DOBROVOLSKAYA N., KHARCHENKO A. DEVELOPMENT OF METACOGNITIVE SKILLS THROUGH DIALOGUE PATTERNS OF INTERACTION WITH AN AI ASSISTANT IN PROGRAMMING EDUCATION. LIFELONG EDUCATION: The 21st Century.
2026. Vol. 14. No. 2. DOI: 10.15393/j5.art.2026.12246


Vol. 14. No. 2.

Lifelong learning in the modern world: the research and design methodology
pdf-version

DEVELOPMENT OF METACOGNITIVE SKILLS THROUGH DIALOGUE PATTERNS OF INTERACTION WITH AN AI ASSISTANT IN PROGRAMMING EDUCATION

DOBROVOLSKAYA Natalia Yu.
PhD in Pedagogical Sciences, Associate Professor of the Department of Information Technologies
Kuban State University
(Krasnodar, Russian Federation)
dnu10@mail.ru
KHARCHENKO Anna V.
PhD in Pedagogical Sciences, Associate Professor of the Department of Information Technologies
Kuban State University
(Krasnodar, Russian Federation)
fz@mail.ru
Keywords:
large language models
artificial intelligence in education
programming instruction
metacognitive skills
prompt engineering
educational technologies.		 Abstract: the widespread integration of large language models (LLMs) into education, including programming instruction, creates the risk of students using ready-made solutions, which may hinder the development of critical metacognitive skills: planning, monitoring, and analysis. This study tests the hypothesis that pedagogically structured interaction with an AI assistant, as opposed to directive assistance, fosters deeper and more sustainable acquisition of these skills. Materials and Methods. A taxonomy of dialogue prompt patterns implementing metacognitive scaffolding strategies was developed and practically tested. The effectiveness of the patterns was evaluated through an exper-iment involving 90 first-year students studying C++, divided into three groups: an experimental group interacting with an LLM via pedagogical patterns, a control group receiving directive LLM assistance, and a group without AI support. Research Results. The results showed that the group which used patterns demonstrated statistically significantly higher scores on tests measuring the ability to apply learned patterns in unfamiliar subject areas, and log analysis of the dialogues re-vealed a qualitatively deeper thinking process. Conclusion. It is shown that the educational effect of LLMs is determined not by the technology itself, but by the pedagogical strategy of interaction, and a specific set of tools for its implementation is proposed.
Paper submitted on: 01/28/2026; Accepted on: 05/08/2026; Published online on: 06/26/2026.

