Teaching note describes an innovative virtual patient simulation for practicing motivational interviewing skills in graduate students, enhancing their learning experience.
Teaching motivational interviewing (MI) skills requires repeated practice with realistic patient interactions, yet traditional approaches – standardized patients and peer role-plays – face barriers of cost, scalability, and consistency. This Teaching Note describes the design and classroom integration of “Sam,” a ChatGPT-powered virtual patient simulation developed to provide graduate students with on-demand MI practice and real-time, individualized feedback. Built on GPT-4o architecture within OpenAI’s Custom GPT builder and grounded in experiential learning principles, Sam simulated a patient navigating ambivalence about smoking cessation, responding dynamically to students’ use of open-ended questions, affirmations, reflections, and summaries (OARS). Fifty interdisciplinary behavioral health doctoral students in an online clinical course engaged with Sam across two structured assignments, each requiring a minimum of ten conversational exchanges. Preliminary student feedback suggests the tool provided a low-stakes environment for iterative skill-building, with students reporting increased confidence in MI techniques and appreciation for the realism of Sam’s responses. The pedagogical rationale, technical development, and implementation steps of this project are detailed to enable other educators in clinical social work and allied health fields to adapt this approach for their own programs.
No takes yet. Share an insight, caveat, or question.
Cordes et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: