Laboratory studies have demonstrated that acute alcohol intoxication can disrupt performance on neuropsychological tests of executive cognitive functioning such as the Wisconsin Card Sorting Test (WCST). However, the generalizability of such findings to typical self-regulated alcohol intake in social settings can be questioned. In the present study, 86 young adults were recruited at Australian bars to perform a computer version of the WCST. Participants displayed blood alcohol concentrations (BACs) across a range from 0 to 0.15%. Although self-report measures of typical alcohol consumption, impulsivity, and frontal lobe related everyday functioning were all intercorrelated in line with other recent findings, multiple regression indicated that these measures did not predict perseverative errors (PE) nor non-perseverative errors (NPE) on the WCST, whereas BAC uniquely predicted PE but not NPE. The results were consistent with a dose-dependent selective disruption of prefrontal cortical functioning by alcohol. There were no differences in performance between participants tested on the ascending limb of the BAC curve and those tested on the descending limb. Alcohol-associated perseveration may reflect the inhibitory effect of alcohol-induced dopamine release in the prefrontal cortex.