All demographic and behavioral data was analyzed in SPSS version 18.0 (SPSS Inc, Chicago, IL) and are presented as mean ± standard deviation unless otherwise indicated. Behavioral responses to the anticipation of reward task were analyzed for percentage correct response and mean response time with MANOVA for group and MPH interaction. MRI scans were analyzed using FSL 5.1 (FMRIB-Software-Library, Functional Magnetic Resonance Imaging of the Brain Center, Dept. of Clinical Neurology, University of Oxford, Oxford, UK, http://www.fmrib.ox.ac.uk). this website Non-brain structures were removed from 3D T1 anatomical scans using the Brain Extraction Tool (BET; Smith,

2002). Scans were analyzed using FSL’s fMRI Expert Analysis Tool (FEAT; Beckmann et al., 2003) and Motion Correction using FMRIB’s Linear Image Registration Tool (MCFLIRT; Jenkinson et al., 2002), BET brain extraction, spatial smoothing set at 5 mm FWHM, high-pass filter cut-off at 100 s. Scans were registered to the high resolution structural image and to standard MNI space (MNI152_T1_2 mm_brain from the FSL atlas Imatinib ic50 library). The general linear model (GLM) used for first-level analysis was adapted from Knutson et al. (2001), with contrasts set to reward versus neutral, loss versus neutral, reward versus

loss and large reward (€5.00) versus small reward (€0.20). The obtained first-level analysis was entered into a second-level (group effect) analysis. Main task effect was determined by examining first level effects for all scans available, both baseline and post-challenge scans were used for this analysis. Next a mask of the corpus striatum based on the Harvard–Oxford brain atlas, including putamen, caudate, nucleus accumbens and globus pallidus, was

applied to the data, to determine ROI specific activation. Because cocaine also influences the DA system and cocaine use was common in dAMPH we decided to incorporate cocaine in the analysis. Cocaine use in the last 12 months was demeaned and added as a covariate to the higher-level analysis. MRIP In order to keep covariate use as low as possible in this explorative study we decided against correcting for age and IQ, seeing as we do not consider these to have a significant influence on DAergic function in this relatively young population. For illustrative purposes we determined striatal task-activation for both groups and conditions separately, uncorrected with p < 0.01. Baseline group differences, in addition to the effect of the challenge per group were calculated. The interaction effect of group × drug challenge was calculated. For these comparisons (baseline effect, challenge effect and interaction) statistical threshold was set at p = 0.