Skip to content

A dual-focus framework for preparing ChatGPT prompts for training, distinguishing general LLM prompting principles from application-specific pedagogical principles

Design · All designs
Evidence · 1 claim (1 for) · 1 study (1 causal), q3 · 0 of 1 report an effect size · 1 claim rests on one study

Description

The article introduces a framework for structuring ChatGPT prompts used as training tools, built on "a dual focus, incorporating general LLM prompting principles, which apply universally across multiple contexts, and application-specific (pedagogical) principles tailored to particular training scenarios". The general principles cover goal and scene setting, structured instruction, and regular reiteration of objectives to counter token-limit context loss. The pedagogical principles cover personalised responses, proactive knowledge checking via follow-up questions, and phrasing of output format such as bullet points. The framework was demonstrated by equipping ChatGPT with an AV user manual for conversation-based driver training.

Design Implications

Context

Requirements

  • ChatGPT must be instructed to base responses solely on the provided training material and not draw from external knowledge sources, to avoid inconsistent or invented information

Constraints

  • The article notes the brief ~1100-word manual "does not adequately represent the complexities of processing larger documents", which could challenge ChatGPT's token capacity and increase the risk of inaccurate or invented information

Target Learners

  • adult licensed drivers learning ADAS and AV functions

Learning Goals

  • accurate operation of advanced driver assistance functions
  • formation of accurate mental models of complex systems

Claims

Examples

-

Key Sources

  • Murtaza, M., Cheng, C.-T., Fard, M., & Zeleznikow, J. (2024). Transforming Driver Education: A Comparative Analysis of LLM-Augmented Training and Conventional Instruction for Autonomous Vehicle Technologies. International Journal of Artificial Intelligence in Education. https://doi.org/10.1007/s40593-024-00407-z
Page metadata
Field Value
type design
id chatgpt-prompt-framework-general-pedagogical-principles
title A dual-focus framework for preparing ChatGPT prompts for training, distinguishing general LLM prompting principles from application-specific pedagogical principles
description The article introduces a framework for structuring ChatGPT prompts used as training tools, built on "a dual focus, incorporating general LLM prompting principles, which apply universally across multiple contexts, and...
status review
generated by process:settle-candidates on 2026-10-27
trust unverified — no one has confirmed this page's content
sources 1 — murtaza-2024

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