Personalized Learning Based on Individual ChatGPT-Feedback in an MR-enhanced quantum cryptography experiment in physics lab course

Quantum cryptography is a central topic in the field of quantum technologies and is particularly important for secure communication. However, the abstract and complex nature of quantum physics makes it difficult to understand. Augmented Reality (AR) offers significant potential for improving quantum physics education in university lab courses by visualizing abstract concepts and enabling interactive learning. Additionally, providing personalized feedback on challenging concepts can further enhance learning, and Large Language Models (LLMs) like ChatGPT are effective in delivering such feedback. This study combines AR and formative feedback via LLMs and explores the impact of an AR-based quantum cryptography experiment. During the experimentation process in the AR environment, we record students’ gaze data to investigate the relevance of virtual and real experimental components for cognitive learning processes. The study involved 38 students in a physics laboratory course of a German university. To investigate the effect of the feedback a crossover design was used. Therefore the students were divided into two groups and the groups received feedback on different questions. Statistical analysis compared scores between feedback and non-feedback questions, and the effects of ChatGPT feedback on eye-tracking data were examined. The results showed that ChatGPT feedback significantly improved learning outcomes and guided the visual attention on relevant virtual and real experimental components depending on task demands.