AI-Based Estimation of the Zone of Proximal Development in Programming Education

Nagy, Bendegúz, Horváth, Győző (2026) AI-Based Estimation of the Zone of Proximal Development in Programming Education In: Proceedings of the 13th International Conference on Applied Informatics. Eger, Eszterházy Károly Catholic University Líceum Publisher. pp. 202-215.

pdf ICAI2026-pp202-215.pdf Download (2MB) [error in script]

Absztrakt (kivonat)

The Zone of Proximal Development (ZPD), introduced by Vygotsky [11], describes the range of learning tasks that learners can successfully complete with appropriate support. Closely related to this concept, the notion of flow experience emphasizes the balance between task challenge and learner skill as a key condition for sustained engagement and intrinsic motivation [1]. Recent research suggests that artificial intelligence (AI) can support ZPDoriented adaptation in modern educational environments by dynamically adjusting task difficulty and feedback [10]. However, practical and lightweight implementations of such approaches in programming education remain limited. This paper presents a proof-of-concept AI-based system for estimating learners’ ZPD using a minimal set of background information and task interaction data. The approach records simplified learner profiles from task interaction data – task success, error frequency, and completion time – and derives a skill estimate from task success and difficulty; this estimate defines a difficulty band that a large language model uses to generate subsequent programming tasks at an adjusted difficulty level. The system is developed as part of the continued evolution of the Codia gamified programming education platform, building on earlier findings related to student motivation in secondary school programming education [6]. The design focuses on feasibility and pedagogical plausibility rather than predictive accuracy. The study examines whether a simplified AI-supported approach can approximate ZPD-aligned task selection in introductory programming education.

Mű típusa:	Könyvrészlet - Book section
Szerző:	Szerző neve Email MTMT azonosító ORCID azonosító Közreműködés Nagy, Bendegúz NEM RÉSZLETEZETT NEM RÉSZLETEZETT NEM RÉSZLETEZETT Szerző Horváth, Győző NEM RÉSZLETEZETT NEM RÉSZLETEZETT NEM RÉSZLETEZETT Szerző
Kapcsolódó URL-ek:	Befoglaló kötet
Kulcsszavak:	zone of proximal development, adaptive learning, ELO rating, large language models, programming education, task generation
Nyelv:	angol
DOI azonosító:	10.17048/icai.2026.202
Felhasználó:	Tibor Gál
Dátum:	22 Szep 2026 07:53
Utolsó módosítás:	22 Szep 2026 07:53
URI:	http://publikacio.uni-eszterhazy.hu/id/eprint/9447

Műveletek (bejelentkezés szükséges)

Tétel nézet