Design of AI-Powered Tool for Self-Regulation Support in Programming Education
Project Overview
The document examines the innovative use of generative AI in education through the introduction of CodeRunner Agent, an AI-driven tool aimed at improving self-regulated learning (SRL) within programming courses. By integrating seamlessly with the Moodle learning management system (LMS), CodeRunner Agent delivers personalized, context-sensitive feedback to students, thereby fostering a more tailored learning experience. Additionally, it enables educators to customize the feedback based on specific course materials, enhancing the instructional process. This tool addresses significant challenges associated with existing large language model (LLM)-based programming assistants, particularly in terms of LMS integration and the cultivation of self-regulation skills among students. Overall, the findings highlight the potential of generative AI to not only support individualized learning but also to empower educators in their teaching strategies, ultimately leading to improved educational outcomes in programming education.
Key Applications
CodeRunner Agent
Context: University programming courses, specifically targeting freshman students learning programming.
Implementation: CodeRunner Agent is integrated into Moodle LMS, leveraging the CodeRunner plugin to provide contextual feedback based on students' learning logs and programming tasks.
Outcomes: Enhances self-regulated learning by providing tailored feedback and scaffolding strategies, while also allowing educators to customize feedback based on course materials.
Challenges: Students may become overly reliant on LLMs, which can hinder the development of self-regulation and problem-solving skills. Additionally, existing tools often operate in isolation from LMS, limiting their effectiveness.
Implementation Barriers
Integration Barrier
Many LLM-based tools operate independently from institutional LMS, creating a disconnect between the tool and educational context.
Proposed Solutions: Integrate LLM-powered tools like CodeRunner Agent within LMS platforms to ensure alignment with course objectives and facilitate tracking of student interactions.
Skill Development Barrier
Students may become overly reliant on AI tools, potentially hindering the development of self-regulated learning skills.
Proposed Solutions: Implement pedagogically sound scaffolding within AI tools to promote independence and critical thinking in problem-solving.
Project Team
Huiyong Li
Researcher
Boxuan Ma
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Huiyong Li, Boxuan Ma
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai