Other Titles

AI-Guided Psychodrama Materials for Psychiatric Nursing: Feasibility of AI-Validated Synthetic Cases: A Six-Stage APP Validation Report [Title Slide]

Abstract

Background: Psychodrama is widely used in psychiatric–mental health nursing to help learners recognize and reorganize dysfunctional family-of-origin or role patterns, but faculty still lack standardized, reproducible, and ethically safe scripts (Giacomucci, 2021; Beauvais, Özbas, & Wheeler, 2019; Çataldas, Atkan, & Eminoğlu, 2024). Recent work on AI in nursing education and on privacy-preserving synthetic data suggests that high-fidelity, non–patient-derived materials can be generated if outputs are quality-checked (Lifshits & Rosenberg, 2024; Cucci, 2025; Qian et al., 2024; Mendes, Barbar, & Refaie, 2025).

Purpose: To test the feasibility of an AI workflow that generates a synthetic psychodrama case, converts it into multimodal teaching assets, and releases media products only when fidelity indicators reach ≥ .80.

Methods: A six-stage workflow using a large language model and a multimodal note tool was run.

  • Stage 1 created an English scenario of an adult daughter unable to express needs to a critical mother.
  • Stage 2 expanded it into a psychodrama script with scene-by-scene directions, named roles (protagonist, mother, internalized critical voice), and explicit techniques (soliloquy, externalization, role reversal, doubling).
  • Stage 3 auto-generated narration text, a slide-style outline, and scene-level visual descriptors.
  • Stage 4 compared the original script (TEXT A) and each auto-generated product (TEXT B) on three indicators: role/character consistency, therapeutic-intention alignment, and emotional-intensity retention.
  • Stage 5 regenerated only the sections whose scores were < .80.
  • Stage 6 produced narration/audio text and scene-by-scene video prompts once all products met the ≥ .80 requirement.

Results: The pipeline ran entirely on synthetic data. Early outputs showed small losses of emotional tone and occasional omission of technique labels; targeted regeneration corrected these issues and raised all indicators to ≥ .80. Media-ready assets were aligned with the instructional outline, and indicator reports provided faculty with an auditable basis for release.

Conclusions/Implications: This workflow is an instructional-design innovation that can shorten faculty preparation, preserve core psychodrama elements, and standardize multimodal materials in psychiatric–mental health nursing. It also offers a repeatable QA loop for LMS/hybrid delivery without exposing identifiable clinical material; effectiveness claims should await expert validation and outcome data.

Notes

References:

Beauvais, A. M., Özbas, A. A., & Wheeler, K. (2019). End-of-life psychodrama: Influencing nursing students’ communication skills, attitudes, emotional intelligence and self-reflection. Journal of Psychiatric Nursing, 10(3), 103–110. https://doi.org/10.14744/phd.2019.96636

Çataldas, S. K., Atkan, F., & Eminoğlu, A. (2024). The effect of psychodrama-based intervention on therapeutic communication skills and cognitive flexibility among nursing students: A 12-month follow-up study. Nurse Education in Practice, 80, 104118. https://doi.org/10.1016/j.nepr.2024.104118

Giacomucci, S. (2021). History of sociometry, psychodrama, group psychotherapy, and Jacob L. Moreno. In S. Giacomucci (Ed.), Social work, sociometry, and psychodrama (pp. 31–52). Springer. https://doi.org/10.1007/978-981-33-6342-7_3

Lifshits, I., & Rosenberg, D. (2024). Artificial intelligence in nursing education: A scoping review. Nurse Education in Practice, 80, 104148. https://doi.org/10.1016/j.nepr.2024.104148

Cucci, F., Marasciulo, D., Romani, M., Soldano, G., Cascio, D., De Nunzio, G., Caldararo, C., Rubbi, I., Vitale, E., Lupo, R., & Conte, L. (2025). The Contribution of Artificial Intelligence in Nursing Education: A Scoping Review of the Literature. Nursing Reports, 15(8), 283. https://doi.org/10.3390/nursrep15080283

Qian, Z., MacDonald, P., Wickramasinghe, N., Denaxas, S., & Titericz, G. (2024). Synthetic data for privacy-preserving clinical risk prediction. Scientific Reports, 14, 25676. https://doi.org/10.1038/s41598-024-72894-y

Mendes, J. M., Barbar, A., & Refaie, M. (2025). Synthetic data generation: A privacy-preserving approach to accelerate rare disease research. Frontiers in Digital Health, 7, 1563991. https://doi.org/10.3389/fdgth.2025.1563991

Description

This presentation presents a six-stage AI workflow that turns a synthetic psychodrama script into validated, media-ready teaching materials for psychiatric–mental health nursing, using automated ≥ .80 checks to preserve roles, therapeutic intention, and emotional tone while reducing faculty workload.

