Figure | Data structure | Type of test | Sample size | Statistical data |

2C (left + right)Open field habituation following VEH or CNO treatment (hM4Di, rM3Ds and control mice^{*}) | Normal distribution | Mixed model analysis with two between-subject factors (group and treatment) and one repeated measure (time), followed by Pairwise Comparisons using the Planned Comparisons procedure (InVivoStat) | 15 CNO hM4Di, 12 CNO rM3Ds, 13 CNO control; 12 VEH hM4Di, 17 VEH rM3Ds, 6 VEH control | F_{TIME}(17,1173) = 47.95, p < 0.0001F_{GROUP}(2,69) = 12.18, p < 0.001F_{TREATMENT}(1,69) = 0.50, p = 0.48F _{GENOTYPE×TREATMENT}(2,69) = 7.36, p = 0.0013Planned Comparisons, relative to control (WT), df = 1242: VEH treated: No significant difference CNO treated: 25: hM4Di, *, p = 0.0064, t = 2.731 & rM3Ds,p = 0.324, t = 0.9930: hM4Di, *, p = 0.042, t = 2.03 & rM3Ds,p = 0.14, t = 1.4735: hM4Di, *, p = 0.016, t = 2.41 & rM3Ds,*, p = 0.021, t = 2.3140: hM4Di, ***, p = 0.0003, t = 3.59 & rM3Ds,p = 0.2827, t = 1.0745: hM4Di, ****, p < 0.0001, t = 3.95 & rM3Ds,p = 0.07, t = 1.850: hM4Di, ****, p < 0.0001, t3.93 & rM3Ds,p = 0.068, t = 1.8255: hM4Di, ****, p < 0.0001, t = 5.1 & rM3Ds,p = 0.15, t = 1.4260: hM4Di, ****, p < 0.0001, t = 4.49 & rM3Ds,*, p = 0.0171, t = 2.3865: hM4Di, ****, p < 0.0001, t = 4.55 & rM3Ds,p = 0.13, t = 1.4970: hM4Di, **, p = 0.0054, t = 2.79 & rM3Ds,**, p = 0.0073, t = 2.6875: hM4Di, **, p = 0.007, t = 2.70 & rM3Ds,p = 0.51, t = 0.6680: hM4Di, ***, p = 0.0032, t = 2.96 & rM3Ds,p = 0.34, t = 0.9585: hM4Di, *, p = 0.016, t = 2.42 & rM3Ds,p = 0.2, t = 1.28 |

2DTotal distance traveled during open field habituation with VEH or CNO pretreatment (hM4Di, rM3Ds and control mice) | Normal distribution | Two-way ANOVA followed by Pairwise Comparisons using the ‘Planned Comparisons’ procedure (InVivoStat) | 15 CNO hM4Di, 12 CNO rM3Ds, 13 CNO control 12 VEH hM4Di, 17 VEH rM3Ds, 6 VEH control | F_{TREATMENT}(1,69) = 0.02, p = 0.89F_{GROUP}(2,69) = 14.93, p < 0.0001F_{INTERACTION}(2,69) = 5.37, p = 0.0068Planned Comparison, Within-treatment, relative to control, df = 69; CNO - hM4Di: **, p = 0.0012, t = 3.39CNO - rM3Ds: *, p = 0.010, t = 2.66VEH - hM4Di: p = 0.908, t = 0.12VEH - rM3Ds: p = 0.255, t = 1.15Within-group, df = 69; Control: p = 0.680, t = 0.41hM4Di: *, p = 0.0103, t = 2.63rM3Ds: p = 0.0605, t = 1.9 |

3AHabituation phase prior to treatment (0–90 min) (hM4Di and control mice). | Normal distribution | Mixed-model analysis with two between subject factors (group and treatment) and one repeated measure (time) (InVivoStat) | 9 saline control, 12 cocaine control, 12 saline hM4Di, 21 cocaine hM4Di | F_{TIME}(8,400) = 86.89, p < 0.0001F_{GROUP}(1,50) = 2.46, p = 0.123F_{TREATMENT}(1,50) = 2.06, p = 0.157 |

