<jats:p>Three experiments investigated the processing costs of watching television messages. Processing costs were indexed with a secondary task reaction time measure in which subjects were asked to pay attention to commercial messages while responding with button presses to randomly occuring tones or flashes. Response time to the secondary tasks was used as a measure of attention to the primary task (watching the messages). Audio and video complexity of the messages were within-subject variables, and the channels presented to subjects (audio-only, video-only, or both) was a between-subjects variable. Results indicated that: (1) for a tone secondary task, multiple-channel presentations demanded more capacity than single-channel presentations (video or audio channel only); (2) more capacity was required to process simple video and auditory information than complex information; and (3) complexity of information in an absent channel (e.g., visual information in the audio-only condition) produced the same slowed reaction times as those occurring when the channel was present.</jats:p>