Author Details

Yen-Chung Ho, PhD, MSN, RN; Chien Mei Sung, PhD, MSN, RN

Sigma Membership

Lambda Beta at-Large

Type

Presentation

Format Type

Text-based Document

Study Design/Type

Other

Research Approach

Other

Keywords:

Teaching and Learning Strategies, Faculty Development, Curriculum Development, Artificial Intelligence, Psychodrama

Conference Name

37th International Nursing Research Congress

Conference Host

Sigma Theta Tau International

Conference Location

Toronto, Ontario, Canada

Conference Year

2026

Rights Holder

All rights reserved by the author(s) and/or publisher(s) listed in this item record unless relinquished in whole or part by a rights notation or a Creative Commons License present in this item record. All permission requests should be directed accordingly and not to the Sigma Repository. All submitting authors or publishers have affirmed that when using material in their work where they do not own copyright, they have obtained permission of the copyright holder prior to submission and the rights holder has been acknowledged as necessary.

Review Type

Abstract Review Only: Reviewed by Event Host

Acquisition

Proxy-submission

Date of Issue

2026-07-27

Click above link to access the slide deck.

Share

COinS
 

AI-Guided Psychodrama Materials for Psychiatric Nursing: Feasibility of AI-Validated Synthetic Cases

Toronto, Ontario, Canada

Background: Psychodrama is widely used in psychiatric–mental health nursing to help learners recognize and reorganize dysfunctional family-of-origin or role patterns, but faculty still lack standardized, reproducible, and ethically safe scripts (Giacomucci, 2021; Beauvais, Özbas, & Wheeler, 2019; Çataldas, Atkan, & Eminoğlu, 2024). Recent work on AI in nursing education and on privacy-preserving synthetic data suggests that high-fidelity, non–patient-derived materials can be generated if outputs are quality-checked (Lifshits & Rosenberg, 2024; Cucci, 2025; Qian et al., 2024; Mendes, Barbar, & Refaie, 2025).

Purpose: To test the feasibility of an AI workflow that generates a synthetic psychodrama case, converts it into multimodal teaching assets, and releases media products only when fidelity indicators reach ≥ .80.

Methods: A six-stage workflow using a large language model and a multimodal note tool was run.

  • Stage 1 created an English scenario of an adult daughter unable to express needs to a critical mother.
  • Stage 2 expanded it into a psychodrama script with scene-by-scene directions, named roles (protagonist, mother, internalized critical voice), and explicit techniques (soliloquy, externalization, role reversal, doubling).
  • Stage 3 auto-generated narration text, a slide-style outline, and scene-level visual descriptors.
  • Stage 4 compared the original script (TEXT A) and each auto-generated product (TEXT B) on three indicators: role/character consistency, therapeutic-intention alignment, and emotional-intensity retention.
  • Stage 5 regenerated only the sections whose scores were < .80.
  • Stage 6 produced narration/audio text and scene-by-scene video prompts once all products met the ≥ .80 requirement.

Results: The pipeline ran entirely on synthetic data. Early outputs showed small losses of emotional tone and occasional omission of technique labels; targeted regeneration corrected these issues and raised all indicators to ≥ .80. Media-ready assets were aligned with the instructional outline, and indicator reports provided faculty with an auditable basis for release.

Conclusions/Implications: This workflow is an instructional-design innovation that can shorten faculty preparation, preserve core psychodrama elements, and standardize multimodal materials in psychiatric–mental health nursing. It also offers a repeatable QA loop for LMS/hybrid delivery without exposing identifiable clinical material; effectiveness claims should await expert validation and outcome data.