3AAcute cocaine response after CNO and habituation (120–180 min) (hM4Di and control mice). | Normal distribution | Mixed-model analysis with two between-subject factors (group and treatment) and one repeated measure (time), followed by Pairwise Comparisons using the ‘Planned Comparisons’ procedure. (InVivoStat) | 9 saline control, 12 cocaine control, 12 saline hM4Di, 21 cocaine hM4Di | F_{TIME}(6,300) = 11.57, p < 0.0001F_{GROUP}(1,50) = 7.68, p = 0.0078F_{TREATMENT}(1,50) = 36.17, p < 0.0001F_{GROUP × TREATMENT}(1,50) = 3.52,p = 0.0667Planned Comparison, relative to control, df = 350?: Cocaine hM4Di vs cocaine control: 120: p = 0.1362, t = 1.49130: *, p = 0.0432, t = 2.03140: **, p = 0.001, t = 3.33150: ***, p = 0.0008, t = 3.39160: **, p = 0.0027, t = 3.01170: *, p = 0.0304, t = 2.17180: p = 0.0635, t = 1.86 |

3B Total distance traveled after cocaine (120–180 min) (hM4Di and control mice). | Normal distribution | Two-way ANOVA followed by pairwise comparisons using the ‘Planned Comparisons’ procedure. (InVivoStat) | 9 saline control, 12 cocaine control, 12 saline hM4Di, 21 cocaine hM4Di | F_{GROUP}(1,50) = 9.71, p = 0.003F_{TREATMENT}(1,50) = 25, p < 0.0001F_{INTERACTION}(1,50) = 2.52, p = 0.119Planned Comparison, df = 50? Within-treatment, hM4Di vs control; Cocaine: **, p = 0.0005, t = 3.70Saline: p = 0.3276, t = 0.99Within-group, saline vs cocaine; Control: ****, p < 0.0001, t = 4.25hM4Di: **, p = 0.0097, t = 2.69 |

3EAMPAR/NMDAR ratio (hM4Di mice) | Normal distribution | Unpaired t test | 6 saline-saline: 4 CNO-cocaine | p = 0.02, t = 2.89, df = 8 |

4DAcute cocaine response after CNO including habituation (hM4Di_{VTA→NAc} and control mice). | Normal distribution | Two-way repeated-measures ANOVA followed by Bonferroni’s post-hoc test | 6 control, 5 hM4Di _{VTA→NAc} | F_{TIME}(17,153) = 11.55, p < 0.0001,F_{GROUP}(1,9) = 6.754, p = 0.0288 andF_{INTERACTION}(17,153) = 3.455, p < 0.0001Multiple comparison, Group, df = 162 130: p = 0.0120, t = 3.471140: p < 0.0001, t = 5.643150: p = 0.0002, t = 4.587 |

4E, left Total distance traveled during habituation (0–90 min) (hM4Di_{VTA→NAc} and control mice). | Normal distribution | Unpaired t test | 6 control, 5 hM4Di _{VTA→NAc} | p = 0.0681, t = 2.072, df = 9 |

4E, rightTotal distance traveled after cocaine (120–180 min) (hM4Di_{VTA→NAc} and control mice). | Normal distribution | Unpaired t test | 6 control, 5 hM4Di _{VTA→NAc} | p = 0.037, t = 2.439, df = 9 |

5B Cocaine sensitization of hM4Di and control mice, day 1 and day 2 of induction (saline control mice) | Normal distribution | Two-way repeated-measures ANOVA followed by Bonferroni’s post-hoc test on induction day 1 | 7 CNO control saline, 4 CNO hM4Di saline, 4 VEH hM4Di saline | Induction day 1 (left);F_{TIME}(5,60) = 27.52, p < 0.0001,F_{GROUP}(2,12) = 4.688, p = 0.0313F_{INTERACTION}(10,60) = 1.338,p = 0.2316Multiple comparison, relative to CNO hM4Di saline, df = 72 VEH hM4Di saline 10: p = 0.016, t = 2.72820: p = 0.0341, t = 2.442CNO control saline 10: p = 0.0024, t = 3.36720: p = 0.0075, t = 2.998Induction day 2 (right); F_{TIME}(5,60) = 16.46, p < 0.0001,F_{GROUP}(2,12) = 3.249, p = 0.0745,F_{INTERACTION}(10,60) = 1.344, p = 0.2288 |

5CTotal distances traveled during cocaine sensitization of hM4Di and control mice, induction days and challenges | Normal distribution | Two-way repeated-measures ANOVA followed by Bonferroni’s post-hoc test | 15 saline controls (7 CNO controls, 4 CNO hM4Di, 4 VEH hM4Di) 10 CNO control cocaine, 10 CNO hM4Di cocaine, 7 VEH hM4Di cocaine | F_{TIME}(8,304) = 38.71, p < 0.0001,F_{GROUP}(3,38) = 35.84, p < 0.0001 andF_{INTERACTION}(24,304) = 5.546, p < 0.0001Multiple comparison, Group, df = 342 CNO hM4Di cocaine vs VEH hM4Di cocaine: Day1: p = 0.171, t = 2.20Day2: p = 0.0003, t = 4.095Day3: p = 0.121, t = 2.335Day4: p = 0.0096, t = 3.182Day5: p = 0.0439, t = 2.698Day6: p < 0.0001, t = 4.513Challenge I: p > 0.999, t = 0.9513 |

