Provide explicit prompting instruction and expert scaffolding before or during novice LLM activities¶

Strategy · All strategies
Evidence · no claims cited

Description¶

Based on the observed limits of trial-and-error prompting, the authors recommend that educators guide trial-and-error efforts through explicit instruction: encouraging more effective prompts using expert-informed techniques, deliberate control over LLM outputs, metacognitive planning of which phases suit LLM support, and solid subject-specific knowledge for evaluating outputs. They state directly that "Future studies should replicate this intervention, first providing students with basic information about prompting and its principles, in order to examine whether explicit guidance enhances their ability to engage effectively with LLMs." The recommendation responds to students' limited use of advanced strategies and their overgeneralized conclusions about model incapability.

Design Implications¶

Context¶

Requirements¶

• Instructor familiarity with prompting strategies to provide expert-informed guidance

Constraints¶

• The article notes the effectiveness of explicit guidance is a hypothesis for future replication, not an established finding of this study

Target Learners¶

• genAI novices in higher education

Target Learning Goals¶

• prompt design competence
• metacognitive planning of LLM-supported work
• critical output evaluation

Related Strategies¶

• Explicit Teacher Modeling
• Modeling And Think Alouds
• Search Strategy Scaffolds
• Scaffold question formulation and active help-seeking in AI-supported tutoring systems
• Teachers should make task demands explicit, compelling, varied, and metacognitively supported when assigning text-study tasks

Examples¶

-

Key Sources¶

• Nicola Brocca, Davide Garassino. (2026). LLMs in text linguistics teaching: an exploratory study with genAI novices in higher education. Computers and Education Open 11 (2026) 100414. https://doi.org/10.1016/j.caeo.2026.100414