Skip to main content
ABCT 2025 59th Annual Convention Main banner
Psychology Tools Ad
Final Program


Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Presentation Icons
Recorded Session
Recorded Session
Ticketed Session
Ticketed Session
ViewAttendees 2

Symposia

Technology/Digital Health

Artificial Intelligence (AI) as a means of supporting evidence-based practices in psychological treatments

3 - (SYM 66) Role Play or Real Play? Development of a Machine Learning-based Course with Behavioral Rehearsal and Feedback for Clinicians Using DBT Strategies with Simulated Clients

Sunday, November 23, 2025
2:32 PM - 2:46 PM CST
Location: Imperial 11, Level 4
Keywords: Technology / Mobile Health, Training / Training Directors, Competence
Recommended Readings: Stade, E., Stirman, S. W., Ungar, L. H., Yaden, D. B., Schwartz, H. A., Sedoc, J., ... & DeRubeis, R. (2023). Artificial intelligence will change the future of psychotherapy: A proposal for responsible, psychologist-led development.,   Olawade, D. B., Wada, O. Z., Odetayo, A., David-Olawade, A. C., Asaolu, F., & Eberhardt, J. (2024). Enhancing mental health with Artificial Intelligence: Current trends and future prospects. Journal of medicine, surgery, and public health, 100099.,   Kopelovich, S. L., Slevin, R., Brian, R. M., Shepard, V., Baldwin, S. A., Ben-Zeev, D., ... & Imel, Z. (2025). Preliminary investigation of an artificial intelligence-based cognitive behavioral therapy training tool. Psychotherapy.,   Imel, Z. E., Pace, B., Pendergraft, B., Pruett, J., Tanana, M., Soma, C. S., ... & Atkins, D. C. (2024). Machine Learning–Based Evaluation of Suicide Risk Assessment in Crisis Counseling Calls. Psychiatric services, 75(11), 1068-1074.,   NARAYANAN, A., & KAPOOR, S. (2024). AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference. Princeton University Press. https://doi.org/10.2307/jj.14736606

    Speaker(s)

  • AC

    Adam Carmel, Ph.D.

    Clinical Professor
    University of Washington, Seattle
    Seattle, WA, United States

Abstract Body


EBP adherence and competence requires active and ongoing learning strategies, such as in vivo coaching, behavioral rehearsal, and direct observation with performance-based feedback. This is particularly the case for EBPs that are more challenging to deliver due to having multiple strategies and nuances stylistic approaches. 

 

Dialectical Behavior Therapy (DBT) is the gold standard treatment for borderline personality disorder and has disseminated on a relatively large scale for the treatment of suicidal behaviors. DBT contains multiple modes of treatment and a range of strategies woven in, utilizing a range of nuanced stylistic approaches to engage a difficult-to-treat clinical population. Thus, DBT can be as challenging to learn as it is to deliver with fidelity. 

 

Advances in machine learning (ML) and artificial intelligence (AI) show promise in strengthening models of training to increase clinician adherence and competence. This project will discuss the development of a ML and AI-based platform to learn DBT.  Considerations that went into the development of client profiles and vignettes will be discussed in order for participants to hone their skills in utilizing a range of strategies to intervene with a range of problematic in-session behaviors. The selection and application of 6 DBT modules used in the course will be discussed: Describing Problems Specifically, Emotion Focus, Accurate Reflection, Validating as Reasonable in the Moment, Direct Confrontation, and Modeling Dialectical Thinking. The process of building a coding scale to capture fidelity and establishing calibration among raters will be discussed. Finally, we will highlight the challenges of evaluating participants’ demonstration of stylistic strategies such as therapist irreverence and validation techniques and the provision of corrective feedback. Considerations for use of ML and AI compared to the standard models of DBT training will be discussed.