5ECocaine sensitization of hM4Di and control mice, total distance traveled during the initial 30 min on day 6 of induction | Normal distribution | One-way ANOVA followed by Bonferroni’s post- hoc test | 15 saline controls, 10 CNO control cocaine, 10 CNO hM4Di cocaine, 7 VEH hM4Di cocaine | F(3,38) = 36.06, p < 0.0001Multiple comparison, df = 38 CNO hM4Di vs CNO control: p = 0.0004, t = 4.518CNO hM4Di vs VEH hM4Di: p = 0.0002, t = 4.710CNO hM4Di vs saline control: p = 0.0041, t = 3.697VEH hM4Di vs saline control: p < 0.0001, t = 8.368CNO control vs saline control: p < 0.0001, t = 8.646 |

5GCocaine sensitization of hM4Di and control mice, total distance traveled during the initial 30 min of Challenge I | Normal distribution | One-way ANOVA followed by Bonferroni’s post- hoc test | 15 saline controls, 10 CNO control cocaine, 10 CNO hM4Di cocaine, 7 VEH hM4Di cocaine | F(3,38) = 6.385, p = 0.0013Multiple comparison, df = 38 CNO hM4Di vs saline control: p = 0.0132, t = 3.209VEH hM4Di vs saline control: p = 0.0119, t = 3.321CNO control vs saline control: p = 0.0081, t = 3.462 |

5ICocaine sensitization of hM4Di and control mice, total distance traveled during the initial 30 min of Challenge II | Normal distribution | One-way ANOVA | 10 CNO control cocaine, 10 CNO hM4Di cocaine, 7 VEH hM4Di cocaine | F(2,24) = 0.7356, p = 0.4897 |

5J Cocaine sensitization of hM4Di and control mice, total distance traveled during Challenge I and Challenge II | Normal distribution | Two-way repeated-measures ANOVA | 10 CNO control cocaine, 10 CNO hM4Di cocaine, 7 VEH hM4Di cocaine | F_{TIME}(1,24) = 4.145, p = 0.0530,F_{GROUP}(2,24) = 0.5531, p = 0.5823,F_{INTERACTION}(2,24) = 0.2762, p = 0.7610 |

5K Effect of CNO on cocaine response in open field test after repeated cocaine administration in home-cage, no habituation (hM4Di and control mice). | Normal distribution | Two-way repeated measures ANOVA | 5 hM4Di cocaine; 5 control cocaine | Day after last injectionF_{TIME}(5,40) = 13.59, p < 0.0001,F_{GROUP}(1,8) = 0.1655, p = 0.6948,F_{INTERACTION}(5,40) = 0.3572, p = 0.8746 |

6BHabituation before and after chronic CNO/VEH administration (hM4Di mice) | Normal distribution | Two-way repeated measures ANOVA | 4 hM4Di CNO, 4 hM4Di VEH | F_{TIME}(5,30) = 8.266, p < 0.0001,F_{GROUP}(1,6) = 0.1646, p = 0.699F_{INTERACTION}(5,30) = 0.3483, p = 0.8793 |

6C-D Protein levels of pTH and DAT after chronic CNO (hM4Di mice) | Assuming normality | Unpaired t-test | 4 hM4Di CNO, 4 hM4Di VEH | C; pTH:p = 0.9225, t = 0.1014, df = 6;D; DAT: p = 0.713, t = 0.3857, df = 6 |

6EStriatal dopamine levels DAT after chronic CNO (hM4Di mice) | Assuming normality | Unpaired t-test | 4 hM4Di CNO, 4 hM4Di VEH | NAc:p = 0.547, t = 0.63, df = 6dStr: p = 0.45, t = 0.807, df = 6 |

7CCocaine conditioned place preference in hM4Di mice, time in paired (% of baseline) | Normal distribution | One-way ANOVA followed by Bonferroni’s post-hoc test | 8 saline, 6 cocaine CNO, 8 cocaine VEH | F(2,19) = 6.549, p = 0.0069Multiple comparison, relative to saline, df = 19 Cocaine VEH: p = 0.0060, t = 3.401Cocaine CNO: p = 0.0302, t = 2.672 |

