During perceptual tasks involving the discrimination of musical intervals, event-related potentials, specifically the P3, were measured for three subject groups: musicians without absolute pitch, musicians with absolute pitch, and nonmusicians. The two interval-discrimination tasks were a simple two-note contour task and a difficult interval-size discrimination task. Clear effects on the neural waveforms were found for both training and the presence of the absolute pitch ability. In general, training increases the amplitude and shortens the latency of the P3, while the absolute pitch ability reduces the amplitude and shortens the latency, or eliminates the P3 altogether. The absolute pitch effect may be due to the use of a long-term memory strategy involved in the correct performance of the discrimination task rather than performing the task by updating working memory each time a target occurs. Finally, these data are contrasted with those from studies involving sine tones and timbrediscrimination tasks.