JH
Jonathan Haber

Articulation prompts

Ask learners to explain their reasoning before, during, and after a task to surface and strengthen tacit knowledge.

Key takeaways

  • What it is: Ask learners to explain their reasoning before, during, and after a task to surface and strengthen tacit knowledge.
  • Why it works: Generating an explanation requires the learner to identify what they know, detect gaps, and organize knowledge into a communicable structure — all of which deepen encoding. Articulation also produces material the coach can respond to: a correct explanation means the concept is internalized; an incorrect one reveals the exact misconception to address. Collins et al. identify articulation as the complement to modeling — the learner does for themselves what the expert did in the think-aloud.
  • Evidence: Backed by randomized trials / meta-analyses.
  • Avoid: Accepting surface articulation ("I did it because it seemed right") rather than probing for the actual reasoning, which lets genuine misconceptions hide behind confident-sounding vagueness.

Why it works

Generating an explanation requires the learner to identify what they know, detect gaps, and organize knowledge into a communicable structure — all of which deepen encoding. Articulation also produces material the coach can respond to: a correct explanation means the concept is internalized; an incorrect one reveals the exact misconception to address. Collins et al. identify articulation as the complement to modeling — the learner does for themselves what the expert did in the think-aloud.

How to do it

  1. 1Before a task: ask "What is your plan and why?"
  2. 2During the task: ask "Why did you just do that?" at a decision point.
  3. 3After the task: ask "What would you do differently and why?"
  4. 4Listen for gaps between stated reasoning and actual behavior — those gaps are the coaching targets.

What the evidence says

RCT / meta-analysis

Generating explanations (self-explanation effect) consistently outperforms re-reading and passive study in learning research. The self-explanation effect has been documented in worked-example learning across mathematics, physics, and biology.

Honest caveat: Self-explanation research typically uses text and worked examples; articulation during coached performance tasks relies on the same mechanism but is less directly replicated in that exact form.

References
  • — Chi et al. (1989), "Self-explanations: How students study and use examples in learning," Cognitive Science

Common mistake

Accepting surface articulation ("I did it because it seemed right") rather than probing for the actual reasoning, which lets genuine misconceptions hide behind confident-sounding vagueness.

IX Coach prompts you to explain your reasoning at key decision points during guided practice, then reflects the gaps between your stated plan and your actual moves back to you.

Practice this with IX Coach →

Related practices