Fig. 3: Whole-brain activations coupled with Amyg-EFP compared to VS-EFP signals.

Results of linear regression of voxel-wise BOLD to EFP time-course for the Amyg-EFP and VS-EFP during passive viewing of movie scenes. (A) Above-chance voxels (P < 0.05, FDR corrected) associated with the Amyg-EFP (orange) and VS-EFP (blue), overlayed on an inflated brain template image. (B) Voxels showing significantly higher (P < 0.05, FDR corrected) GLM effect sizes for regression to the Amyg-EFP than to the VS-EFP were observed mainly in salience-network regions (black outline) and sensory (visual, auditory, somato-motor) regions (dotted outline), as defined by Schaefer et al., (2018) [65]. (C) Mean effect size in major nodes of the salience network [65], for regression to the Amyg-EFP (orange) and to the VS-EFP (blue), presented as mean ± SEM across subjects. Replication data presented in the second column, denoted ‘rep’.