Use PEA as a design vocabulary to specify policy goals, enforcement mechanisms, and authority arrangements when building LLM-based programming support¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article recommends that instead of describing a system only as a tutor, chatbot, hint generator, or feedback tool, designers specify the Policy goal it encodes, the Enforcement mechanism enacting it, and the Authority arrangement determining who can adjust it during use. A design walkthrough shows the same pedagogical goal can produce different learning conditions depending on enforcement and authority. Future directions include collaborative help routing, grounded AI literacy, and governed content generation.

Design Implications¶

Context¶

Requirements¶

• Designers must specify policy, enforcement, and authority explicitly rather than relying on generic system labels

Constraints¶

• Different configurations require different evidence, including learning outcomes, help-seeking quality, peer interaction, AI literacy, instructional alignment, and teacher workload

Target Learners¶

• Introductory programming learners
• CS educators and tool designers

Target Learning Goals¶

• Pedagogically bounded, configurable, and accountable LLM assistance for programming learning

Affordances¶

• Pea Assistance Governance Lens

Related Strategies¶

Examples¶

-

Key Sources¶

• Minsun Kim, S. Moonwara A. Monisha, Zihan Wu, David H. Smith IV. (2026). Exploring the Design Space of LLM-Based Programming Support in CS Education: A Scoping Review through the Lens of Assistance Governance. https://arxiv.org/abs/2607.21257