Pre-Prompting Patterns
- Pre-prompting patterns are teacher-initiated prompts used before student activity in programming workshops, often with AI, to structure discussion, clarify tasks, and allocate roles.
- Specific patterns like Code-Based, Reverse, and Role-Driven pre-prompting are documented to enhance student collaboration, critical problem interpretation, and balanced division of labor.
- Carefully implemented pre-prompting integrates AI as a scaffold for inclusive participation and critical thinking, rather than just a source of answers, by making pedagogical goals explicit.
Pre-prompting patterns refer to systematic, teacher-initiated prompts administered prior to student activity in programming pair workshops, particularly in educational settings involving generative AI tools. In the paper of university-level systems development courses, these pre-prompts are crafted to influence early collaborative processes, structure discussions, clarify task understanding, and allocate roles, thereby shaping the trajectory and quality of student engagement. Distinct pre-prompting patterns have been documented and analyzed for their impact on fostering productive peer collaboration and knowledge construction.
1. General Conceptual Pre-Prompting
General conceptual pre-prompting provides students with a broad, foundational prompt focusing on a core technical or theoretical concept. Typically, these are formulated to be accessible to all students, minimizing the cognitive and social risk of participation. Prompts of this sort often request non-programming examples or plain-language explanations, intentionally de-coupling from code to encourage inclusive group discussion.
Example:
Can you explain the difference between something synchronous and something asynchronous to me, please. Keep your answer short and give a simple example that is not related to programming.
Impact: This pattern lowers participation barriers and ensures all students discuss a common, neutral anchor rather than relying on self-generated or expert-supplied examples. It supports the early phase of problem interpretation by prompting collective evaluation of the AI’s response and encourages all group members to contribute. On occasion, some students perceive this approach as lessening the need to develop their own reasoning, viewing the exercise as passive; however, the pattern consistently achieves fast, lively engagement and minimizes uneven participation.
2. Code-Based Conceptual Pre-Prompting
This pattern leverages students’ engagement with concrete code by prompting AI to explain a relevant concept in direct relation to code that the group has examined or generated. Students are instructed to supply sample code and request an explanation targeting a particular abstraction illustrated by that code.
Example:
Please explain the concept of synchronous programming using the following code: [your pasted code]
Impact: By anchoring explanations in code, this pattern fosters connections between abstract concepts and tangible syntax or execution. It reliably increases students’ willingness to pose clarifying or critical questions, both to the instructor and AI, and supports mapping theory to practice. Discourse is more equal, as all participants share responsibility for interpreting the code and evaluating the AI response.
3. Reverse Pre-Prompting (Post-Discussion Verification)
Reverse pre-prompting instructs students to first collaboratively analyze and answer a question—often practical or conceptual—followed by submitting the same query to the AI for comparison. The objective is to encourage students to critique or validate their group-derived understanding against that provided by AI.
Example:
After coding a LINQ expression and discussing its behavior, students are directed: Ask ChatGPT the same questions as above, then compare its answers with your own reasoning.
Impact: This method strengthens source-critical thinking and metacognition. Students systematically reflect upon potential discrepancies between their interpretations and the AI’s output, which often leads to a heightened awareness of common misconceptions and alternative perspectives. The group thus practices verifying and, if necessary, correcting their own reasoning.
4. Exploratory/Iterative Pre-Prompting
Exploratory or iterative pre-prompting embeds instructions for students to progressively refine their inquiry through cycles of follow-up prompts. The pre-prompt explicitly encourages further questioning to address ambiguity or deepen understanding.
Instructional Strategy:
Add follow-up instructions to the prompt if you feel it is necessary.
Impact: This approach models discourse patterns common in scientific and technical exploration, with students collectively negotiating the direction of subsequent queries. It equalizes participation by allowing multiple group members to take initiative in interacting with the AI, not just those most confident in their knowledge. This pattern supports deeper inquiry, especially during debugging or conceptual unpacking tasks.
5. Role-Driven/Task-Specific Pre-Prompting
Role-driven or task-specific pre-prompting distributes explicit prompting and discussion duties according to group roles (e.g., “driver” and “navigator” in pair programming). Typically, one student executes code while the other is charged with crafting and submitting the AI prompt, interpreting the output, and facilitating reflection.
Case Structure:
| Stage | Student Action | AI/Prompted Action |
|---|---|---|
| Preparation | Review problem or code | -- |
| Prompting | Formulate/copy pre-prompt | Generate AI response |
| Collaboration | Discuss, critique, iterate | Field follow-up as needed |
| Reflection | Compare and validate answers | -- |
Impact: This structured rotation requires all students to engage with both technical (code) and communicative (prompt) aspects of the task. It curtails dominance by a single member and reinforces an equitable division of labor. The alternation of responsibilities provides varied perspectives on both the programming and the inquiry process.
6. Effects on Collaboration, Interpretation, and Division of Labor
Systematic use of pre-prompting patterns has demonstrable effects on the dynamics of pair and group programming sessions:
- Student Collaboration: Patterns offering ready-made prompts lower the social and linguistic barriers for less confident students, supporting more inclusive discussion. Exploratory and role-driven designs further ensure wide participation by explicitly rotating initiative and responsibility.
- Problem Interpretation: By foregrounding either general concepts, code, or post-discussion critique, these patterns ensure students jointly engage in critical evaluation and substantive sense-making. The reflective (reverse) pattern, in particular, heightens self-awareness and exposes students to alternative reasoning processes.
- Division of Labor: Pre-prompting, especially when role-based, formalizes balanced labor by assigning and rotating distinct cognitive and technical duties. This minimizes social loafing and increases the likelihood that all students develop both programming skills and inquiry competence.
7. Pedagogical Considerations and Reflections
The documented patterns indicate several broader educational consequences. When carefully introduced and purposefully explained, pre-prompting integrates AI tools as scaffolds for inclusive participation, critical discussion, and collaborative meaning-making. However, the effectiveness of these interventions is contingent upon the instructor’s clarity in communicating not just what prompt to use, but why—making explicit the intended cognitive and collaborative outcomes.
Structured pre-prompting can transform previously passive or uneven groups into more balanced, critically engaged teams. A plausible implication is that over-reliance on AI-generated content, without explicit pedagogical framing, may risk diminishing students’ perceived need for peer reasoning and instructor guidance. Strategic implementation ensures that AI’s role is complementary, amplifying rather than replacing core educational interactions.
Table: The Five Documented Pre-Prompting Patterns and Core Effects
| Pattern | Primary Focus | Collaboration Outcome |
|---|---|---|
| General Conceptual Pre-Prompting | Core concept with non-programming ex. | Lowers barriers, shared group anchor |
| Code-Based Conceptual Pre-Prompting | Code-driven, concept explanation | Anchors abstraction in code, high equity |
| Reverse (Post-Discussion Verification) | Compare self vs. AI solution | Promotes reflection, critical thinking |
| Exploratory/Iterative Pre-Prompting | Continuous inquiry, follow-ups | Rotates initiative, deepens engagement |
| Role-Driven/Task-Specific Pre-Prompting | Explicit role rotation, prompting assignment | Forces equitable participation |
In sum, teacher-engineered pre-prompting patterns provide robust scaffolds in programming pair workshops, enabling structured discussion, clarified problem interpretation, and balanced division of labor, with clear potential for enhancing both cognitive and collaborative learning outcomes.