Generative AI-Assisted Personalized Feedback for Laparoscopic Skill Learning in Medical Undergraduates

Laparoscopic training involves learning to control surgical instruments while viewing movements on a screen. Feedback can help learners identify technical problems and plan further practice, but providing individualized guidance can require substantial instructor time. This study compares personalized feedback assisted by generative artificial intelligence (AI) with feedback produced using predefined rules and fixed statements. The purpose is to determine whether AI assistance provides additional benefits for laparoscopic skills learning. The investigators hypothesize that AI-assisted personalized feedback will help medical students improve laparoscopic skills more than structured rule-based feedback.

Inclusion Criteria: 1. Age ≥18 years and able to provide written informed consent. 2. Fourth-year undergraduate medical students enrolled at Shandong University Qilu Second Hospital. 3. No or limited previous laparoscopic simulation skills training, defined as no systematic completion of Fundamentals of Laparoscopic Surgery (FLS) or equivalent core laparoscopic simulation training reaching proficiency. 4. Able to complete the scheduled study procedures, including baseline assessment, training sessions, post-training assessments, and follow-up assessment. 5. Voluntary participation with written informed consent. Exclusion Criteria: 1. Previous systematic FLS training or equivalent laparoscopic simulation skills training with established proficiency. 2. Previous laparoscopic operative experience or simulation experience likely to result in substantial ceiling effects. 3. Conditions that may substantially affect fine motor skills, visuospatial ability, or safe completion of simulation tasks. 4. Any other condition considered by investigators to make participation inappropriate or unsafe.