Fig 1.
Overview of the location of the amygdala.
A,B: Amygdala in a medial view of the brain. C,D: The location of the nine different amygdaloid nuclei within the amygdala.
Table 1.
Amygdala subnulcei and their abbreviations.
Fig 2.
Overview of the analysis pipeline in our data.
See text for details.
Table 2.
Description of the sample.
Table 3.
Table of correlations of the whole-brain FC maps for each IC with known resting-state networks.
Fig 3.
Results from srICA and dual regression.
srICA left colum and Dual Regression right column. A: IC2 (Ventral Attention), B: IC3 (Somatomotor), C: IC5 (Default Mode). Note how different IC hotspots inside the amygdala connect to separate whole brain networks.
Fig 4.
Detailed view of results from srICA.
C2 (red, associated with Ventral Attention), IC3 (green, associated with Somatomotor) and IC5 (blue, associated with Default Mode) in saggital (A), coronal (B), and axial (C) views.
Fig 5.
Results from group-level analyses for all ROIs.
Representation of estimated marginal mean group-level FC values between IC hotspots inside the amygdala and each cortical and subcortical region using a spiderplot (A), as well as the z-ratios representing group-level FC between IC hotspots inside the amygdala and subcortical (B) and cortical (C) regions. Note how different hotspots of activity detected inside the amygdala are functionally connected with clearly different whole-brain networks.
Table 4.
Cortical and subcortical areas showing reliable co-activity with IC2 (correlated with ventral attention network) relative to the mean co-activity value of all other areas.
p-values corrected for multiple comparisons using Bonferroni correction.
Table 5.
Cortical and subcortical areas showing reliable co-activity with IC3 (correlated with Somatomotor network) relative to the mean co-activity value of all other areas.
p-values corrected for multiple comparisons using Bonferroni correction.
Table 6.
Cortical and subcortical areas showing reliable co-activity with IC5 (correlated with default mode network) relative to the mean co-activity value of all other areas.
p-values corrected for multiple comparisons using Bonferroni correction.
Fig 6.
Results from group-level analyses for amygdala nuclei.
Estimated marginal mean co-activity values for each amygdaloid nucleus across the three IC hotspots detected in the amygdala (A), as well as the relative contributions of each amygdaloid subnucleus within each IC hotspot in terms of summed group-level pairwise comparisons (B). Note how the different IC hotspots that are associated with different resting-state networks rely on different configurations of subnucleus co-activity.
Table 7.
Relative contributions of each amygdaloid subnucleus within the different resting-state networks.
Rankings based on the sum of pairwise z-ratio differences for all subnuclei within a given IC.
Fig 7.
Results from fear affect analyses.
Distribution of Fear Affect scores (A) with low and high FearAffect groups defined by the median value (indicated by the vertical line), and marginal means of co-activity scores between the low and high fear group for each amygdaloid nucleus. * indicates significant difference (p < 0.001).