Exploring the Impacts and Challenges of Vibe Coding Paradigm to Children's Programming Learning and Practices

Programming Education & Computational ThinkingChildren's AI Literacy & Data LiteracyHuman-LLM CollaborationK-12 TeachersUniversity Professors & ResearchersEarly Childhood Educators

Paper Title

Exploring the Impacts and Challenges of Vibe Coding Paradigm to Children's Programming Learning and Practices

Publication Info

  • Topic area: The application of generative AI in children's programming education, focusing on the vibe coding paradigm.
  • Keywords: Vibe coding, generative AI, Scratch, children's programming, computational thinking, educational technology, child-centered design, programming challenges, creative ideation, AI-assisted learning.

Background and Problem

  • Problem / challenge: Existing research on vibe coding focuses primarily on adult users in text-based programming environments, leaving a gap in understanding how children engage with this paradigm in block-based environments like Scratch. Challenges include children's developing cognitive and linguistic abilities, which may not align with the assumptions of vibe coding systems.
  • Significance: Understanding how vibe coding impacts children's programming learning and practice is crucial for designing educational technologies that support computational thinking (CT), creativity, and learner autonomy.
  • Motivation and related work: Prior studies have explored GenAI tools for programming, focusing on efficiency and creative outcomes for adults. Some tools have extended vibe coding to Scratch, but these efforts remain tool-specific and overlook broader impacts on children's learning behaviors and challenges. This paper addresses these gaps by examining vibe coding as a paradigm in children's programming education.

Solution

  • Proposed approach: The study investigates the impacts and challenges of vibe coding on children's programming learning and practices through workshops with children and interviews with Scratch teachers, using the GenAI-supported tool Stax.
  • Novelty:
    1. Extends vibe coding research to child-centered programming contexts, identifying mismatches between system design and children's developmental needs.
    2. Examines vibe coding's influence across three programming contexts: concept acquisition, knowledge application, and project creation.
    3. Identifies challenges children face at different interaction stages and proposes design implications for child-centered GenAI tools.
  • Procedure and key techniques:
    • Conducted three workshops with 41 children (ages 6–11) grouped by Scratch experience.
    • Used Stax, a vibe coding tool integrated with Scratch, to observe interactions.
    • Collected data through video recordings, prompt logs, and focus group interviews.
    • Conducted semi-structured interviews with five Scratch teachers to triangulate findings.
    • Analyzed data using thematic coding, interaction logs, and computational thinking frameworks.

Results

  • Concrete findings:
    • Positive impacts: Increased self-efficacy, improved problem-solving efficiency, and expanded creative ideation.
    • Negative impacts: Reduced exploration motivation, confused conceptual construction, and weakened knowledge consolidation.
    • Challenges: Difficulty articulating goals, interpreting AI outputs, debugging AI-generated code, and balancing AI support with independent problem-solving.
  • Advantage over baselines: Vibe coding enabled rapid prototyping, creative exploration, and immediate feedback, which traditional Scratch environments lack. However, it introduced unique challenges not present in manual coding.
  • Experiments / evaluation:
    • Workshops: Observed children's interactions with Stax across three programming contexts (concept acquisition, knowledge application, project creation).
    • Teacher interviews: Provided pedagogical insights and contextualized children's challenges.
    • Metrics: Analyzed prompt lengths, response times, and interaction patterns.
  • Limitations and future work:
    • Participant homogeneity: All children were from a single city in China.
    • Tool-specific findings: Results may reflect Stax-specific features.
    • Short-term study: Long-term impacts of vibe coding remain unexplored.
    • Future work: Extend research to diverse populations, compare multiple vibe coding tools, and investigate long-term effects on learning.

Summary

This study explores the impacts and challenges of the vibe coding paradigm in children's programming using the GenAI-supported tool Stax. Findings reveal both positive effects, such as enhanced creativity and efficiency, and negative effects, including reduced exploration and conceptual confusion. Challenges arise across three interaction stages: goal formulation, information interpretation, and evaluation. The study highlights the need for child-centered design in GenAI tools, proposing scaffolding strategies, exploratory outputs, and multimodal interfaces to better align with children's developmental needs. These insights contribute to the design of educational technologies that balance AI assistance with meaningful learning.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/chi/223401/2026

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3772318.3790366
At a Glance

Paper Snapshot

fact_check
dataset
Source
CHI
calendar_month
Year
2026
emoji_events
Award
No award tagged
group
Authors
9 authors
sell
Subtopics
Programming Education & Computational Thinking, Children's AI Literacy & Data Literacy, Human-LLM Collaboration
work
Professions
K-12 Teachers, University Professors & Researchers, Early Childhood Educators
article
Content Status
Full text indexed
hub
Related Papers
5 related papers