Please use this identifier to cite or link to this item: https://repository.isls.org/handle/1/11936
Title: Generative Failure: Learning Design in Programming with Generative AI
Authors: Kondo, Hideki
Tohyama, Sayaka
Iseri, Shuntaro
Ishii, Masaaki
Keywords: CSCL
Issue Date: 2025
Publisher: International Society of the Learning Sciences
Citation: Kondo, H., Tohyama, S., Iseri, S., & Ishii, M. (2025). Generative Failure: Learning Design in Programming With Generative Ai. In Oshima, J., Chen, B., Vogel, F., & Järvelä, J. (Eds.), Proceedings of the 18th International Conference on Computer-Supported Collaborative Learning - CSCL 2025 (pp. 625-627). International Society of the Learning Sciences.
Series/Report no.: Proceedings of the 18th International Conference on Computer-Supported Collaborative Learning - CSCL 2025
Abstract: This study proposes the Generative Failure Method (GFM) to address challenges in applying Productive Failure to novice programming. Beginners often struggle with syntax errors, rarely compare diverse solutions, and lack tasks aligned with interests. Leveraging large language models (LLMs), GFM helps novices overcome initial barriers, adapt project goals, and incrementally cultivate open-ended exploration. Rather than imposing open-ended tasks, GFM expands problem contexts as learners progress. A pilot study found that LLM support enabled learners to produce functional applications, engage in iterative trial-and-error, and reframe “failure” as a constructive resource for learning.
Description: Poster
URI: https://doi.org/10.22318/cscl2025.784519
https://repository.isls.org//handle/1/11936
Appears in Collections:ISLS Annual Meeting 2025

Files in This Item:
File SizeFormat 
CSCL2025_625-627.pdf306.23 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.