Figure 3. Test-retest reliabilities (ICCs) of significant task-related activations varied from none-to-high across different parcels. The FDR corrected significant clusters were further divided into anatomical segments using the Human Connectome Project Multi-Modal Parcellation (MMP1.0) and the Freesurfer subcortical segmentation. ICCs were mapped separately for the decision making and outcome (feedback) phases of the BART for the categorical and the parametric designs. On cortical surface view, Gray outlines depict the boundaries of the MMP1.0 cortical parcellation and Black outlines depict the thresholded subset. Cortical parcels with negative ICC values are not plotted. On subcortical volume view, Black outlines depict the Freesurfer segmentation. All cortical and subcortical ICCs were lower than .8. L: left, R: right. Note that this scene file includes ICC estimates of significant task-related activations for the categorical design only. Each tab presents ICCs for a specific condition (i.e. ChooseInflate) either for cortical (*_Surface) or subcortical (*_Volume) regions. For the ICC estimates of parametric design, see scene4.
Surface Mesh:32k fs LR, Modality:T2-weighted, Modality:T1-weighted, Species:Human, Parcellation:HCP_MMP1.0, Atlas:HCP-S1200, Modality:Task fMRI