This paper presents an empirical evaluation of a digital music instrument (DMI) for electroacoustic vocal performance, the Tibetan Singing Prayer Wheel (TSPW). Specifically, we study audience preference for the way it maps horizontal spinning gestures to vocal processing parameters. We filmed six songs with the singer using the TSPW, and created two alternative soundtracks for each song: one desynchronized, and one with the mapping inverted. Participants viewed all six songs with either the original or desynchronized soundtrack (Experiment 1), or either the original or inverted-mapping soundtrack (Experiment 2). Participants were asked several questions via questionnaire after each song. Overall, they reported higher engagement and preference for the original versions, suggesting that audiences of the TSPW prefer more highly synchronized performances, as well as more intuitive mappings, though level of perceived expression of the performer only significantly differed in Experiment 1. Further, we believe that our experimental methods contribute to how DMIs can be evaluated from the audience’s (a recently noted underrepresented stakeholder) perspective.