Author ORCID Identifier

https://orcid.org/0000-0003-2097-4868

Date of Award

Winter 1-31-2025

Document Type

Thesis (Master's)

Department or Program

Computer Science

First Advisor

Michael Casey

Second Advisor

Lorie Loeb

Third Advisor

Tam Vu

Abstract

This study investigates the integration of real-time physiological data with AI-generated music to enhance emotional well-being, stress regulation, and focus, using Heart Rate Variability (HRV) as a biomarker of autonomic function. Conducted in two phases—Stable Audio Open (SAO) and Suno (SUNO)—the research evaluates biofeedback-driven music interventions across varying daily music-listening habits.

In the SAO phase, short AI-generated instrumental tracks were compared with Spotify recommendations and guided meditation. Modest HRV improvements were observed in biofeedback conditions, but participants noted emotional limitations, citing short track lengths and abrupt transitions.

The SUNO phase addressed these limitations with longer, more complex AI-generated compositions combined with mixed Spotify and AI recommendations. Notably, the condition with four AI-generated tracks (0S4G) resulted in a significant HRV decrease ($p = 0.0006$), indicating heightened sympathetic activation and reduced parasympathetic modulation. Additionally, AI-generated calming tracks significantly promoted parasympathetic activation, evidenced by a median RMSSD increase of 2.48 ($p < 0.05$) compared to the 0.79 increase observed in energizing tracks.

Stratifying participants based on daily music-listening duration (0--1, 2--3, and 4--5 hours) revealed high inter-individual variability, particularly among those with very low or high daily listening habits. These findings underscore the critical role of personal familiarity and preference in physiological responses, highlighting that AI-generated music may fall short in evoking the emotional resonance required for relaxation.

These mixed results point to the need for deeper personalization and dynamic adaptation to listeners’ preferences and physiological states. While statistical trends support the potential of targeted music interventions to positively influence autonomic regulation, challenges persist in achieving emotional authenticity and sustained HRV improvements.

Future research should focus on refining AI models for greater emotional depth, extending session durations, and integrating multimodal biometric data to optimize personalized, real-time interventions for emotional and physiological well-being.

Download

Share

COinS