Date of Award


Degree Name

Doctor of Philosophy


Interdisciplinary Health Sciences

First Advisor

Dr. Linda I. Shuster

Second Advisor

Dr. John M. Spitsbergen

Third Advisor

Dr. David J. Colson


Music improvisation, brain, oxytocin, music therapy, social bonding, flow


The three studies in this dissertation explore the biological and phenomenological under-pinnings of music improvisation related to bonding, empathy, and flow. Musical improvisation is a multi-dimensional behavior encompassing motor, cognitive, affective, and often social domains, and as such, it is reasonable to consider that music improvisation shares many characteristics with other forms of non-musical extemporaneous behavior, such as carrying on a verbal dialogue. The primary aims of this three-study dissertation were to (1) determine the feasibility of the study designs and interdisciplinary methods; (2) examine potential differences in the neuroanatomical and neurophysiological correlates of improvisation between non-musicians and musicians; and (3) better understand the lived experiences of the participants during musical and verbal improvisation related to social bonding, emotional synchrony, and flow.

Study 1 used functional-magnetic resonance imaging (fMRI) to examine brain activity associated with free and interactive improvisation on a keyboard in a musically naïve child. Study procedures were determined to be feasible as the participant was able to carry out all study tasks as directed and tolerated the procedures for the duration of data collection. Brain areas associated with self-awareness were significantly deactivated (p < .001) during free improvisation as well as during interactive improvisation; however, they were coupled with activations of areas associated with expressive language and empathy during interactive improvisation (p < .001).

Study 2 examined the molecular underpinnings of group instrumental music improvisation, as well as verbal improvisation, in musicians and non-musicians using blood-based measurements of oxytocin, vasopressin, dopamine, and cortisol with verbal improvisation serving as a control. Again, all participants (N = 8) were able to successfully complete study tasks as directed and tolerate the blood draws. On average, regardless of musicianship status, males had a greater change in cortisol than females (males M = -.033; females M = .025) after improvising musically (p < .04). Males also had a significant difference between the decrease in cortisol after music improvisation (M = -.033) compared to an increase in cortisol after verbal improvisation (M = .026; p < .01).

In Study 3, emotional synchrony and flow were examined using group instrumental vs. verbal improvisation paradigms. No statistically significant differences were found between musicians and non-musicians for either condition and mean scores on the Perceived Emotional Synchrony and Shared Flow scales were all above 4.00 (on a scale from 1 to 7) indicating that all participants experienced reasonably high levels of synchrony and flow. Interview data indicated that members from each group felt more connected to each other as the musical experience progressed.

The research designs and methods employed across the three studies were determined to be feasible and appropriate for full-scale follow-up implementation. They provide a guide to structure interdisciplinary methods of studying health-related outcomes in populations served by music therapists.

Access Setting

Dissertation-Open Access