Design and Effect Evaluation of Vocational Education Training System Based on Virtual Reality

Virtual Reality (VR) offers a powerful medium for bridging the gap between theoretical knowledge and hands-on skill acquisition in vocational education. This paper presents a VR-based training system that integrates adaptive learning, real-time feedback, and dynamic content generation to deliver personalized and scalable instruction. The system is anchored by two key components: Persona Skill Net, which models learners’ cognitive and behavioral states through hierarchical encoding and graph-based task representations for individualized competence tracking; and Reflective Curriculum Flow, which adapts task sequencing and difficulty via feedback-driven optimization and stability regularization. Together, these modules form a closed-loop learning environment that aligns instructional content with evolving learner proficiency. Experiments on benchmark educational datasets demonstrate significant improvements in accuracy, recall, and engagement over state-of-the-art adaptive learning methods. This work offers a robust, data-driven framework for enhancing realism, interactivity, and pedagogical effectiveness in next-generation VR-based vocational education.