AI-Supported Empathy Mapping to Enhance Communication and Grit in Pediatric Nursing Students

The nurse-patient communication environment in pediatric care is characterized by high uncertainty and complexity. Due to children's limited language development and emotional regulation abilities, coupled with parents' high level of involvement, nursing students often experience anxiety, lack of confidence, and avoidance behaviors, which negatively affect their clinical learning outcomes and the establishment of therapeutic relationships. Therefore, providing effective communication support strategies is essential in pediatric nursing education. This study aims to implement an instructional scaffolding model using artificial intelligence (AI)-generated empathy maps to enhance the communication skills, empathy performance, and grit of nursing students during pediatric clinical practicums when encountering communication challenges.

A mixed-methods research design was adopted, and the participants were third-year nursing students enrolled in a pediatric nursing practicum course. The teaching intervention included AI-assisted generation of age-appropriate communication strategies, the construction of a grit-oriented empathy map, small group scenario-based exercises, and the application of learned strategies in clinical settings. Quantitative data were collected using pre- and post-intervention assessments, including an empathy scale, a communication skills scale, and a grit scale, to evaluate changes in learning outcomes. Qualitative data, including reflective journals, clinical observations, and focus group interviews, were analyzed to explore students' learning processes and strategy adaptations. Triangulation was applied to strengthen the validity of the findings.

It is anticipated that this teaching model will enhance students' understanding of pediatric patients' emotional needs, strengthen their communication strategy application and clinical interaction quality, and promote persistence and adaptability in challenging situations. Through evidence-based teaching practice, this study is expected to provide a feasible and scalable innovative instructional model that supports the effective integration of AI into clinical nursing education, thereby improving pediatric nursing competence and the quality of care for children.