Connectopic mapping with resting-state fMRI

Abstract

Brain regions are often topographically connected: nearby locations within one brain area connect with nearby locations in another area. Mapping these connection topographies, or 'connectopies' in short, is crucial for understanding how information is processed in the brain. Here, we propose principled, fully data-driven methods for mapping connectopies using functional magnetic resonance imaging (fMRI) data acquired at rest by combining spectral embedding of voxel-wise connectivity 'fingerprints' with a novel approach to spatial statistical inference. We apply the approach in human primary motor and visual cortex, and show that it can trace biologically plausible, overlapping connectopies in individual subjects that follow these regions' somatotopic and retinotopic maps. As a generic mechanism to perform inference over connectopies, the new spatial statistics approach enables rigorous statistical testing of hypotheses regarding the fine-grained spatial profile of functional connectivity and whether that profile is different between subjects or between experimental conditions. The combined framework offers a fundamental alternative to existing approaches to investigating functional connectivity in the brain, from voxel- or seed-pair wise characterizations of functional association, towards a full, multivariate characterization of spatial topography.

Situating the default-mode network along a principal gradient of macroscale cortical organization

“…Rather than delineating discrete network parcellations, we implemented a method that captures gradients in connectivity patterns over space-a cortical feature termed "connectopies" (41). This method, known as diffusion embedding (42), allows local and long distance connections to be projected into a common space more effectively than approaches that use linear dimensionality reduction, such as principal component analysis (SI Materials and Methods).…”

Large-Scale Gradients in Human Cortical Organization

“…Connectivity analyses have often taken the form of dividing the cortex into discrete networks of strongly interconnected areas, but recent studies have demonstrated continuous spatial patterns of connectivity in the human cerebral cortex [31][32][33][34][35][36][37]. These studies simplify the complex connectivity matrix -representing how each cortical area is connected to every other area -to a small set of connectivity gradients.…”

“…In a more data-driven approach, connectivity gradients can be obtained by identifying the main axes of variance in the connectivity matrix [33][34][35][36][37] (Box 2). Each of these axes represents a gradient, along which cortical locations are ordered according to their similarity in connections to the rest of the cortex.…”

“…Sometimes these different aspects are straightforward to interpret. For example, decomposition of primary visual cortex functional connectivity yields two dominant gradients, each reflecting a different aspect of retinotopic organization -eccentricity and polar angle [37]. Other times, the spatial distribution of different superimposed gradients is less obviously related to known anatomical or functional properties and their interpretation can be more challenging [25].…”

“…Describing cortical features in terms of large-scale gradients also makes way for novel analytic approaches. For example, spatial statistics have been used to compare superimposed aspects of connectivity, represented by individual gradients, across sessions or subjects [37]. In addition, as basic cortical gradients appear to be conserved across T r a n s v e r s e s .…”