Self Explanation Prompts¶

Strategy · All strategies
Evidence · 1 claim (1 against) · 2 studies (1 causal, 1 quant-synthesis), q3–q4 · 1 of 2 report an effect size

Description¶

Self explanation prompts ask learners to generate explanations to themselves — why a step in a worked solution was taken, why a statement is true, or how new material relates to what they already know. The prompt supplies the question; the learner supplies the reasoning. This differs from receiving an explanation (which is instructor-generated) and from Think-Aloud (which verbalizes ongoing processing rather than justifying specific content).

Design Implications¶

Self explanation forces the learner to actively integrate new information with prior knowledge, exposing gaps and repairing flawed mental models during study rather than at assessment. A meta-analysis of 64 studies found self explanation prompts produce moderate learning gains across domains and ages Bisra et al. (2010) [+M]. The quality of the explanation matters more than its mere production: prompts that direct attention to why and how yield deeper learning than prompts that merely restate content.

Context¶

Requirements¶

• Content with explainable structure — worked solutions, proofs, diagrams, causal claims, or procedures with justifiable steps
• Prompts tied to specific content ("Why is this step valid here?") rather than generic ("Explain what you learned")
• Enough working memory headroom that explaining does not crowd out comprehension of the material itself
• A way for learners to check their explanations against expert reasoning (Worked Examples with step annotations, or instructor feedback)
• Study material with enough structure that correct explanations are discoverable (e.g., worked examples, well-organized texts)
• Prompt stems that direct attention to why and how, not just what
• Some mechanism for learners to detect and repair flawed explanations — feedback, Comparing Cases, or access to correct reasoning — since self-generated explanations can be wrong and errors go uncorrected [-M]
• Time: self explanation slows study; schedules must accommodate it

Constraints¶

• Explaining familiar or trivial content wastes time and adds load without benefit [~M] — prompts must target genuinely non-obvious steps
• Learners with low prior knowledge may generate inaccurate explanations and encode the errors; explanations should be verifiable against authoritative content [-M]
• Prompting explanation of already well-understood material can disrupt fluent processing (the "over-explanation" cost seen with high-knowledge learners) [~M]
• Poorly timed prompts during reading can interrupt comprehension flow; place them after meaningful segments, not mid-sentence [~W]
• Inaccurate self explanations can entrench misconceptions when no corrective mechanism exists [-M]
• Learners with high prior knowledge may gain little and experience the prompts as redundant, consistent with the expertise reversal effect [~M]
• Overly frequent or rigid prompting adds extraneous load and interrupts comprehension of the base material Cognitive overload degrades learning. [-M]
• For rote or procedural fluency goals, explanation adds little beyond direct practice [~M]

Implementation Variability¶

• Prompted during study — embedded questions after each worked-example step (the classic Chi paradigm)
• Learner-initiated — training learners to spontaneously ask "why" while studying; slower to establish but more durable
• Structured formats — sentence starters, selection from menu of reasons, or Concept Mapping as explanation scaffolds for younger or lower-prior-knowledge learners
• Interactive — pairs take turns explaining and questioning each other, blending self explanation with Collaborative Learning
• Prompted during study (inline prompts after each step or paragraph) vs. prompted after study (end-of-section explanation tasks); inline prompting produces stronger effects but costs more time [+M]
• Open prompts ("explain what this means to you") vs. directed prompts ("why is this step valid?"); directed prompts are more reliable for novices [~M]
• Training in self explanation followed by unprompted use — the long-term goal is a self-initiated study habit, part of self-regulated learning

Target Learners¶

• Novices and intermediate learners; effects shrink or reverse for high-prior-knowledge learners for whom explanation is redundant [~M]
• Learners prone to illusions of understanding from fluent reading — explaining exposes what they cannot actually justify [+M]
• Young learners need more scaffolded prompt formats (menu-based or sentence-starter prompts) [~W]
• Novices and low-prior-knowledge learners, who otherwise read worked examples superficially Bisra et al. (2018) [+S]
• Middle school through adult learners; effects are weaker but present for younger children, who may need more scaffolded stems [+M]
• Less valuable for experts, for whom prompted explanation of familiar material is redundant [~M]

