Table S1. Right IFG activitywithin the anatomical maskduring response inhibition (“No-Go versus Go trials” and “No-Go versus oddball” contrast) and oddball-driven attention allocation (“oddball versus Go” contrast) after the administration of saline orheroin.

Contrast / Treatment / MNI Coordinates
(x, y, z) / Cluster size / T-value at voxel-level
No-Go versus Go / Saline / (52, 14, 38) / 34 / 4.12
Heroin / (34, 16, 28) / 66 / 3.67
No-Go versus Oddball / Saline / (42, 18, 6) / 31 / 4.19
Heroin / (60, 16, 12) / 59 / 3.26
Oddball versus Go trials / Saline / (54, 16, 34) / 36 / 3.54
Heroin / (32, 14, 28) / 38 / 3.35

Note:All reported activations survived peak and voxel-level correction for multiple comparisons after adjustment for small volume across the ROI (FWE-corrected at p<0.05).

Figure S1. Activity in the right IFG within the anatomical maskafter saline (upper row) and heroin (lower row) administration during successful response inhibition as expressed either by the “No-GoversusGo” (column 1) or “No-Goversusoddball” contrasts (column 2) and during oddball attention allocation as expressed by the “oddballversusGo” contrast(column 3). The right side of the brain is shown on the right side of the images. All activities survived small volume correction at peak and cluster-level (FWE-corrected at p<0.05).

Table S2. Difference in right IFG activitywithin the anatomical maskbetween saline and heroin during response inhibition and oddball-driven attention allocation.

Cognitive Function / Treatment / MNI Coordinates
(x, y, z) / Cluster size / T-value at voxel-level
Response Inhibition
(“No-Go versus Go”) / Saline>heroin / (58, 20, 2) / 21 / 3.23
Oddball attention allocation
(“Oddball versus Go”) / Saline>heroin / (54, 28, 0) / 18 / 3.47

Note:All reported activations survived peak and voxel-level correction for multiple comparisons after adjustment for small volume across the ROI (FWE-corrected at p<0.05).

Figure S2. Significant heroin-induced reduction (saline>heroin)of right IFG activity during oddball attention allocation (“oddballversusGo” contrast, row 1) and response inhibition (“No-Go versus Go”, row 2). The right side of the brain is shown on the right side of the images. All activities survived small volume correction at clusterlevel (FWE-corrected at p<0.05).

Table S3. Relation between changes in brain activity and task performance, symptom ratings and consumption of psychostimulant substances.

Effect / F-value / p-value / Partial eta squared
Smoking behaviour / 0.339 / 0.718 / 0.043
Changes in drug carving / 0.699 / 0.513 / 0.085
Changes in state-anxiety / 0.773 / 0.479 / 0.093
Changes in probability of inhibition / 0.726 / 0.500 / 0.088
Changes in sensitivity index d’ / 0.159 / 0.855 / 0.021
Changes in response bias c / 0.697 / 0.514 / 0.085
Cocaine consumption / 1.719 / 0.213 / 0.186
Cannabis consumption / 0.182 / 0.835 / 0.024
Cocaine*cannabis interaction / 1.182 / 0.334 / 0.136

Multivariate analysis, with heroin-induced changes in right IFG activity (“No-Go-Go” and “Oddball-Go”) as dependent variables, cocaine and cannabis consumption as fixed factors and changes in drug craving, state-anxiety, behavioural performance, and nicotine consumption as covariates.