The Show Must Go Wrong: Towards an understanding of audience perception of error in digital musical instrument performance

Abstract

This thesis is about DMI (digital musical instrument) performance, its audiences, and their perception of error. The goal of this research is to improve current understanding of how audiences perceive DMI performance, where performers and their audiences often have no shared, external frame of reference with which to judge the musical output. Further complicating this audience-performer relationship are human-computer interaction (HCI) issues arising from the use of a com- puter as a musical instrument. In current DMI literature, there is little direct inquiry of audience perception on these issues. Error is an aspect of this kind of audience perception. Error, a condition reached by stepping out of bounds, appears at first to be a simple binary quantity, but the location and nature of those boundaries change with con- text. With deviation the locus of style and artistic progress, understanding how audiences perceive error has the potential to lend important insight to the cultural mechanics of DMI performance. In this thesis I describe the process of investigating audience perception and unpacking these issues through three studies. Each study examines the relative effects of various factors on audience perception — instrument familiarity and musical style, gesture size, and visible risk — using a novel methodology combining real-time data collected by mobile phone, and post- hoc data in the form of written surveys. The results have implications for DMI and HCI researchers as well as DMI performers and composers, and contribute insights on these confounding factors from the audience’s perspective as well as important insights on audience perception of error in this context. Further, through this thesis I contribute a practical method and tool that can be used to continue this audience-focused work in the future.