The influence of self-prescribed exercise intensity on information processing in the central nervous system (CNS) was investigated using P300, no-go P300, and contingent negative variation (CNV) event-related potentials (ERP). Thirteen subjects aged 22-35y performed a go/no-go reaction time task under a control condition, and again after self- and externally prescribed pedaling exercises. The average rating of perceived exertion (RPE), heart rate (HR), and work rate (WR) for 20min of both exercises were equivalent. The comfort level after self-prescribed and externally prescribed exercises was measured by a visual analog scale (VAS). VAS after self-prescribed exercise was increased compared with that after externally prescribed exercise, indicating that the comfort level after self-prescribed exercise was larger. That is, we only manipulated comfort levels because exercise intensity and duration were fixed for each subject. The P300 latency shortened after self-prescribed exercise compared to the control condition, and the P300 amplitude after self-prescribed exercise was larger than under the control condition. The larger and earlier P300 after self-prescribed exercise suggested that when larger amounts of attentional resources were allocated to a given task, the stimulus evaluation time was faster. However, early CNV amplitude was not influenced by either type of pedaling exercise. From these results, we suggest that P300 changes after self-prescribed exercise due to psychological factors such as comfort level, but not due physiological factors such as arousal level. In addition, no-go P300 also showed the same changes as P300 after self-prescribed exercise. This indicates that self-prescribed factors (comfort level) influence not only the stimulus evaluation processing the requirement for a go response, but also processing of the need for a no-go response. In conclusion, the present findings suggest that differences in comfort level obtained after exercise influence information processing in the CNS.