Measurement Fidelity in AI-Mediated Physics Experimentation: A Systematic Review
Dea Nur Hafifah1, Arita Marini1, Yinghuei Chen2

1Faculty of Basic Education, Universitas Negeri Jakarta, 13220, Jakarta, Indonesia
2College of Humanities and Social Sciences, Asia University, Taichung, 41354, Taiwan

Abstract

Artificial intelligence (AI) is increasingly integrated into experimental physics for detector reconstruction, signal interpretation, and automated measurement processing. However, the extent to which AI-mediated systems preserve measurement fidelity during automated experimental interpretation remains unclear. This study systematically reviews how measurement fidelity is evaluated and preserved within AI-mediated physics experimentation systems. Using the PRISMA 2020 framework, literature published between 2020 and 2025 was retrieved from Scopus, Web of Science, and IEEE Xplore. Following screening and eligibility assessment, thirteen studies were included in the final qualitative synthesis. The studies were analyzed according to measurement context, fidelity dimension, validation method, and key limitation. Results show that fidelity was primarily evaluated through detector benchmarking, reconstruction consistency, uncertainty calibration, simulation-to-experiment comparison, and validation against independent reference measurements. Five recurring fidelity dimensions were identified: reconstruction, detector, simulation, uncertainty, and detector-response fidelity. Persistent challenges included simulation-to-experiment mismatch, detector-boundary effects, uncertainty instability, and incomplete representation of physical processes. The review concludes that measurement fidelity remains a central challenge in AI-mediated physics experimentation, highlighting the need for more experimentally grounded validation strategies to support reliable AI-assisted interpretation in increasingly automated experimental environments.

Keywords: measurement fidelity- AI-mediated experimentation- experimental validation- automated measurement interpretation- applied physics

Topic: Instrumentation and Computational Physics