Music is an inherently emotional feat of human creativity. Previous studies of music perception have followed the path of speech perception research, in which subjects were asked to simultaneously attend to multiple signals in an effort to determine how much information one could consciously perceive. While these studies have led to the development of models that seek to explain the role of attention in music perception, they lack external validity. Many of these studies have required subjects to detect errors in simultaneously played melodies. While this presents certain demands on attention, it does not accurately portray the way that music is listened to. Music is born out of a burst of emotion and inspiration, and is designed to stir such feelings in those who listen to it. Therefore, the present study sought to study music perception in a more naturalistic setting. In Experiments 1 and 2, participants listened to pieces of music. Participants in Experiment 1 were asked to detect changes in musical arrangement, while participants in Experiment 2 were asked to rate their enjoyment of the music. The results of the present study revealed that participants struggled to identify changes in musical arrangement. Changes that were more noticeable received a significantly greater mean number of emotional responses than changes that were less noticeable, which suggests they had a stronger impact on emotional responses. Further research is needed to determine whether existing models of music perception can be applied to more naturalistic tasks like those used in the present study.