Four-phase iterative cGPT co-creation process with SPAR reflective logging

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

Description

The article describes a reusable co-creation pattern with four construction phases: Creation (writing the system prompt and selecting knowledge-base files), Refinement (iterative testing, prompt editing, and SPAR reflections), Consolidation (finalising prompts and merging versions), and Evaluation (scoring responses to ten evaluation questions). Throughout, students wrote SPAR reflections — "Stimulus (Prompt sent to GPT), Processing (GPT's response), Analysis (Evaluation of GPT response based on design principles) and Reflection (Next steps based on GPT response)" — which informed system prompt adjustment until satisfactory accuracy and instruction adherence were reached.

Design Implications

Context

Requirements

• Regular student-academic team meetings (weekly to twice weekly) for testing and refinement, and design principles focused on clear, accurate and consistent responses

Constraints

• The overall process from construction to implementation lasted two to six weeks, depending on the cGPT's topic, students' prior experience, and availability of resources such as system prompt templates

Target Learners

• undergraduate students without extensive prior research experience

Learning Goals

• AI literacy, teamwork, and project management
• iterative design and evaluative judgment

Claims

• Continued Interaction Improves Cgpt Instruction Adherence [+M]
• Cgpt Small Edits Unpredictable Response Effects [+M]

Related Designs

Examples

-

Key Sources

• Lo S, Lea T, Mavromoustakos A, Joseph B, Yii M, Exintaris B, Karunaratne N, Pearson DL, Ritchie T, Yuriev E and Pyun J (2026) Co-creating custom GPTs: an autoethnographic study of undergraduate students as partners in generative AI innovation. Front. Educ. 11:1818781. https://doi.org/10.3389/feduc.2026.1818781

Page metadata

Field	Value
type	design
id	four-phase-cgpt-co-creation-spar-pattern
title	Four-phase iterative cGPT co-creation process with SPAR reflective logging
description	The article describes a reusable co-creation pattern with four construction phases: Creation (writing the system prompt and selecting knowledge-base files), Refinement (iterative testing, prompt editing, and SPAR refl...
status	review
generated	by process:settle-candidates on 2026-10-13
trust	unverified — no one has confirmed this page's content
sources	1 — lo-s-2026

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