As of 2020, approximately 40 million adults, or 18.1% percent of the adult population, in the United States suffer from anxiety disorders, making them the country’s most common mental health condition.
Nonetheless, only 36.9% of this population receive treatment [1]. Traditionally, interventions such as pharmaceuticals, meditation and biofeedback have been used to treat anxiety disorders. However, recently, music has emerged as a powerful tool for controlling negative affective states, including stress and anxiety, due to its portability, accessibility, and unique ability to modulate mood [3]. As a new potential therapy, there is still much to be explored in the realm of music-centered anxiety research. In particular, the majority of research appears to bias towards either self-reported data or physiological measurement, and rarely assesses these measures together. To address this deficiency, we designed our study using two methods of data collection: self-reported survey data and participants’ heart rate (HR) and electrodermal activity (EDA). We sought to investigate whether we could more effectively shift users’ affective state from anxious and stimulated to calm and relaxed by modulating musical parameters, including, tempo, rhythm, and instrumentation, of a provided music piece [6], [7].
In order to do so, we created an interactive musical interface, populated with fourteen 30-45 second long musical pieces specifically written for this experiment. Each musical fragment led into the next due to gradual changes in tempo, rhythm, and instrumentation. Following a stress-inducing task, we recorded HR and EDA, while participants in the intervention group were asked to navigate through the interface and participants in the control group were asked to remain seated, quiet and still. Physiological data was supported with periodic self-reported questionnaires pertaining to stress and anxiety.
Despite the fact that participants' pre-study stress was reportedly higher in the control group, there was no significant difference in reported stress between both groups after the stressor task. At the end of the experiment, participants in the intervention group reported significantly less stress than participants in the control group. Physiological results were promising as well, though not as clear-cut, and must be considered alongside a discussion of our experimental design.