Skip to main content
ABCT 2024 58th 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
Back
Evaluation ViewAttendees 4

Technology/Digital Health
Symposia

Symposium 54 - Generative Artificial Intelligence in PTSD Treatment: Exploring Different Use Cases of Novel Technologies

Friday, November 15, 2024
2:30 PM – 3:30 PM EST
Location: Liberty B, Level 3, Marriott Downtown
CE: 1

Keywords: Technology / Mobile Health, PTSD (Posttraumatic Stress Disorder), CBT
Level of Familiarity: Basic to moderate
Recommended Readings: Stade, E. C., Stirman, S. W., Ungar, L. H., Boland, C. L., Schwartz, H. A., Yaden, D. B., Sedoc, J., DeRubeis, R., Willer, R., & Eichstaedt, J. C. (2023). Large language models could change the future of behavioral healthcare: A proposal for responsible development and evaluation. PsyArXiv. https://doi.org/10.31234/osf.io/cuzvr,   Stirman, S. W., Gutner, C. A., Gamarra, J., Suvak, M. K., Vogt, D., Johnson, C., Wachen, J. S., Dondanville, K. A., Yarvis, J. S., Mintz, J., Peterson, A. L., Young-McCaughan, S., & Resick, P. A. (2021). A Novel Approach to the Assessment of Fidelity to a Cognitive Behavioral Therapy for PTSD Using Clinical Worksheets: A Proof of Concept With Cognitive Processing Therapy. Behavior Therapy, 52(3), 656–672. https://doi.org/10.1016/j.beth.2020.08.005,   Stirman, S. W., Gutner, C. A., Suvak, M. K., Adler, A., Calloway, A., & Resick, P. A. (2018). Homework Completion, Patient Characteristics, and Symptom Change in Cognitive Processing Therapy for PTSD. Behavior Therapy, 49(5), 741–755. https://doi.org/10.1016/j.beth.2017.12.001,   ,  

Trauma-focused cognitive behavioral therapies (CBTs), such as Cognitive Processing Therapy (CPT), Prolonged Exposure (PE), and Written Exposure Therapy (WET) are recommended PTSD treatments. Research has demonstrated that the delivery of CBTs with fidelity increases their effectiveness. Fidelity ratings generally involve the review of sessions recordings by a clinical supervisor who then provides the treating clinician with feedback on the delivery of the intervention. Effective training can also further enhance the delivery of CBTs. Trainings are commonly facilitated by experts combining didactic and experiential methods. Key challenges with these outlined processes that facilitate the effective delivery of CBTs are that 1) patients often do not find cognitive restructuring activities (e.g., worksheet completion) engaging, 2) the review of sessions to obtain fidelity ratings or provide personalized guidance for clinicians is extremely time consuming, costly, and therefore not scalable, 3) real-time feedback to help clinicians learn is often limited to the trainings but drastically reduces thereafter, potentially impacting skill development.

Artificial intelligence (AI) offers promise in supporting these aspects of effective CBTs and have the potential to offer more engaging and ultimately effective treatment. For example, users can be engaged in Socratic dialogue via generative AI tools, which may thereby facilitate cognitive restructuring in a more engaging manner than commonly used worksheets. Likewise, using labeled data, AI tools can be trained to conduct fidelity ratings of PE sessions on their own, thereby increasing the efficiency of training efforts and providing the opportunity for previously trained therapists to self-monitor their competencies. Generative AI tools can be nearly universally accessible and can be developed to help train clinicians who want to implement WET, providing clinicians with real-time feedback of how to improve the delivery of the intervention, even when they do not have access to expert trainers. However, due to the sensitive nature of psychotherapy with trauma survivors, close collaboration with patients and clinicians in the development of generative AI tools as well as rigorous feasibility and safety testing is essential.

In this symposium Drs. Held, Sherrill, and Wiltsey-Stirman will detail three distinct use cases of generative AI tools: engaging in Socratic dialogue, characterizing PE delivery including fidelity ratings, and supporting WET training through simulation exercises with feedback, respectively. The presentations will cover theoretical background and detailed information about the development process for each of these applications. Several presentations will also include initial feedback from users who have interacted with early versions of these tools to gauge their feasibility. Dr. Dondanville will discuss the three presentations and critically examine the roles, potential benefits, and challenges of using generative AI in PTSD treatment. Ethical considerations, integration concerns, and future research directions will be discussed to foster a comprehensive understanding of how AI can augment how PTSD treatments are learned and delivered.

Learning Objectives:

At the end of this session, the learner will be able to:
  1. Explore the potential of generative AI to facilitate specific therapeutic tasks and support therapists in delivering effective PTSD treatment.
  2. Describe steps taken to ensure generative AI tools are feasible and safe to use for both patients and therapists.
  3. Evaluate ethical considerations, potential benefits, challenges, and integration concerns associated with using generative AI in treatment.
  4. Analyze user feedback and empirical evidence related to the effectiveness of generative AI tools in enhancing patient engagement and outcomes.
  5. Identify future research directions and technological advancements needed to improve the scalability and impact of generative AI in treatment.

Presentations:

  • 2:30 PM – 3:30 PM EST
    (SYM 54) Socrates 2.0: A Novel Generative Artificial Intelligence Tool to Facilitate Socratic Dialogue

    Speaker: Philip Held, Ph.D. (he/him/his) – Rush University Medical Center

    Co-author: Sarah Pridgen, M.A. – Rush University Medical Center

    Co-author: Yaozhong Chen, B.S. – Rush University Medical Center

    Co-author: Zuhaib Ahtar, M.S. – Rush University Medical Center

    Co-author: Darpan Amin, B.S. – Rush University Medical Center

    Co-author: Sean Pohorence, PhD – Independent Researcher

  • 2:30 PM – 3:30 PM EST
    (SYM 54) A Generative Ai-based Simulated Patient for Post-workshop Written Exposure Therapy Therapist Practice and Support

    Speaker: Shannon Wiltsey Stirman, Ph.D. – National Center for PTSD and Stanford University

    Co-author: Stade Elizabeth, PhD (she/her/hers) – Stanford University

    Co-author: Aadesh Salecha, M.S. (he/him/his) – Stanford University

    Co-author: Courtney Worley, PhD – National Center for PTSD, Dissemination and Training Division

    Co-author: Stefanie T. LoSavio, ABPP (she/her/hers) – University of Texas Health Science Center at San Antonio

    Co-author: Zoe Tait, B.A. (she/her/hers) – Stanford

    Co-author: Johannes Eichstaedt, PhD (he/him/his) – Stanford University

    Co-author: Craig Rosen, PhD – National Center for PTSD, Dissemination and Training Division

    Co-author: Debra Kaysen, ABPP, Ph.D. (she/her/hers) – Stanford University

  • 2:30 PM – 3:30 PM EST
    (SYM 54) Teaming with Artificial Intelligence (AI) to Learn and Maintain Clinical Skills in Prolonged Exposure: Development of TEAMMAIT (Trustworthy, Explainable, and Adaptive Monitoring Machine for AI Teams)

    Speaker: Andrew M. Sherrill, Ph.D. (he/him/his) – Emory University

    Co-author: Saeed Abdullah, Ph.D. – The Pennsylvania State University

    Co-author: Rosa Arriaga, PhD – Georgia Institute of Technology

    Co-author: Christopher Wiese, PhD (he/him/his) – Georgia Institute of Technology