Acute effects of ethanol (0.55 g/kg) and the opioid antagonist naltrexone (50 mg) on auditory event-related brain potentials (ERP) (i.e., electrical brain activity time-locked to sensory stimuli) were investigated in 13 healthy social drinkers, using a double-blind, placebo-controlled, design. The subjects' task was to attend to tones presented to a designated ear while ignoring tones to the other, and to detect deviant tones among the attended tones. When administered alone, naltrexone significantly reduced the amplitude of the later part of negative difference (Nd[l]), suggesting impaired selective attention. However, this effect might have been caused by naltrexone-induced nausea. Ethanol, when ingested alone, attenuated the amplitude of the N1, and increased the peak latencies of the mismatch negativity (MMN) and N2b that have been suggested to reflect automatic change detection in audition and allocation of attentional resources to processing of stimulus deviance, respectively. In contrast, the P1 amplitude was augmented by alcohol, but only when the tones were attended. When ethanol and naltrexone were simultaneously ingested, however, the alcohol-induced P1 amplitude augmentation was canceled, thus tentatively suggesting opioidergic mediation of this alcohol effect. In contrast, the MMN peak latency was increased significantly more in the interaction condition than in the ethanol condition, thus suggesting that the detrimental effects of alcohol on involuntary attention switching were augmented by naltrexone. Furthermore, the N2b amplitude was significantly suppressed in the interaction condition, suggesting attentional impairment.