Convergent functional architecture of the superior parietal lobule unraveled with multimodal neuroimaging approaches

Abstract

The superior parietal lobule (SPL) plays a pivotal role in many cognitive, perceptive, and motor-related processes. This implies that a mosaic of distinct functional and structural subregions may exist in this area. Recent studies have demonstrated that the ongoing spontaneous fluctuations in the brain at rest are highly structured and, like coactivation patterns, reflect the integration of cortical locations into long-distance networks. This suggests that the internal differentiation of a complex brain region may be revealed by interaction patterns that are reflected in different neuroimaging modalities. On the basis of this perspective, we aimed to identify a convergent functional organization of the SPL using multimodal neuroimaging approaches. The SPL was first parcellated based on its structural connections as well as on its resting-state connectivity and coactivation patterns. Then, post hoc functional characterizations and connectivity analyses were performed for each subregion. The three types of connectivity-based parcellations consistently identified five subregions in the SPL of each hemisphere. The two anterior subregions were found to be primarily involved in action processes and in visually guided visuomotor functions, whereas the three posterior subregions were primarily associated with visual perception, spatial cognition, reasoning, working memory, and attention. This parcellation scheme for the SPL was further supported by revealing distinct connectivity patterns for each subregion in all the used modalities. These results thus indicate a convergent functional architecture of the SPL that can be revealed based on different types of connectivity and is reflected by different functions and interactions.

Keywords: behavioral domains analyses; coactivation; functional connectivity; parcellation; structural connectivity.

Figure 1

Superior parietal lobule (SPL) parcellation results using multimodal neuroimaging methods and selection of the optimal number of SPL subregions. (A) The degree of overlap between the structural, resting‐state functional, and coactivation connectivity‐based parcellation results of the left (SPL) was calculated for each number of clusters using the generalized Dice coefficient. (B) The degree of overlap between the structural, resting‐state functional, and coactivation connectivity‐based parcellation results of the right SPL was computed as in A. (C) The maximum probability maps for the SPL subregions were obtained using structural and resting‐state functional connectivity‐based parcellation in the first two column. The third and last column showed the SPL parcellation result obtained on the basis of a coactivation connectivity‐based parcellation and the overlap between the parcellation results across the different modalities. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 2

The overlap with the cytoarchitectonic map of the superior parietal lobule (SPL). (A) The maximum probability map for each SPL subregion as defined using cytoarchitecture and extracted using the SPM Anatomy Toolbox. (B) The overlap between the cytoarchitectonic map of each SPL subregion and the corresponding SPL subregion derived using the different neuroimaging modalities of connectivity‐based parcellation as calculated for each hemisphere. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 3

Whole‐brain structural, resting‐state functional, and coactivation connectivity patterns for each subregion. Whole‐brain population maps of the probabilistic tractography results for each subregion of the SPL. The main tract pathways include the SLF, extreme capsule (EmC), and corpus callosum (CC). Whole‐brain resting‐state functional connectivity patterns for each cluster were obtained using one sample t‐tests (thresholded at P < 0.05, cluster‐level FEW‐corrected, cluster‐forming threshold at voxel‐level P < 0.001). The whole‐brain coactivation connectivity pattern for each subregion of the SPL was obtained using meta‐analytical connectivity modeling (MACM) analyses (thresholded at P < 0.05, cluster‐level FEW‐corrected, cluster‐forming threshold at voxel‐level P < 0.001). [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 4

Specific resting‐state functional connectivity pattern of each SPL subregion. Regions show significantly more resting‐state connectivity with a given cluster than with any of the other four clusters. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 5

Specific coactivation connectivity pattern of each SPL subregion. Regions significantly more coactivated with a given subregion than with any of the other subregions. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 6

Overlapping connectivity between resting‐state functional and coactivation connectivities. The intersection connectivity was calculated with whole‐brain resting‐state functional and coactivation connectivities. We first obtained thresholded whole‐brain resting‐state functional connectivity and coactivation connectivity maps for each SPL subregion and then computed the intersection connectivity between the two modalities. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 7

The Pearson cross‐correlation analyses between the resting‐state network and the coactivation network for each SPL subregion on the basis of resting‐state fMRI data. First, the resting‐state functional network and the coactivation network of each SPL subregion were established. Then, the Pearson correlation coefficient between the resting‐state network and coactivation network was calculated.

Figure 8

Behavioral domains and paradigm classes of the left SPL subregions. Forward inference and reverse inference were used to determine the functional organization of each subregion. The significant activation probabilities for each subregion with respect to a given domain or paradigm and the significant probability of a domain's or paradigm's occurrence given activation in a cluster are depicted separately. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 9

Behavioral domains and paradigm classes of the right SPL subregions. Forward inference and reverse inference were used to determine the functional organization of each subregion. The significant activation probabilities for each subregion with respect to a given domain or paradigm and the significant probability of a domain's or paradigm's occurrence given activation in a cluster are depicted separately. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]

Figure 10

Summary of the parcellation schemes for the superior parietal lobule (SPL). Scheperjans et al. [2008] parcellated the SPL into different subregions on the basis of different cytoarchitectonic properties. Nelson et al. [2010] applied resting‐state functional connectivity to subdivide the left lateral parietal cortex (LLPC) into different parts and identified the SPL. Subsequently, Barnes et al. [2010] used similar procedures to parcellate the LLPC and identified two similar subregions in the SPL in both adults and children. Mars et al. [2011] parcellated the right parietal cortex into subregions, identifying five subregions in the SPL. With permission. [Color figure can be viewed in the online issue, which is available at http://wileyonlinelibrary.com.]