References

  1. Dolinsky M. S. Directions for Using Generative Artificial Intelligence in Initial Programming Training at Universities [Electronic resource]. Komp'yuternye instrumenty v obrazovanii [Computer Tools in Education]. 2024. No. 2. P. 85–96. Electron. dan. URL: https://cyberleninka.ru/article/n/napravleniya-ispolzovaniya-generativnogo-iskusstvennogo-intellekta-pri-nachalnom-obuchenii-programmirovaniyu-v-universitetah (date of access 18.12.2025). (In Russ.)
  2. Yashina I. A. Artificial Intelligence in Teaching Programming to Pedagogical University Students [Electronic resource]. Otkrytoe obrazovanie [Open Education]. 2024. No. 4. P. 23–32. Electron. dan. URL: https://cyberleninka.ru/article/n/iskusstvennyy-intellekt-v-obuchenii-programmirovaniyu-studentov-pedagogicheskogo-vuza (date of access 18.12.2025). (In Russ.)
  3. Belenkova Yu. S. Teaching Metacognitive Skills and Methods for Assessing Their Development [Electronic resource]. Gumanitarnye, sotsial'no-ekonomicheskie i obshchestvennye nauki [Humanities, Socio-Economic and Social Sciences]. 2015. No. 3-2. P. 20–22. Electron. dan. URL: https://cyberleninka.ru/article/n/obuchenie-metakognitivnym-navykam-i-metody-otsenki-ih-sformirovannosti (date of access 18.12.2025). (In Russ.)
  4. Abdulvahabova B. B.-A., Magomedova P. K., Batchaeva Z. B. Pedagogical Conditions as a Tool for Developing the Cognitive Abilities of a University Graduate [Electronic resource]. Problemy sovremennogo pedagogicheskogo obrazovaniya [Problems of Modern Pedagogical Education]. 2024. No. 85-4. P. 7–10. Electron. dan. URL: https://cyberleninka.ru/article/n/pedagogicheskie-usloviya-kak-instrument-razvitiya-kognitivnyh-sposobnostey-vypusknika-vuza (date of access 18.12.2025). (In Russ.)
  5. Nepesova S. Development of Students' Cognitive Skills through the Integration of Digital Technologies into the Educational Process [Electronic resource]. Mezhdunarodnyy zhurnal gumanitarnykh i estestvennykh nauk [International Journal of Humanities and Natural Sciences]. 2025. No. 3-2 (102). P. 95–100. Electron. dan. URL: https://cyberleninka.ru/article/n/razvitie-kognitivnyh-navykov-studentov-cherez-integratsiyu-tsifrovyh-tehnologiy-v-obrazovatelnyy-protsess (date of access 18.12.2025). (In Russ.)
  6. Alfer'eva-Termsikos V. B. Prompt Engineering as a Strategy for Forming the Information Culture of Learners [Electronic resource]. Mezhdunarodnyy zhurnal gumanitarnykh i estestvennykh nauk [International Journal of Humanities and Natural Sciences]. 2024. No. 9-1 (96). P. 10–15. Electron. dan. URL: https://cyberleninka.ru/article/n/promt-inzhiniring-kak-strategiya-formirovaniya-informatsionnoy-kultury-obuchayuschihsya (date of access 18.12.2025). (In Russ.)
  7. Lukinsky I. S., Gorsheneva I. A. Prompt Engineering in the Educational Process and Scientific Activity or on the Issue of the Need for Training in Working with Artificial Intelligence. [Electronic resource]. Psikhologiya i pedagogika sluzhebnoy deyatel'nosti [Psychology and Pedagogy of Official Activities]. 2024. No. 4. P. 148–154. Electron. dan. URL: https://cyberleninka.ru/article/n/promt-inzhiniring-v-obrazovatelnom-protsesse-i-nauchnoy-deyatelnosti-ili-k-voprosu-o-neobhodimosti-obucheniya-rabote-s (date of access 18.12.2025). (In Russ.)
  8. Rasulova N. Yu. Multi-Agent Approach in Creating Adaptive Intelligent Tutoring Systems [Electronic resource]. Ekonomika i sotsium [Economics and Society]. 2021. No. 2-2 (81). P. 524–531. Electron. dan. URL: https://cyberleninka.ru/article/n/multiagentnyy-podhod-v-sozdanii-adaptivnyh-intellektualnyh-obuchayuschih-sistem (date of access 04.12.2025). (In Russ.)
  9. Gubasheva Kh. A., Magamedova D. M., Magazieva Z. A. Innovative Methods of Teaching Programming and IT in Russian Universities [Electronic resource]. Mezhdunarodnyj nauchno-issledovatel'skij zhurnal [International research journal]. 2022. No. 5-3 (119). P. 69–71. Electron. dan. URL: https://cyberleninka.ru/article/n/innovatsionnye-metody-obucheniya-programmirovaniyu-i-it-v-rossiyskih-vuzah (date of access 04.12.2025). (In Russ.)
  10. Skvortschevsky K. A., Dytlova O. V. Modern Adaptive and Intelligent Digital Learning Systems: Mechanisms and Potential [Electronic resource]. Voprosy obrazovaniya [Educational Studies]. 2024. No. 3 (2). P. 237–299. Electron. dan. URL: https://cyberleninka.ru/article/n/sovremennye-adaptivnye-i-intellektualnye-tsifrovye-sistemy-obucheniya-mehanizmy-i-potentsial (date of access 04.12.2025). (In Russ.)
  11. Popova Yu. B. From LMS to Adaptive Learning Systems. [Electronic resource]. Sistemnyy analiz i prikladnaya informatika [System Analysis and Applied Informatics]. 2019. No. 2. P. 58–64. Electron. dan. URL: https://cyberleninka.ru/article/n/ot-lms-k-adaptivnym-obuchayuschim-sistemam (date of access 04.12.2025). (In Russ.)
  12. Platov A. V., Gavrilina Yu. I. Artificial Intelligence in Education: Evolution and Barriers. Nauchnyy rezul'tat. Pedagogika i psikhologiya obrazovaniya [Research Result. Pedagogy and Psychology of Education]. 2024. No. 1. P. 26–43. (In Russ.)
  13. Yartseva E. Ya. Integration of Artificial Intelligence into Education. Problemy sovremennogo pedagogicheskogo obrazovaniya [Problems of Modern Pedagogical Education]. 2024. No. 85-2. P. 398–401. (In Russ.)
  14. Finnie-Ansley J., Denny P., Becker B. A. [et al.]. The robots are coming: Exploring the implications of OpenAI Codex on introductory programming. Proceedings of the 24th Australasian Computing Education Conference. 2022. P. 10–19.
  15. Kazemitabaar M., Chow J., Ma C. K. T. [et al.]. Studying the effect of AI code generators on supporting novice learners in introductory programming. Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. 2023. P. 1–23.
  16. Arora S., Narayan A., Chen M. F. [et al.]. Ask me anything: A simple strategy for prompting language models. arXiv preprint arXiv:2210.02441. 2022.
  17. Kiesler N., Schiffner D. Large language models in introductory programming education: ChatGPT’s performance and implications for assessments. arXiv:2308.08572. 2023.
  18. Pearce H., Tan B., Krishnamurthy P. [et al.]. Pop quiz! Can a large language model help with reverse engineering? arXiv:2202.01142. 2022.

Displays: 94; Downloads: 18;

DOI: http://dx.doi.org/10.15393/j5.art.2026.12246