Date of Award

Spring 1-1-2017

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Angela D. Bryan

Second Advisor

Kent E. Hutchison

Third Advisor

Sona A. Dimidjian


Regular exercise confers tremendous benefits for health, yet nearly half of American women are insufficiently active. Increasingly, researchers have been turning to the study of affective/experiential-based determinants of exercise maintenance, but how best to address these factors and maximize health outcomes is not well understood. The novel perspective of this dissertation was that an Acceptance and Commitment Therapy (ACT)-based intervention targeting improvement on constructs of psychological flexibility (e.g., experiential acceptance, defusion, discomfort intolerance) would differentially impact exercise behavior over time.

A sample of N = 113 insufficiently active women aged 21 – 65 were randomly assigned to one of the three conditions: (1) an ACT-based health coaching intervention, (2) an education-based health coaching intervention, or (3) a no-health coaching control intervention. The study was divided into two phases, assessing exercise in terms of adoption (Phase 1) and maintenance (Phase 2). During Phase 1, all participants completed a 30-day exercise program and submitted daily exercise journals to the study team. Participants assigned to the ACT- or education-based conditions additionally took part in workshops at the beginning and end of Phase 1. All participants were contacted 3-months post-intervention for a follow-up assessment.

The central hypotheses were generally supported. Feasibility analyses revealed no between group differences, with the exception that participants assigned to the ACT-based condition rated their intervention highest on credibility/expected impact, and also tended to report greater satisfaction as compared to participants assigned to the other two conditions.

Planned contrasts tended to suggest participants assigned to the ACT-based condition completed more exercise over Phase 1 compared to participants assigned to the other two conditions, but, evidence for between condition effects on exercise maintenance at 3-months follow-up were mixed. However, our results showed that the ACT-based condition was particularly effective at improving experiential acceptance scores over the course of the intervention, and higher scores on experiential acceptance at the end of the intervention positively predicted exercise maintenance at 3-months follow-up. Future work should assess the optimal intervention “dose” for yielding the greatest impacts on experiential acceptance scores, and explore the relationship between change in experiential acceptance and behavior over a longer-term follow-up period.