It has been shown that saccade-related neurons in the superior colliculus (SC) display an increased level of prestimulus activity and a higher stimulus-related burst in action potentials preceding direction errors in the anti-saccade task compared with correct anti-saccades. From this, it has been hypothesized that errors occur when the incoming visual signal in the SC passes a threshold and triggers a reflexive saccade. This hypothesis predicts that an attenuated visual signal will reduce the number of direction errors. Since ethanol has been shown to have a suppressive effect on cortical visual event-related potentials (ERPs), the purpose of the present study was to investigate the effects of moderate ethanol consumption on anti-saccade performance. Under both placebo and ethanol conditions, we recorded ERPs and measured eye movements in male subjects during the performance of an anti-saccade task in which the fixation point disappeared 200 ms prior to stimulus presentation. Compared with the placebo condition, we found in the ethanol condition: (1). a decrease in ERP amplitudes during the gap period and after stimulus presentation, (2). an increase in the latencies of anti-saccades, and (3). a decrease in the percentage of direction errors. These data demonstrate the effects of ethanol on anti-saccade task performance and provide further support for the hypothesis that errors in the anti-saccade task are triggered by the incoming visual signal.