Rats were trained in a previously validated task for the assessment of sustained attention, or vigilance. This task required the animals to discriminate between signals of variable lengths and non-signal events by making an appropriate lever press. The performance of sham-lesioned animals in this task was characterized by a signal-length dependent number of hits. Also, approximately 70 percent of the non-signals were correctly rejected. Ibotenic acid-induced lesions of the medial prefrontal cortex decreased the relative number of hits and correct rejections and, in essence, resulted in random lever selection. The lesion did not affect the number of omissions or side bias. Furthermore, the performance of lesioned animals was insensitive to the detrimental effects of distractors. The effects of the lesions do not support an interpretation in terms of sustained attention. Rather, the pattern of the lesioned animals' performance is speculated to reveal a fundamental disruption of decisional processes, reminiscent of the executive dysfunctions observed in patients with damage to ventromedial prefrontal areas or with schizophrenia.