Artificial Intelligence and Science Education: Assessing ChatGPT’s Understanding of Thermal Phenomena

PDF (ελληνικά)
Published: Apr 19, 2026
DOI: https://doi.org/10.12681/codiste.9893
Keywords:
artificial intelligence (AI) heat misconceptions science education temperature
Georgia Vakarou
University of Ioannina
https://orcid.org/0009-0001-1235-8275
Georgios Stylos
University of Ioannina
https://orcid.org/0000-0001-8036-8427
Kostas Georgopoulos
University of Ioannina
Konstantinos Kotsis
University of Ioannina
https://orcid.org/0000-0003-1548-0134
Abstract

Artificial intelligence (AI) has already begun to shape educational processes, with the potential to offer personalized learning and direct assistance to students. This study examines the effectiveness of AI, and especially ChatGPT, in understanding and teaching fundamental concepts such as heat and temperature. Through the Thermal Concept Evaluation (TCE) questionnaire, the research evaluates the scientific answers provided by ChatGPT and identifies any alternative perceptions. The study highlights the importance of critical thinking when using artificial intelligence technologies in science teaching, as the answers it provides are not always accurate and free from common misconceptions. Although there is an improvement in ChatGPT performance at a later stage, the limitations presented by AI technology highlight the need for continuous development and careful integration in training.


