Asking questions matters

Comparing the effect of learning through dialogue with peers, ChatGPT, and intelligent tutoring system

Bibliographic Data

ID21631385
AuthorsJi Won Yang (0000-0002-9885-7701, Seoul National University), Kiroong Choe (0000-0002-6084-2539, Seoul National University), Jaeseo Lim (0000-0003-0978-9098, Jeonju University), Jooyong Park (0000-0002-5113-5111, Seoul National University, corresponding author)
Year2025
Volume33
Issue10
Pages5870-5886
Publication date2025-11-26
Open AccessNo
TypeARTICLE
VenueInteractive Learning Environments (JOURNAL)
Journal identifiersISSN: 1049-4820 • E-ISSN: 1744-5191
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1080/10494820.2025.2488984
OpenAlexW4409514093
LanguageEN
Citations received2
References cited36

The ways students learn through dialogue are diversifying, in terms of whom they engage with and how. With the recent advancement of AI chatbots like ChatGPT, learners can leverage aspects of dialogue, such as asking and answering questions, without a peer or instructor being present. Given the lack of experimental studies validating their effect, this study compares three dialogue conditions involving Peer, ChatGPT, and Intelligent Tutoring System. A chat interface was developed where students could exchange questions with a Peer, ask questions to a GPT agent, or answer questions displayed by an ITS agent. After reading a biology passage, 114 university students from Republic of Korea solved two practice items and then engaged in dialogue with their respective counterparts about what they had studied. Finally, they were instructed to re-answer the practice items and solve two new isomorphic test problems. The Peer and GPT groups, where students could ask questions, demonstrated significantly higher post-dialogue performance in both absolute score and conditional correct rate compared to the ITS group, which only answered questions. The number of questions asked and the level of trust in the dialogue were positively correlated with individual differences in performance. Our findings highlight the importance of asking questions over asking many when learning through dialogue, laying the groundwork for designing more effective interactive learning systems

Educational technology · Human–computer interaction · Intelligent tutoring system · Mathematics education · Multimedia · AI in Service Interactions · Computer Science · Innovative Teaching and Learning Methods · Intelligent Tutoring Systems and Adaptive Learning · Psychology

Tools

Open DOI