7DCocaine conditioned place preference in hM4Di mice, time in paired | Normal distribution | Two-way repeated measures ANOVA followed by Bonferroni’s post-hoc test | 8 saline, 6 cocaine CNO, 8 cocaine VEH | F_{TIME}(8.152) = 8.656, p < 0.0001,F_{GROUP}(2,19) = 2.174, p = 0.1412F_{INTERACTION}(16,152) = 1.833, p = 0.0315;Multiple comparison, relative to baseline, df = 152: Cocaine VEH: Post-test: p < 0.0001, t = 5.320Ext1: p = 0.0003, t = 4.267Ext2: p = 0.0084, t = 3.341Reinstatement: p = 0.0350, t = 2.893Cocaine CNO: Post-test: p = 0.0001, t = 4.451Ext1: p < 0.0001, t = 4.981Ext2: p = 0.0001, t = 4.421Ext3: p = 0.0002, t = 4.322Ext4: p = 0.0033, t = 3.608Ext5: p = 0.0117, t = 3.242 |

7ELocomotor activity during baseline-, post-, and reinstatement-test of the cocaine conditioned place preference | Normal distribution | Two-way repeated measures ANOVA followed by Bonferroni’s post-hoc test | 8 saline, 6 cocaine CNO, 8 cocaine VEH | F_{TIME}(2,38) = 38.60, p < 0.0001,F_{GROUP}(2,19) = 9.012, p = 0.0018F_{INTERACTION}(4,38) = 5.036, p = 0.0023Multiple comparison, relative to saline, df = 57: Cocaine VEH: Baseline: p > 0.999, t = 0.4795Post-test: p = 0.0009, t = 3.714Reinstatement: p < 0.0001, t = 5.410Cocaine CNO: Baseline: p > 0.999, t = 0.1025Post-test: p = 0.3743, t = 1.335Reinstatement: p = 0.1354, t = 1.863 |

7FLocomotor activity during conditioning sessions in paired compartment | Normal distribution | Two-way repeated measures ANOVA followed by Bonferroni’s post-hoc test | 8 saline, 6 cocaine CNO, 8 cocaine VEH | F_{TIME}(3,57) = 10.22, p < 0.0001,F_{GROUP}(2,19) = 15.21, p = 0.0001 andF_{INTERACTION}(6,57) = 7.515, p < 0.0001;Multiple comparison, relative to cocaine VEH, df = 76: Saline VEH: 1st: p = 0.0005, t = 3.8232nd: p = 0.0061, t = 3.0623rd: p < 0.0001, t = 6.1214th: p < 0.0001, t = 6.768Cocaine CNO: 3rd: p = 0.0002, t = 1.7874th: p = 0.0002, t = 4.124Multiple comparison, relative to 1st session, df = 57: Cocaine VEH: 3rd: p = 0.001, t = 4.4084th: p < 0.0001, t = 5.843 |

7H Conditioned place preference in hM4Di mice following CNO treatment, time in paired (% of baseline) | Normal distribution | Two-way repeated measures ANOVA followed by Bonferroni’s post-hoc test | 7 saline (+CNO), 8 saline (+CNO/VEH) 8 saline (+VEH) from Fig. 7C,D | F_{TIME}(4,80) = 3.582, p = 0.0097,F_{GROUP}(2,20) = 0.1793, p = 0.8372F_{INTERACTION}(8,80) = 0.8799, p = 0.5371 |

8C+DReward preference test in hM4Di mice, total volumes consumed | Normal distribution | Unpaired t-test | 8 hM4Di CNO, 8 hM4Di VEH | p = 0.22, t = 1.27, df = 14 |

8EReward preference test in hM4Di mice, reward preference ratio (%) | Normal distribution | Unpaired t-test | 8 hM4Di CNO, 8 hM4Di VEH | p = 0.30, t = 1.08, df = 14 |

8GTouchscreen-based motivational assay, number of touch trials | Normal distribution | Unpaired t-test | 8 hM4Di CNO, 8 hM4Di VEH | p = 0.01, t = 2.83, df = 14 |

8HTouchscreen-based motivational assay, mean correct latency | Normal distribution | Unpaired t-test | 8 hM4Di CNO, 8 hM4Di VEH | p = 0.04, t = 2.28, df = 14 |

8ITouchscreen-based motivational assay, mean correct latency | Normal distribution | Unpaired t-test | 8 hM4Di CNO, 8 hM4Di VEH | p = 0.29, t = 1.11, df = 14 |

Statistical analyses were performed in GraphPad Prism 6, except for behavioral data in Figs. 2 and 3, which were analyzed InVivoStat (Invivostat.co.uk).

↵* Control mice are wild-type mice injected with the Cre-dependent AAV and do not express the DREADD transgenes.