Target Learning Goals¶

• Conceptual understanding: why procedures work, not just how to execute them
• Schema construction: connecting new material to prior knowledge
• Metacognitive monitoring: discovering gaps and misconceptions during study
• Transfer: principle-based explanations support applying methods to novel problems [+M]
• Integration: connecting new material to prior knowledge and across topics
• Metacognitive monitoring: explaining exposes gaps the learner didn't know they had [+M]
• Transfer: principled explanations support applying methods to novel problems [+M]

Instructions¶

1. Select content with explainable reasoning steps — a worked solution, demonstration, or Case Study.
2. Write a prompt for each non-obvious step, focused on justification ("Why does this step follow?") rather than description.
3. Present the content, then the prompt, giving adequate time before revealing any expert explanation.
4. Let learners compare their explanation to an expert explanation or annotated solution, and revise.
5. Fade the prompts as expertise develops, shifting toward learner-initiated explanation (Fading).

Related Strategies¶

• Worked Examples — the most common carrier; step-by-step self explanation prompts turn passive example study into active processing
• Think-Aloud Modeling — instructor version of the same move; models the explaining learners are later prompted to do
• Elaborative Interrogation — the "why is this true?" variant applied to factual text rather than procedures
• Use elaborative interrogation (answering why a fact is true) as a low-resource strategy for science content that conflicts with student misconceptions
• Worked example study — self explanation prompts are the standard companion to worked examples, converting passive example reading into active principle extraction
• Retrieval practice — complementary generative activity; explanation targets understanding, retrieval targets accessibility
• Comparing contrasting cases — comparison prompts often elicit self explanations as a byproduct
• Induce self-explanation by prompting learners during study and problem solving

Examples¶

• Chi et al.'s classic paradigm — Studying physics worked examples with each step annotated, prompted "What is this line telling you? Explain why it was done."
• Khan Academy (https://www.khanacademy.org) — hint sequences progressively reveal solution steps, each functioning as a check on the learner's own explanation of the next move.
• Cognitive Tutor / Carnegie Learning (https://www.carnegielearning.com) — adaptive math tutors require justification selections ("Why did you choose this operation?") at each problem-solving step.
• Chi et al.'s (1989) physics studies — the foundational paradigm: students studying worked mechanics examples who spontaneously self explained learned far more; prompted versions replicate the effect for non-explainers.
• Khan Academy — hint sequences in practice exercises function as scaffolded explanation prompts, revealing one reasoning step at a time before the full solution.
• ASSISTments — math homework platform that inserts "explain your reasoning" prompts and uses explanation quality to route feedback.
• Reading comprehension instruction — reciprocal teaching (Palincsar & Brown) embeds self explanation within summarizing and questioning routines.

Key Sources¶

• Chi, M. T. H., Bassok, M., Lewis, M. W., Reimann, P., & Glaser, R. (1989). Self-explanations: How students study and use examples in learning to solve problems. Cognitive Science, 13(2), 145–182. doi:10.1207/s15516709cog1302_1
• Bisra, K., Liu, Q., Nesbit, J. C., Salimi, F., & Winne, P. H. (2018). Inducing self-explanation: A meta-analysis. Educational Psychology Review, 30(3), 703-725. doi:10.1007/s10648-018-9434-x
• Atkinson, R. K., Renkl, A., & Merrill, M. M. (2003). Transitioning from studying examples to solving problems: Effects of self-explanation prompts and fading worked-out steps. Journal of Educational Psychology, 95(4), 774–783. doi:10.1037/0022-0663.95.4.774
• Fonseca, B. A., & Chi, M. T. H. (2011). Instruction based on self-explanation. In R. E. Mayer & P. A. Alexander (Eds.), Handbook of Research on Learning and Instruction (pp. 296–321). Routledge.
• Bisra, K., Liu, Q., Nesbit, J. C., Salimi, F., & Winne, P. H. (2018). Inducing self-explanation: A meta-analysis. Educational Psychology Review, 30(3), 703–725. doi:10.1007/s10648-018-9434-x
• Atkinson, R. K., Renkl, A., & Merrill, M. M. (2003). Transitioning from studying examples to solving problems: Effects of self-explanation prompts and fading worked-out steps. Journal of Educational Psychology, 95(4), 774–789. doi:10.1037/0022-0663.95.4.774
• Roy, M., & Chi, M. T. H. (2005). The self-explanation principle in multimedia learning. In R. E. Mayer (Ed.), The Cambridge handbook of multimedia learning (pp. 271–286). Cambridge University Press.