Fig. 3: Test – Retest reliability of deviation scores for faces>shapes models. | Communications Biology

Fig. 3: Test – Retest reliability of deviation scores for faces>shapes models.

From: Dissecting task-based fMRI activity using normative modelling: an application to the Emotional Face Matching Task

Fig. 3

Normative Probability Maps illustrate the voxels wherein 2 or more participants had positive (a, hot colours) or negative deviations (b, cool colours) > ±2.6 for the faces>shapes normative models in the Test (top rows) and Re-Test (bottom rows) samples. Intra-class correlation coefficients unthresholded (left) and thresholded to show only regions that had a moderate ICC or higher ( > 0.5; c). Mean within-subject difference per voxel (histogram) illustrated thresholded at > 0.5 (i.e., a change greater than half a standard deviation between Test and Retest scans (d). The correlation coefficients (rho) between Test and Retest deviation scores (histogram) illustrated thresholded by the coefficients of determination (rho2 > 0.3, e). x, y, z = −4, −6, −16.

Back to article page