Muses in the Making: Composing and Producing Modular Music to Identify Unique Audio-Sensory Preferences in Individuals on the Autism Spectrum

Researcher(s)

  • Elise Ruggiero, Music - Applied, University of Delaware
  • Connor Fone, Music Composition, University of Delaware
  • Justin Eichenberg, Music Composition, University of Delaware

Faculty Mentor(s)

  • Daniel Stevens, School of Music, University of Delaware
  • Matthew Mauriello, Computer and Information Sciences, University of Delaware

Abstract

While research correlating Autism Spectrum Disorder (ASD) and specific musical preferences is sparse, prior studies have identified atypical responses from individuals with ASD to sound-based stimuli (Dotch et al. 1). For example, individuals with ASD have shown higher sensitivity to environmental sounds (Stiegler & Davis 73) and pitch variance (Hisaizumi & Tantam 14), as well as favorance for lower pitched (Santos et al. 2) and non-vocal sounds (Michel et al. 7). In 2022, we developed an interactive musical interface (uCue) to be operated by participants with ASD to help identify preferences by providing users agency over certain compositional elements. Building on prior research and user data from  recent pilot studies, we aim to further isolate such preferences in future studies by producing songs that test for specific variables. The songs provide contrasting versions of multiple layers that are designed to facilitate A/B testing for preferences. In order to test listener preferences for pitch level (high, medium, low), one potential melody layer in our arrangement of “Rain, Rain, Go Away” features a high instrument (violin) and another layer features a low instrument (cello) from the same instrument family (strings). “London Bridge” tests for timbral preferences, featuring clarinet for Melody 1 and viola for Melody 2. While the instruments share similar ranges, they host different timbres–the clarinet produces a bright sound, whereas the viola has a dark, warm tone. Lastly, our rendition of “Hush Little Baby” tests for rhythmic subdivision level. Melody 1 utilizes eight rhythmic subdivisions per meter, Melody 2 utilizes twelve, and Melody 3 utilizes sixteen. Each line features identical synthesizers to control for timbre and similar melodic structures. In the fall, we intend to hold additional studies using the newly expanded music library in local libraries and schools to expand our data set, analyze trends, and identify emergent preference patterns.