Supporting self-directed learning and self-assessment using TeacherGAIA, a generative AI chatbot application: Learning approaches and prompt engineering
The DOI matches its Crossref record; 14 of the 14 references it cites were found in a registry or an index.
Abstract
Self-directed learning and self-assessment require student responsibility over learning needs, goals, processes, and outcomes. However, this student-led learning can be challenging to achieve in a classroom limited by a one-to-many teacher-led instruction. We, thus, have designed and prototyped a generative artificial intelligence chatbot application (GAIA), named TeacherGAIA, that can be used to asynchronously support students in their self-directed learning and self-assessment outside the classroom. We first identified diverse constructivist learning approaches that align with, and promote, student-led learning. These included knowledge construction, inquiry-based learning, self-assessment, and peer teaching. The in-context learning abilities of large language model (LLM) from OpenAI were then leveraged via prompt engineering to steer interactions supporting these different learning approaches. These interactions contrasted with ChatGPT, OpenAI’s chatbot which by default engaged in the traditional transmissionist mode of learning reminiscent of teacher-led instruction. Preliminary design, prompt engineering and prototyping suggested fidelity to the learning approaches, cognitive guidance, and social-emotional support, all of which were implemented in a generative AI manner without pre-specified rules or “hard-coding”. Other affordances of TeacherGAIA are discussed and future development outlined. We anticipate TeacherGAIA to be a useful application for teachers in facilitating self-directed learning and self-assessment among K-12 students.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Reciprocal Teaching of Comprehension-Fostering and Comprehension-Monitoring Activities
Palinscar, A. S., & Brown, A. L. (1984) · Cognition and Instruction · doi:10.1207/s1532690xci0102_1 · Cited by 5,186
Phases of inquiry-based learning: Definitions and the inquiry cycle
Pedaste, M., Mäeots, M., Siiman, L. A., de Jong, T., van Riesen, S. A. N., Kamp, E. T., Manoli, C. C., Zacharia, Z. C., & Tsourlidaki, E. (2015) · Educational Research Review · doi:10.1016/j.edurev.2015.02.003 · Cited by 2,119 · cited 1×
“…learn knowledge new to them by undergoing phases of asking questions, posing hypotheses/claims, investigating involving evidence work of gathering, synthesizing, and inferring, as well epistemic agency to decide on, reflect on, and share, various aspects of the inquiry (Pedaste et al., 2015).”
BackgroundWhy Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts
Zamfirescu-Pereira, J. D., Wong, R. Y., Hartmann, B., & Yang, Q. (2023) · Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems · doi:10.1145/3544548.3581388 · Cited by 1,274 · cited 1×
“However, we were guided by emerging best practices for prompt engineering in the rapidly evolving field (Mishra et al., 2023; White et al., 2023; Zamfirescu-Pereira et al., 2023).”
MethodMeta-Analysis of Inquiry-Based Learning: Effects of Guidance
Lazonder, A. W., & Harmsen, R. (2016) · Review of Educational Research · doi:10.3102/0034654315627366 · Cited by 967 · cited 1×
“For example, teacher guidance in the form of explanations, heuristics (suggestions) and scaffolds (directing students on how to think and what to do) tend to have the largest impact on performance success compared to other forms of teacher guidance (Lazonder & Harmsen, 2016).”
BackgroundInteracting with educational chatbots: A systematic review
Kuhail, M. A., Alturki, N., Alramlawi, S., & Alhejori, K. (2023) · Education and Information Technologies · doi:10.1007/s10639-022-11177-3 · Cited by 841
A Prompt Pattern Catalog to Enhance Prompt Engineering with ChatGPT
White, J., Fu, Q., Hays, S., Sandborn, M., Olea, C., Gilbert, H., Elnashar, A., Spencer-Smith, J., & Schmidt, D. C. (2023) · arXiv · doi:10.48550/arxiv.2302.11382 · Cited by 838
Do AI chatbots improve students learning outcomes? Evidence from a meta‐analysis
Wu, R., & Yu, Z. (2024) · British Journal of Educational Technology · doi:10.1111/bjet.13334 · Cited by 576 · cited 2×
“A recent meta-analysis highlighted the significant benefits of chatbots in education (Wu & Yu, 2023).”
BackgroundMethodConstructive activity and learning in collaborative small groups.
Webb, N. M., Troper, J. D., & Fall, R. (1995) · Journal of Educational Psychology · doi:10.1037/0022-0663.87.3.406 · Cited by 474
Jeong, H., Hmelo-Silver, C. E., & Jo, K. (2019) · Educational Research Review · doi:10.1016/j.edurev.2019.100284 · Cited by 339 · cited 1×
“Meta-analyses suggest that knowledge construction approaches can lead to deeper learning and more enduring outcomes compared to passive transmission of knowledge (Erdem, 2022; Fiorella & Mayer, 2016; Jeong et al., 2019).”
Background
Cited by
Zhai, X. (2025) · Journal of Science Education and Technology · doi:10.1007/s10956-024-10174-0 · Cited by 184
Zhan, Y., & Yan, Z. (2026) · Assessment & Evaluation in Higher Education · doi:10.1080/02602938.2025.2471821 · Cited by 162
“These features enable GenAI tools to deliver personalised, specific, interactive, and timely feedback through engaging, adaptable, and human-like conversational interactions (Ali et al. 2023).”
Generative artificial intelligence as an enabler of student feedback engagement: a framework
Zhan, Y., Boud, D., Dawson, P., & Yan, Z. (2025) · Higher Education Research & Development · doi:10.1080/07294360.2025.2476513 · Cited by 94
Tang, Q., Deng, W., Huang, Y., Wang, S., & Zhang, H. (2025) · Journal of Computer Assisted Learning · doi:10.1111/jcal.70027 · Cited by 55
Generative AI and educational assessments: a systematic review
Graduate School of Education, T. U. O. W. A., Zhao, J., Chapman, E., Sabet, P. G. P., & Global Curtin, C. U. (2024) · Education Research and Perspectives · doi:10.70953/erpv51.2412006 · Cited by 45
Metadata
- Authors
- Ali, F., Choy, D., Divaharan, S., Tay, H. Y., & Chen, W.
- Venue
- Learning: Research and Practice
- Publisher
- Informa UK Limited
- Published
- 2023-07-03
- Volume
- 9
- Issue
- 2
- Pages
- 135-147
- ISSN
- 2373-5082 (ISSN-L)
- 2373-5090
- OpenAlex
- W4387024875
- Cited by
- 99
- References
- 14
- FWCI
- 23.26
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Supporting self-directed learning and self-assessment using TeacherGAIA, a generative AI chatbot application: Learning approaches and prompt engineering. https://app.cochat.ai/papers/10.1080/23735082.2023.2258886
- The paper
- Ali, F., Choy, D., Divaharan, S., Tay, H. Y., & Chen, W. (2023). Supporting self-directed learning and self-assessment using TeacherGAIA, a generative AI chatbot application: Learning approaches and prompt engineering. Learning: Research and Practice. https://doi.org/10.1080/23735082.2023.2258886
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.