Fig. 2: Datasets description and cortical gradient calculation.
From: Functional geometry of the cortex encodes dimensions of consciousness
a Five fMRI datasets consist of participants and patients with normal wakefulness (baseline/healthy controls) and depressed (or altered) states of consciousness. b An overview of cortical gradient analysis. The fMRI time courses were extracted from 400 cortical areas according to a brain parcellation scheme26. A 400 × 400 functional connectivity matrix was calculated for each participant and each condition. A normalized cosine angle affinity matrix was calculated to capture the similarity of connectivity profiles between cortical areas. Cortical gradients were computed using a diffusion map embedding algorithm. c Gradient-1 ranges from unimodal primary sensory areas to transmodal cortex. Gradient-2 ranges from visual to somatomotor cortices. Gradient-3 ranges from visual/default-mode to areas commonly involved in multiple-demand tasks. d Connectome-level variance explained by the gradients (mean ± SD across all scans). The first three gradients explained ~37% of the variance in the functional connectivity matrices. Source data are provided as a Source Data file.
