Preprint · Research Square · September 1, 2026
The comparative evaluation of explanations for algorithmic programming problems provided by a Human vs. ChatGPT: an exploratory online user study
Alfarabi Imashev, Zhanna Abisheva, Inara Tusseyeva, Zhanibek Kozhirbayev, Tohid Alizadeh, Bakhyt T. Matkarimov
Abstract
Abstract The rapid advancement of generative artificial intelligence (AI) systems like ChatGPT has the potential to revolutionize education. However, using these technologies requires an understanding of the factors that motivate students to use them. Investigations on the acceptability of ChatGPT have mainly focused on academic environments, resulting in a lack of exploration within school contexts. This research examines the gap by surveying 37 upper secondary students. The aim of this paper is twofold: to figure out whether students are able to distinguish codes (solutions) provided by GPT over the ones provided by humans and to evaluate how students may react to and perceive a 3D avatar as a tutor explaining how to solve non-trivial programming tasks. A 3D avatar that "explained" the solution codes was provided to students having experience in programming: some solutions were ChatGPT-driven, and others were human-provided solutions from www.codeforces.com. Previous programming (starting age, highest olympiad participation level) and experience of previous interaction with avatars were supposed to be important for their acceptance of such technologies and will of future use, but the user study revealed that in this case these factors have no significant effect on participants' decisions. According to feedback, participants were quite optimistic about the future of such technologies, however also emphasized the demerits of proposed avatars. The findings may guide educators and policymakers seeking to promote the thoughtful and successful integration of generative AI in teaching.
More by Zhanibek Kozhirbayev
- Sequential Domain Adaptation of Whisper for Kazakh Academic Lecture Speech RecognitionLecture notes in computer science · 2026
- Enhancing Personalised Learning with Graph-Based Ensemble Prediction and Skill Cluster Mapping for Student Knowledge CompletenessComputers · 2026
- Evaluating Large Language Models for Text-to-Gloss Translation in Kazakh-Russian Sign Language: A Pilot StudyLanguage Resources and Evaluation Conference · 2026
- An Empirical Comparison of Cascade and Direct End-to-End Speech Translation for Low-Resource Language PairComputers · 2026 · Citations: 2