Dissertation Index

Author: Baker, David J

Title: Modeling Melodic Dictation

Institution: Louisiana State University

Begun: August 2015

Completed: August 2019


Melodic dictation is a cognitively demanding process that requires students to hear a melody, then without any access to an external reference, transcribe the melody within a limited time frame. Despite its ubiquity in curricula within School of Music settings, exactly how an individual learns a melody is not well understood. This dissertation aims to fill the gap in the literature between aural skills practitioners and music psychologists in order to reach conclusions that can be applied systematically in pedagogical contexts. In order to do this, I synthesize literature from music theory, music psychology, and music education in order to demonstrate how tools from both cognitive psychology as well as computational musicology can be used to help inform pedagogical practices.

In the second chapter, I discuss factors that might play a role in a student\'s ability to take melodic dictation and put forward a taxonomy of factors that are assumed to contribute to an individual’s ability to take melodic dictation. The third chapter of the dissertation investigates individual factors that are theorized to contribute to melodic dictation using a cross-sectional experimental design. The chapter corroborates claims on the importance of understanding individual differences in working memory capacity in research on melodic dictation. The fourth chapter discusses how aural skills pedagogy can incorporate methodologies from computational musicology in order to inform teaching practice. In my fifth chapter, I introduce the MeloSol corpus, a new collection of 783 digitized melodies encoded in the **kern format. In the sixth chapter, I synthesize the previous research in a melodic dictation experiment to show how using robust statistical methods can be used to model melodic dictation. Finally, in my seventh chapter, I introduce a computational, cognitive model of melodic dictation with the goal of helping explain how students improve at melodic dictation. I demonstrate how modeling the cognitive decision process during melodic dictation helps provide a precise framework for pedagogues to understand student’s inner cognition during melodic dictation and can help inform teaching practice.


Keywords: melodic dictation, aural skills, computational musicology, cognitive psychology, pedagogy, music perception, music cognition, corpus study, computational model


I. Significance of Study
II. Context, Literature, and Rationale
III. Individual Differences
IV. Musical Differences
V. Hello, Corpus
VI. Experiment
VII. Computational Model


@DavidJohnBaker | davidjohnbaker1@gmail.com

     Return to dissertations