Iterative prompt refinement shifted a generative-AI tutor from tutorial-style explanation delivery toward dialogic Socratic support for K-12 beginners¶

Claim · All claims
Evidence · 1 study · design r? · q2
Currency · no dated study recorded · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Across four weekly revisions informed by learner feedback, the SocratiCode prompt evolved from adaptive explanatory tutorialing toward a dialogic tutoring flow with pauses, reflection, learner response requirements, and hard stopping points. → evolution-1

Evidence¶

evolution-1¶

Cassandra Lucas, Anshul Bihani, Rohini Kukka, Chun-Hua Tsai, Jaydeb Sarker, and Mia Mohammad Imran. 2026. TowardsSocratiCode: Designing a Generative AI-Based Programming Tutor for K-12 Students through a 4-Week Participatory Design Study. https://arxiv.org/abs/2605.17857 doi:10.48550/arxiv.2605.17857

q2 · i? · design · r?

Participatory design study in which two K-12 interns gave daily and weekly feedback on a GPT-based Python tutor over four weeks, with prompt revisions introduced at the start of following weeks. The authors report that "the prompt had stabilized" by W4 and that no further modifications were necessary in W4.

"By W4, the prompt had stabilized into a structured interactive flow characterized by pauses, reflection, learner response requirements, and hard stopping points."

Discussion¶

Related Claims¶

• Human oversight remained necessary alongside AI-based tutoring, especially for misconceptions and advanced topics — related
• Dialogic competence is a prerequisite for meaningful engagement with LLM-based chatbots — related
• Learner engagement was strengthened when the AI tutor adapted to learners' pace and prior knowledge — related
• Repeated negative alerts for one student prompted the teacher to revisit her and shift from explanation to inquiry, and reorganization of her engagement occurred only after this second intervention — related
• SPED-adaptive prompting shifts tutor behavior toward We-Do scaffolding, naturalistic language support, and participation-affirming feedback across three codebooks — related
• Qualitatively, Socratic prompts supported reflective monitoring while directive feedback was associated with overload and resistance — related
• Student participation in validating and iteratively refining AI-generated simulations promotes physics understanding and problem-solving skills — related
• GRPO alignment collapses Mathematical-Knowledge-for-Teaching utterances and surges Socratic-style Exploratory questions in tutor output — related