What are the factors that determine how long a person chooses to listen to an unfamiliar piece of music? We examined this question across three experiments in which we played participants novel repeating multi-instrument stimuli and recorded their listening times and reasons for their decisions to either continue or stop listening. To influence the habituating effects of repeating musical material drawn from a large stimulus library (> 450 items), we manipulated novelty along several musical dimensions. In Experiment 1, all instruments entered simultaneously. In Experiment 2, instrument entrances were also offset in time. In Experiment 3, we composed core multi-instrument loops and manipulated them to further minimize harmonic variability, minimize rhythmic variability, introduce spatialization, or change timbral characteristics. Novelty introduced by instrument entrances was the strongest determinant of listening times, though harmonic variability and timbral features were also important. Subjective enjoyment was the best predictor of listening times, mediating the effects of the degree of perceived groove in a stimulus, the urge to move, interest in a stimulus, perceived complexity, and congruency with current mood. We conclude that naturalistic looping musical stimuli serve well to examine the diverse psychological and musical determinants of choice behavior underlying music consumption.