Tool, Tutor, or Crutch?: A Grounded Theory of Cognitive Scaffolding and Offloading in AI-Assisted Programming Education

Synthesis: This Constructivist grounded-theory study (N = 24 AI-enabled + 17 contrast undergraduates in Java programming) builds a process-level model of how learners navigate the tension between "Domain Mastery" (conceptualization, explanation, evaluation) and "Tool Mastery" (procedural efficiency with AI) through two recurrent loops — Scaffolding and Offloading — interpreted through cognitive load theory and self-determination theory. It explains how performance and affect gains can co-occur with thinner germane processing and attenuated metacognitive calibration under routine offloading.

Key Findings

1. A core tension between Domain Mastery and Tool Mastery. Learners balance conceptual understanding (Domain Mastery) against procedural efficiency with AI (Tool Mastery), switching dynamically via a "Strategic Dance."
2. Two recurrent loops: Scaffolding and Offloading. The tension model centers on a Scaffolding Loop (AI support that preserves learning) and an Offloading Loop (routine delegation that reduces germane processing), with boundary conditions (time pressure, task complexity/familiarity, scaffolding design) shaping movement between them.
3. Trust-but-Can't-Verify and a Boilerplate Blindspot. Novices struggle to verify AI output; more experienced students develop a blindspot for boilerplate/generic code — two distinct evaluation challenges.
4. Attenuated metacognitive calibration. A mismatch between perceived readiness and independent capability co-occurs with sustained offloading, echoing overconfidence research.
5. Concrete instructional strategies. The model proposes dedicated 'critique-the-AI' phases, planned fading of AI assistance through offline tasks, verification journals, and contrastive prompting — shifting the debate from 'use or ban' to how and when the tool aligns with learning goals.

Implications

This is a direct theoretical contribution to the wiki's Cognitive Offloading thread in programming education, providing a process-level account of why performance and affect gains coexist with thinner learning — the scaffolding-vs-offloading distinction operationalizes the "coach vs. crutch" boundary the wiki documents. The proposed interventions (critique-the-AI, planned fading, verification journals) offer concrete, testable levers aligned with Reducing AI Misuse and AI Literacy — moving beyond bans toward calibrated, self-regulated AI use. The finding connects to Measuring LLM Tutors Teach Vs Solve and Tutoring Specific Vs General AI.

Connected Concepts

• Generative AI
• CS Education
• Cognitive Offloading
• Scaffolding
• Metacognition
• Self Regulated Learning
• AI Literacy
• Reducing AI Misuse

Connected Articles

• AI Making Us Stupid — AI's cognitive effects / overconfidence
• Coach Not Crutch AI Writing — AI as coach not crutch in writing
• Measuring LLM Tutors Teach Vs Solve — Whether LLM tutors teach or solve
• Tutoring Specific Vs General AI — Tutoring-specific vs general AI
• Jost LLM Programming Education Learning Outcomes — LLM reliance and grades in coding
• Stromberg Generative AI Learning Penalty Secondary 2026 — Generative AI learning penalty

Citation

Liu, D., Fan, G., & Pan, L. (2026). Tool, tutor, or crutch?: a grounded theory of cognitive scaffolding and offloading in AI-assisted programming education. International Journal of STEM Education, 13, 10.