Crystallizing Theory Evaluation: A Human–AI Collaborative Framework for Multi‐Perspective Nursing Theory Analysis

The NLN/Jeffries Simulation Theory (JST) occupies an important and debated position in nursing education. It is widely used for its practical value but often criticized for lacking theoretical clarity. Traditional evaluation methods rely on single theoretical perspectives and have not resolved this contradiction, leading to fragmented and incomplete assessments. This study presents a Human-AI collaborative framework for comprehensive meta-theoretical analysis and uses JST as a demonstration case. The framework is based on the concept of crystallization and combines evaluation criteria from Fawcett, Meleis, and Walker and Avant. Artificial intelligence is used to organize and systematically analyze complex and multidimensional theoretical evidence, while the human researcher performs interpretation and synthesis. The findings show that JST's lack of precision is not a weakness but a strength that allows it to function as a boundary object. It serves as a flexible structure that supports collaboration across nursing education, research, and practice. This study provides a practical and transparent model for using Human-Centered Artificial Intelligence in academic research. It shows how technological tools can augment rather than displace the critical, interpretive judgment of nursing scholars.