Article Details
  • Section
  • Προφορικές Ανακοινώσεις
Downloads
Download data is not yet available.
References
Αχλαδιανάκης, Α. & Παππά, Ε. (2019). Τα chatbots και οι δυνατότητες που παρέχουν στο μάρκετινγκ. Υλοποίηση και ενσωμάτωση ενδεικτικού chatbot σε ένα σύστημα διαχείρισης περιεχομένου, (Αδημοσίευτη Πτυχιακή Εργασία). Τεχνολογικό Εκπαιδευτικό Ίδρυμα, Kρήτη. http://hdl.handle.net/20.500.12688/9236
Βασιλικού, Μ. Π. (2023). Το ChatGPT περνάει στις Πανελλήνιες. Οι μαθητές;. Στο Α. Καζανίδης, Α. Τσινάκος (Επιμ.), Πρακτικά Εργασιών 13ου Πανελλήνιου και Διεθνούς Συνεδρίου «Οι ΤΠΕ στην Εκπαίδευση», Διεθνές Πανεπιστήμιο της Ελλάδας, Καβάλα, 29 Σεπτεμβρίου – 1 Οκτωβρίου 2023. ISSN: 2529-0916, ISBN: 978-618-83186-8-7. Ανακτήθηκε στις 01/11/2024 από: https://www.etpe.gr/wp-content/uploads/2024/07/HCICTE23_510-516.pdf
Κασσέτας, Α.Ι. (2004). Το μήλο και του κουάρκ: Διδακτική της Φυσικής. Αθήνα: Σαββάλας
Μάγειρα, Κ. & Πιττού, Δ. (2021). Σχεδίαση και Ανάπτυξη Εκπαιδευτικού Λογισμικού Συζητήσεων με Τεχνητή Νοημοσύνη (Ai Chatbot) με Πολιτιστικό Περιεχόμενο, (Αδημοσίευτη Μεταπτυχιακή Εργασία). Πανεπιστήμιο Αιγαίου, Μυτιλήνη. http://hdl.handle.net/11610/22825
Adiguzel, T., Kaya, M. H., & Cansu, F. K. (2023). Revolutionizing education with AI: Exploring the transformative potential of ChatGPT. Contemporary Educational Technology, 15(3), ep429. https://doi.org/10.30935/cedtech/13152
Ali, D., Fatemi, Y., Boskabadi, E., Nikfar, M., Ugwuoke, J., & Ali, H. (2024). ChatGPT in Teaching and Learning: A Systematic Review. Education Sciences, 14(6), 643. https://doi.org/10.3390/educsci14060643
Almasri, F. (2024). Exploring the impact of artificial intelligence in teaching and learning of science: A systematic review of empirical research. Research in Science Education, 54, 977–997. https://doi.org/10.1007/s11165-024-10176-3
Bahak, H., Taheri, F., Zojaji, Z., & Kazemi, A. (2023). Evaluating ChatGPT as a question answering system: A comprehensive analysis and comparison with existing models. arXiv preprint arXiv:2312.07592. https://doi.org/10.48550/arXiv.2312.07592
Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency, 610-623. https://doi.org/10.1145/3442188.3445922
Çakır, M., & Sarıkaya, M. (2016). Misconceptions about heat and temperature. Journal of Education and Training Studies, 4(5), 105-113. https://doi.org/10.11114/jets.v4i5.1490
Kortemeyer, G. (2023). Could an artificial-intelligence agent pass an introductory physics course?. Physical Review Physics Education Research, 19(1), 010132. https://doi.org/10.1103/PhysRevPhysEducRes.19.010132
Kotsis, K. T. (2024a). Integrating artificial intelligence in science education: Benefits and challenges. International Journal of Educational Innovation, 6(3), 39-49. https://doi.org/10.69685/ICAS1772
Kotsis, K. T. (2024b). ChatGPT as Teacher Assistant for Physics Teaching. Journal of Effective Teaching Methods, 2(4), 18-27. https://doi.org/10.59652/jetm.v2i4.283
Luckin, R., Holmes, W., Griffiths, M., & Forcier, L. B. (2016). Intelligence unleashed: An argument for AI in education. Pearson. Ανακτήθηκε στις 28/11/2024 από: https://static.googleusercontent.com/media/edu.google.com/en//pdfs/Intelligence-Unleashed-Publication.pdf
OpenAI. (2024). GPT-4 technical report. Ανακτήθηκε στις 05/11/2024 από: https://openai.com/research
Ribeiro, M. T., & Lundberg, S. (2022). Adaptive testing and debugging of NLP models. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (τ. 1: Long Papers), 3253–3267. Association for Computational Linguistics. https://doi.org/10.18653/v1/2022.acl-long.230
Roll, I., & Wylie, R. (2016). Evolution and revolution in artificial intelligence in education. International Journal of Artificial Intelligence in Education, 26(2), 582-599. https://doi.org/10.1007/s40593-016-0110-3
Stylos, G., Sargioti, A., Mavridis, D., & Kotsis K.T. (2021). Validation of the thermal concept evaluation test for Greek university students’ misconceptions of thermal concepts. International Journal of Science Education, 43(2), 247-273. https://doi.org/10.1080/09500693.2020.1865587
Suits, J. P., & Mestre, J. P. (2016). Misconceptions in physics and the physics concepts inventory. Στο International Handbook of Research in History, Philosophy and Science Teaching, σσ. 1261-1298.
Springer, Dordrecht. Ανακτήθηκε στις 28/11/2024 από: https://www.academia.edu/125092827/International_Handbook_of_Research_in_History_Philosophy_and_Science_Teaching
Theophilou, E. et al. (2023). Learning to Prompt in the Classroom to Understand AI Limits: A Pilot Study. Στο R. Basili, D.Lembo, C. Limongelli, A. Orlandini, (Επιμ.) AIxIA 2023 – Advances in Artificial Intelligence. Lecture Notes in Computer Science, τ. 14318. Springer, Cham. https://doi.org/10.1007/978-3-031-47546-7_33
Tong, D., Tao, Y., Zhang, K. et al. (2023). Investigating ChatGPT-4’s performance in solving physics problems and its potential implications for education. Asia Pacific Education Review, 25, 1379–1389. https://doi.org/10.1007/s12564-023-09913-6
West, C. G. (2023). Advances in apparent conceptual physics reasoning in GPT-4. Physical Review Physics Education Research, 19(1), 010132. https://doi.org/10.1103/PhysRevPhysEducRes.19.010132
Yang, X., & Evans, C. (2021). The limits of artificial intelligence in scientific education: Examining the understanding of heat and temperature. Physics Education, 56(3), 035011. https://doi.org/10.1088/1361-6552/abc123
Yeo, S., & Zadnik, M. (2001). Introductory thermal concept evaluation: Assessing students’ understanding. The Physics Teacher, 39(8), 496–504. https://doi.org/10.1119/1.1424603