Human-in-the-loop human–AI–human collaboration workflow for critical prompt engineering

Design · All designs
Evidence · 2 claims (2 for) · 1 study (1 design), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description

A six-step structured workflow introduced in Week 3 to scaffold critical human–AI–human collaboration: cognitive activation before prompting, prompt formulation, AI output generation, critical evaluation with fact and bias checking as gatekeeping, iterative refinement via new prompts, and group consensus on final internalization. The article states that "Drawing on the principle of cognitive activation, students were instructed to generate their own initial ideas before querying the AI", a step explicitly intended to mitigate offloading and reduce anchoring effects. Groups applied the workflow in a TikTok script co-creation task using DeepSeek.

Design Implications

Context

Requirements

• Groups treat AI outputs as provisional materials requiring fact-checking and scrutiny for potential bias rather than authoritative answers
• Students generate their own initial ideas before querying the AI

Constraints

Target Learners

• ethnic minority preparatory students in higher education

Learning Goals

• critical evaluation of AI-generated content
• prompt engineering skill

Claims

• Scaffolded Strategy Talk Repurposed Critical Evaluation [+M]
• Prompt Self Efficacy Significant Increase [+M]

Related Designs

Examples

-

Key Sources

• Deliang Wang and Cunling Bian. (2026). Scaffolding Critical Engagement with GenAI: Transforming Ethnic Minority Preparatory Students’ Collaborative Discourse in Prompt Engineering Tasks. https://arxiv.org/abs/2607.28630

Page metadata

Field	Value
type	design
id	human-in-the-loop-critical-collaboration-workflow
title	Human-in-the-loop human–AI–human collaboration workflow for critical prompt engineering
description	A six-step structured workflow introduced in Week 3 to scaffold critical human–AI–human collaboration: cognitive activation before prompting, prompt formulation, AI output generation, critical evaluation with fact and...
status	review
generated	by process:settle-candidates on 2026-10-11
trust	unverified — no one has confirmed this page's content
sources	1 — deliang-wang-and-cunling-bian-2026

Frontmatter is the page's machine-readable record — see the Schema & Guide.