In multimedia learning, the modality effect occurs when students learn better from a lesson containing graphics accompanied by narration than one accompanied by on-screen text. The redundancy effect occurs when students learn better from a lesson containing graphics accompanied by narration than one accompanied by narration and on-screen text. In order to determine the information-processing mechanisms responsible for these effects, 36 students viewed three multimedia lessons in which the words were presented as narration, on-screen text, or both. During the lessons, brief visual distractors were presented and event-related potentials (ERPs) in response were measured. Learners showed a more positive early (P1) ERP response during the graphics+text lesson than during the graphics+narration+text lesson, indicating that more perceptual processing was required for the latter condition. In general, results suggested that perceptual load plays an important role in the modality and redundancy effects, a useful clarification of the cognitive theory of multimedia learning.