2.3. Core Strategies for Prompting
Asha Vas
Prompting is both an art and a science. While simple strategies like role-based or few-shot prompting can help, professionals benefit from a structured framework that ensures clarity, completeness, and relevance. One such framework is the Six-Element Prompting Model, adapted from Liu et al. (2023) and Zamfirescu-Pereira et al. (2023). This model breaks a prompt into six elements:
- Role/Persona – Who is the AI supposed to act as?
- Task/Instruction – What exactly should it do?
- Context/Constraints – What is the clinical or educational scenario?
- Input Data – What case details or variables are needed?
- Output Format – How should the response be structured?
- Tone/Style – What literacy level, audience, or communication style is appropriate?
The Six Element Prompting Model
1. Role / Persona
Weak: “Explain photosynthesis to students.”
Strong: “You are a high school biology teacher. Explain photosynthesis to a group of 9th graders using simple examples.”
Why it matters: Anchors vocabulary and level of explanation so the AI knows who it is “speaking” as.
2. Task / Instruction
Weak: “Write an introduction for a research paper.”
Strong: “Draft a 200-word introduction for a research paper that identifies the problem, explains why it matters, and previews the argument.”
Why it matters: Clarifies the expected deliverable and reduces ambiguity.
3. Context / Constraints
Weak: “Write a speech for a student event.”
Strong: “Write a three-minute opening speech for a student debate competition, addressed to first-year undergraduates.”
Why it matters: Provides situational grounding that helps the AI generate a more relevant response.
4. Input Data
Weak: “Summarize this article into a paragraph.”
Strong: “Summarize this 800-word news article into three key bullet points suitable for a classroom newsletter.”
Why it matters: Ensures the AI response is tied to the actual material or details provided.
5. Output Format
Weak: “Explain climate change and give examples.”
Strong: “Provide a table with two columns: causes of climate change and their effects, with short examples in each row.”
Why it matters: Structures the output for clarity and usability.
6. Tone / Style
Weak: “Describe quantum computing in simple terms.”
Strong: “Describe quantum computing in 200 words for high school students, using clear, age-appropriate metaphors and avoiding jargon.”
Why it matters: Adapts communication to the intended audience.
Analogy: Writing a Recipe (click to expand)
📖 Analogy: Writing a Recipe (click to expand)
Think of prompting as writing a recipe. If you only say “make pasta,” the result could vary wildly—spaghetti, lasagna, or mac and cheese. But if you provide details (ingredients, cooking method, portion size, presentation style), you guide the cook toward a predictable dish. Similarly, the Six-Element Prompting Model gives AI the “recipe” it needs to produce responses that are aligned with your purpose.
📚 Weekly Reflection Journal
Reflection Prompt: Write a short prompt to an AI tool (e.g., “Explain climate change”). Then, revise your prompt by adding one or more of the six elements discussed above. Compare the original and revised outputs. What improved? What surprised you?
Looking Ahead
In the next chapter, we will explore advanced prompting techniques such as chain-of-thought reasoning, iterative refinement, and multi-turn dialogue. These strategies build on the Six-Element Model and illustrate how careful prompting can unlock more accurate and nuanced AI responses.
