This study questions the assumption that a majority of nonmusicians pay attention to key when not specifically directed to do so. The question asked was "what do people hear," rather than "what can people hear." Musicians and nonmusicians listened to a song sung in a different key from its accompaniment. There were no other differences in the music stimuli. Subjects were asked what differences, if any, they heard. In other words, left to their own devices would listeners pay attention to and report the bitonality of the excerpt? Although 100% of the musicians heard the clash of keys, only 40% of the nonmusicians indicated that they heard the difference in keys between singer and accompaniment. Of these, half also mentioned nonexistent differences, for example, in instrumentation, dynamics, and tempo.