Could not find session

2026 ASEE Annual Conference & Exposition

Examining student perceptions of custom GPT Chatbots to support introductory physics instruction

Presented at Engineering Physics and Physics Division (EP2D) Technical Session 3

Introductory physics courses, often required for STEM majors, typically require a minimum grade of C for students to progress. Those who fall short must retake the course. While individualized feedback is crucial for mastering core concepts, student demand often exceeds what instructors can reasonably offer. Existing computer-graded assignments can help but often only provide a one-size-fits-all approach to student feedback. As such, there is a need to provide students with both additional and immediate personalized feedback and guidance to ensure they meet their learning and educational goals. With generative artificial intelligence tools becoming increasingly widespread and the barriers to entry becoming decreasingly prohibitive, chatbots can be leveraged to provide personalized help and promote student success.

Here, we used ChatGPT’s custom GPT feature to develop chatbots for two introductory physics courses at a large, research-intensive university in the Southeastern United States. Each chatbot was solely trained on course materials including the syllabus, handouts, homework solutions, and practice exams. The customized chatbot then serves as an intelligent virtual tutor, providing 24/7 support by answering queries, guiding students through problem-solving, and reinforcing theoretical concepts.

The chatbots were offered as optional resources in three sections of a calculus-based, studio-style introductory physics for engineers course and two sections of an algebra-based, lecture-based introductory physics course. Students could use them for any part of the course except for in-class exams and were told that their instructor could not access the chatbot’s history or logs.

To test the chatbot’s efficacy, we administered an end-of-semester survey on how often students used the chatbot, the types of queries they asked, how often the chatbot was able to resolve their requests, and general feedback. This paper will present the results of that survey and suggestions for future improvements to better support student learning in important pre-requisite courses.

Authors

Dr. Nicholas Young http://orcid.org/https://0000-0003-1396-416X University of Georgia [biography]

Nicholas Young is an assistant professor of physics with a focus on physics education research at the University of Georgia. He earned his PhD in physics and computational mathematics, science, and engineering at Michigan State University, and he was a postdoctoral scholar at the University of Michigan’s Center for Academic Innovation. His current research focuses on assessment in introductory physics courses and graduate education in physics.
Jasmine Freeman University of Georgia [biography]

Jasmine Freeman is a first-year PhD physics student with a focus on quantum technology in condensed matter physics and physics education research at the University of Georgia. She earned her bachelor's degree, in physics and astronomy, at Valdosta State University.
Dr. Sarangan Balasubramaniam http://orcid.org/0000-0003-0053-8196 The University of Georgia [biography]

Sarangan Balasubramaniam is a Lecturer in the Department of Statistics at the University of Georgia, where he has taught since 2022. He holds a Ph.D. in Computational Mathematics from the University of North Carolina Greensboro, with a dissertation focused on spatial prediction for axially symmetric processes on the sphere. His research interests span spatial statistics, machine learning, statistical learning, and biostatistics. At UGA, he teaches courses ranging from introductory statistics to advanced graduate-level data science, and M.S. in Applied Data Science program. He is also an affiliated faculty member of the UGA Statistical Consulting Center.
Dr. Nandana Weliweriya University of Georgia [biography]

Nandana Weliweriya is a Senior Lecturer in the Department of Physics and Astronomy at the University of Georgia, Athens. His teaching and research focus on physics problem-solving, active learning pedagogies, and undergraduate laboratory instruction.

Note

The full paper will be available to logged in and registered conference attendees once the conference starts on June 21, 2026, and to all visitors after the conference ends on June 24, 2026

« View session