Computational Thinking Development in AI Agent Creation_A Mixed-Methods Study

2026-05-13

AI Summary

This study investigates the developmental mechanisms and individual differences in computational thinking (CT) among young learners engaged in no-code AI agent creation. Drawing on a mixed-methods approach—including pre- and post-tests, behavioral logs, and interviews—with 93 upper-primary students participating in a five-day CocoFlow workshop, the research reveals significant gains in abstraction and algorithmic thinking (Cohen’s d ≈ 0.70). Furthermore, engagement in iterative testing positively predicts growth in self-efficacy. A key contribution is the identification of a “zone of optimal development” effect: students with moderate initial CT proficiency demonstrated the greatest improvement (η² = 0.55), challenging assumptions of linear learning progression and underscoring the necessity of differentiated instructional scaffolding.

Abstract

This mixed-methods study examined computational thinking (CT) development among 93 pre-high school students in a five-day AI agent creation workshop using CocoFlow, a no-code platform. Integrating pre-post assessments, behavioral logs, and interviews, we investigated CT development and how initial CT levels shape learning trajectories. Results revealed significant improvements in abstract thinking (effect size d = 0.71) and algorithmic thinking (effect size d = 0.70). Hierarchical regression identified iterative testing engagement as a predictor of self-efficacy gains (beta = 0.20, p = 0.05). Notably, students with moderate initial CT levels demonstrated substantially greater gains than both high-CT and low-CT peers, revealing an Optimal Development Zone effect (eta squared = 0.55). Qualitative analysis showed moderate-CT students exhibited adaptive expertise, while high-CT students risked over-engineering and low-CT students struggled with task decomposition. These findings challenge linear learning assumptions and provide evidence for differentiated scaffolding in CT education.

Problem

• Computational Thinking
• AI Agent Creation
• Learning Trajectories
• Optimal Development Zone
• Differentiated Scaffolding

Innovation

• computational thinking
• AI agent creation
• no-code platform
• Optimal Development Zone
• mixed-methods study