Integration of Generative AI Character Mechanism and Virtual Reality Clues for the Development of Educational Games for Clinical Reasoning Training in Nursing Care

Ward nursing staff must collect information for new patients and perform clinical reasoning for health assessment and nursing diagnosis, which requires realistic ward situations and case studies.Combining game mechanics with standardized patient simulations is expected to improve trainees' communication skills and problem-solving abilities and enhance motivation.This study investigates the effectiveness of an educational game combining GenAI and virtual reality in training nursing diagnostic reasoning.We developed a GenAI dual-role mechanism (GenAI playing the role of a patient uncle Sam and a senior nurse sister) and constructed a virtual reality clues room space through SVVR.The learning objectives of the game: learners take on the role of a nurse and interact through dialogues to perform health assessments and clinical reasoning on Uncle Sam, as well as consulting with another experienced GenAI nurse to gain expertise and solve problems.The study participants were 16 formal nurse practitioners in Taiwan.The results of this study showed that the study was highly useful in that the learners had a high flow and did not have a high external cognitive load, but only an internal cognitive load slightly above the median of the five-point Likert scale (i.e., 3), and that the difficulty of the tasks could be considered for simplification in the future.The qualitative feedback indicated that the mechanism could create realistic clinical situations (81.25%), improve interview assessment skills (93.75%), facilitate clinical reasoning (75%), reduce stress (87.5%), and increase concentration (81.25%).