augmented reality

AI feedback improved answers during an augmented-reality physics lab, while independent transfer was not assessed

Atakan Coban, David Dzsotjan, Stefan Küchemann, Jürgen Durst, Jochen Kuhn, Christoph Hoyer

EPJ Quantum Technology

A physics laboratory study compared feedback and no-feedback questions within an augmented-reality quantum-cryptography activity involving 21 student groups. GPT-4 feedback helped groups improve their immediate explanations, while gaze records suggested changes in attention to relevant representations and equipment. The measured outcome was answer improvement during supported discussion. A small voluntary sample, exploratory gaze analyses and no independent delayed test limit broader claims about learning and the value of the combined technology.