PD049 - A LONGITUDINAL EVALUATION OF AN INTEGRATED ARTIFICIAL INTELLIGENCE (AI)- AND ACTOR-BASED SIMULATION TRAINING MODEL IN NUTRITION EDUCATION

PD049

A LONGITUDINAL EVALUATION OF AN INTEGRATED ARTIFICIAL INTELLIGENCE (AI)- AND ACTOR-BASED SIMULATION TRAINING MODEL IN NUTRITION EDUCATION

D. Chermon^1,*, A. Gur-Arie¹, V. Kaufman-Shriqui¹, R. Foraiter², N. Dgania-Yaroslaviz¹, Y. Haber³, K. Hod¹

¹Nutrition sciences, ²Teaching Innovation, Ariel University, Ariel, ³Bar-Ilan, Ramat Gan, Israel

Rationale: Whether integrating AI chatbot simulations with actor-based simulations within a structured training sequence enhances self-efficacy while reducing performance anxiety and facilitating interpersonal risk-taking among nutrition students remains unclear, despite growing adoption of AI in clinical education.

Methods: A quasi-experimental study was conducted among third-year undergraduate nutrition students(n = 29). Over four weeks, participants completed an AI chatbot-only simulation, an actor-based simulation, and two combined chatbot-actor sessions using the same and different clinical scenarios. Self-efficacy, performance anxiety, and interpersonal risk were assessed repeatedly before and after sessions. Longitudinal trends and simulation modality effects were examined using linear mixed-effects models.

Results: Self-efficacy increased significantly over time(p < 0.001). Both chatbot-based and actor-based simulations were associated with higher self-efficacy compared with pre-simulation assessments(p<0.001), with no significant difference between modalities. Performance anxiety was significantly lower following chatbot-based simulations compared with both pre-simulation and actor-based simulations(p<0.001). Interpersonal risk ratings were higher following chatbot-based simulations compared with pre-simulation assessments(p=0.008). Simulation order did not influence outcomes.Additionally, active participation in actor-based simulations was associated with greater self-efficacy gains compared with observation. 

Conclusion: An integrated, longitudinal simulation model combining AI chatbot and actor-based modalities enhances self-efficacy while reducing performance anxiety in nutrition education. AI-based simulations provide a psychologically safe foundation for early skill development, supporting a complementary approach to preparing students for clinical practice.

Disclosure of Interest: None declared