Building bridges: Exploration of music analysis methods in improvisational music therapy research and clinical work

Abstract

In music therapy, where music is the principal medium of both expression and communication, the use of an appropriate musical analysis method for clinical material is of key importance. An effective approach to music analysis allows accurate assessment of the current state of a client, and adjustment of clinical methods in order to optimise treatment. There are currently a number of analysis methods available, but very little guidance regarding factors that are important in the selection of a specific method for a specific case. This dissertation explores common music analysis methods that are used in music therapy clinical work and research. The goals are 1) to establish how the use of various music analysis methods and the results they provide compare to each other, 2) to explore the factors influencing music therapists' choice, use, and opinion of music analysis in music therapy and 3) to explore computational music analysis in the assessment of depression. This dissertation is based on three studies (a mixed methods study, a qualitative study and a quantitative study), which resulted in four articles. Article I, triangulation of music analysis methods, introduces a novel approach to the classification of music analysis methods. Article II, grounded theory on the use of music analysis methods, proposes that professional freedom – tension between professional responsibilities and creative impulses – is a key factor influencing analysis and its methods in music therapy. Article III, computational music analysis in depression, and Article IV, computational music analysis and emotion in depression, establish a set of links between computationally extracted musical features, depression and felt emotions. Additionally, this dissertation proposes a set of recommendations for using music analysis for clinicians and researchers, and a comparison of the availability, accessibility, efficiency and relevance of various methods. As a body of work, the findings of this project suggest that data triangulation of musical and psychological measures shows the greatest potential for music therapy assessment. Keywords: music therapy, aural music analysis, notation-based music analysis, computational music analysis, data triangulation