second block (p < 0.001), while there was no significant change from the second to the third one. The placebo group, by contrast, slowly increased the number of pumps throughout the task, showing a significant increase only between the first and the last block (p < 0.05). Furthermore, we found marginally significant between-group differences in trials 1–20: participants in the alcohol group made marginally more risky decisions (p = 0.06) during the first 20 trials as compared to the control group. Mean number of pumps for each block of 20 trials, as a function of condition, are depicted in Fig. 2. Table 1Descriptive statistics for BART measures for the alcohol (n = 31) and placebo groups (n = 30)Alcohol groupPlacebo groupMeanSDMeanSDMean pumps62.329.9363.196.27Pumps 1–2064.929.5860.398.01Pumps 21–4060.229.6863.868.40Pumps 41–6061.8212.9965.067.16
