Implement AI-peer integrated feedback as a standardized two-phase revision cycle with rubric-aligned prompting and instructor monitoring¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article's implementable procedure: students draft an essay, submit it to ChatGPT-4 with a standardized core prompt targeting the four rubric dimensions, revise, then complete conventional rubric-based peer review with small-group discussion, and revise again. The core prompt asked the AI to "Provide constructive feedback on this essay focusing on task achievement, coherence and cohesion, lexical resource, and grammatical range and accuracy". The instructor periodically reviewed students' interaction logs, focusing on "the quality and appropriateness of their prompting strategies" to detect ineffective prompting practices.

Design Implications¶

Context¶

Requirements¶

• A 45-minute rubric training session and 30-minute peer feedback training before the intervention
• Standardized core prompts with supplementary prompts restricted to the four pre-defined assessment dimensions
• Individual AI accounts to ensure equal access privileges

Constraints¶

• The procedure was implemented in intact EFL writing classes with a single instructor over five biweekly writing tasks; AI interaction was capped at approximately 30 minutes per revision cycle

Target Learners¶

• tertiary EFL students
• English majors in advanced writing courses

Target Learning Goals¶

• L2 argumentative writing revision
• engagement with feedback across task achievement, coherence and cohesion, lexical resource, and grammatical range and accuracy

Affordances¶

• Sct Ai Peer Feedback Scaffolding Framework

Claims¶

Related Strategies¶

• Guided Peer Response using PeerMark™
• Revision And Resubmission
• Revision Cycles
• Peergrade
• Integrate AI feedback as a supportive resource within process-oriented drafting cycles, with explicit guidance on critical evaluation
• Use ChatGPT as a supplementary tool for grammar, writing, and conversational practice so teachers can focus on higher-order thinking skills, while integrating human interaction to offset emotional detachment
• Collaborative Human-AI Revision Design embedding prompt engineering across drafting, feedback, and rewriting phases
• Classroom exercise cycle: select a task, formulate a baseline prompt, submit it to an LLM, then critically analyse the output before revising instructions
• Implement an AI-Peer-Teacher hybrid integrated feedback model staged across drafting, revision, and polishing phases
• Scaffolded LLM-collaboration assignment cycle: draft, LLM critique, then rebut or incorporate

Examples¶

-

Key Sources¶

• Liu W (2026) AI-peer integrated feedback in second language writing classes: exploring students' engagement and writing performance. Front. Psychol. 17:1871951. https://doi.org/10.3389/fpsyg.2026.1871951