Data Pre-processing

Anatomical images were skull-stripped, segmented, and normalized to the MNI152 template using the Advanced Normalization Tools (ANTs) (39). Standard preprocessing steps were performed and included the following: motion-correction, registration to the anatomical brain, intensity normalization, nuisance signal removal [including 24 motion parameters, five principal components derived from the CSF and white matter signals [according to the compcor method, see Behzadi et al. (40)], the average signal from the CSF, as well as the linear and quadratic trends], and temporal filtering between 0.01 and 0.1 Hz.

Data preprocessing was performed using an alpha version of the Configurable Pipeline for the Analysis of Connectomes (C-PAC version 0.3.9, http://fcp-indi.github.io./). Three seed regions (NAcc, amygdala, and VTA) were defined by centering bilateral spheres with a radius of 5 mm at the coordinates shown in Figures 1–3. The location of these seeds was based on the study by Konova et al. (28). To reduce multiple testing issues, we decided not to include the hippocampus, thalamus, and rostral ACC as seed regions, because these areas are not of primary interest in DBD populations. Time series from these seeds were extracted by warping the preprocessed resting-state images to MNI space and averaging the signal in the seed regions. To test our hypotheses, whole-brain correlation maps were calculated separately for each bilateral seed region by correlating the time series from each voxel (in MNI space) with the seed time course. Subsequently, correlation maps were smoothed by applying a 4-mm FWHM Gaussian filter and transformed using Fisher R-to-Z to improve normality.

NAcc connectivity differences between HC, DBD-PCB, and DBD-MPH groups. (A) Location of the NAcc seed. (B) Clusters showing significant connectivity differences (Z > 2.3, cluster p < 0.05). (C) Average Fisher-Z transformed NAcc connectivity in the significant clusters. Dark gray patches represent the 95% confidence interval and light gray patches represent the standard deviations. Black lines represent the group means. All shown coordinates are in MNI space. NAcc, nucleus accumbens; OccPole, Occipital pole; Prec, Precuneus; MFG, medial frontal gyrus; IPL, inferior parietal lobule; PCC, posterior cingulate cortex; HC, healthy control; PCB, placebo; MPH, methylphenidate; *cluster-p < 0.05; **cluster p < 0.001.

VTA connectivity differences between HC, DBD-PCB, and DBD-MPH groups. (A) Location of the VTA seed. (B) Clusters showing significant connectivity differences (Z > 2.3, cluster p < 0.05). (C) Average Fisher-Z transformed VTA connectivity in the significant clusters. Dark gray patches represent the 95% confidence interval and light gray patches represent the standard deviations. Lines represent the group means. All shown coordinates are in MNI space. VTA, ventral tegmental area; PCG, post-central gyrus; SMC, supplementary motor cortex; HC, healthy controls; PCB, placebo; MPH, methylphenidate; *cluster-p < 0.05.