioral domain metadata in BrainMap. These results were integrated to determine a DMN connectivity model that represents the patterns of interactions observed in task-related increases in activity across diverse tasks. Subnetwork components of this model were identified, and behavioral domain analysis of these cliques yielded discrete functional properties, demonstrating that components of the DMN are differentially specialized. Affective and perceptual cliques of the DMN were identified, as well as the cliques associated with a reduced preference for motor processing. In summary, we used advanced coordinate-based meta-analysis techniques to explicate behavior and connectivity in the default mode network; future work will involve applying this analysis strategy to other modes of brain function, such as executive function or sensorimotor systems.

Figure 1.

Meta-analytic identification of regions in the DMN (p < 0.005, corrected). These results were used to define ROIs for the BD profile analyses and MACM. z values for each axial slice are reported in Talairach space (top panel), while x values denote sagittal slices (bottom panel).

Figure 2.

Behavioral Domain Profiles for the Top 3 Highly Concordant DMN Regions. BrainMap counts (blue histograms) represent values observed across the entire database, while DMN ROIs (green histograms) were restricted to the nodes defined in Figure 1. All values are normalized. A complete listing of significant domains for all regions is reported in Figure 3.

Figure 3.

Behavioral domain profiles for individual DMN regions. Significant behavioral domains are listed for each region in the default mode network. Black upward triangles indicate domains in which the observed regional number of experiments was higher than expected compared with the distribution across the BrainMap database, while gray downward triangles indicate domains that were lower than expected. The entire set of domains tested is available at http://brainmap.org/scribe.

Figure 4.

Composite image of the deactivations meta-analysis (blue) and MACM maps for each DMN region (red). Substantial overlap was observed for some regions (e.g., PCC and RMTG), while other regions showed minimal overlap (e.g., RIPL and vACC).

Figure 5.

Meta-analytic model of connectivity between DMN regions. MACM overlap between ROIs observed in Figure 4 was used to construct a model representing connectivity between regions. Directionality of paths indicates that an ROI was observed (ending point) in another ROIs MACM (starting point). Color-coding of cliques in the model was determined by BD analysis of sets of nodes comprising these subnetworks.

Figure 6.

Behavioral domain profiles for DMN subnetworks. Significant behavioral domains are listed for each DMN subnetwork. Black upward triangles indicate clique domain counts that were higher than expected as compared across the BrainMap database, while gray downward triangles indicate lower than expected domains. A–G depict each of the subnetworks identified in the analysis of DMN coactivations, extracted from the observed connections in Figure 5. A and B are subnetworks composed of 3 nodes, constrained by the direction of paths in the model; C and D include 3 nodes unconstrained by path direction; E includes 4 constrained nodes; F includes 4 unconstrained nodes; G includes 5 constrained nodes; and H includes 5 unconstrained nodes.