Question prompts improve learning¶

Claim · All claims
Evidence · 2 studies · 1 quant-synthesis r?, 1 review r? · q3 · i2 medium
Currency · newest study 2016 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person

Question prompts are questions embedded in learning materials, tasks, or dialogue that ask learners to respond before, during, or after instruction. They are hypothesized to improve learning by prompting active processing — retrieval, self-explanation, and connection to prior knowledge — rather than passive reading or listening.

Subclaims¶

q3 i? Adjunct (inserted) questions accompanying prose text produce a strong facilitative effect on retention of the specific material they ask about, and higher-order adjunct questions generalize this benefit to related and unrelated test material more than factual recall questions do; factual pre-questions can depress learning of unrelated, unquestioned content. → Hamaker 1986

q3 i2 Elaborative interrogation — prompting learners to explain why a stated fact is true — shows a medium average effect on deep-learning outcomes (d = 0.42) across a meta-analysis of 24 studies compiled in a synthesis of learning-strategy research. → Hattie & Donoghue 2016

Evidence¶

Hamaker 1986¶

Hamaker, C. (1986). The Effects of Adjunct Questions on Prose Learning. Review of Educational Research, 56(2), 212–242. doi:10.3102/00346543056002212

q3 · systematic review of the adjunct-questions literature (examines 13 design moderators across studies) · i? · no pooled effect size reported in the abstract · n=NR (narrative/quantitative review; study count not given in the abstract) · quant-synthesis · r?

This is a review of the research literature on factual and higher-order adjunct (inserted) questions — questions embedded before or after sections of text. It finds that adjunct questions of all cognitive levels strongly help learners on repeated (re-asked) test questions and help more weakly on related test questions. Higher-order adjunct questions (versus factual/verbatim ones) generalize further, improving performance on repeated, related, and even unrelated higher-order test questions. Factual pre-questions and (under time pressure) factual post-questions can actually hurt performance on unrelated, unquestioned material — the classic pre-question "narrowing of attention" cost. Effect sizes varied with text length, question density, question/test format, and control-group performance level, but not with learner age or the interval between reading and test. Only the abstract was available (publisher paywall; no open-access copy found), so no overall quantitative effect size can be reported here.

Hattie & Donoghue 2016¶

Hattie, J. A. C., & Donoghue, G. M. (2016). Learning strategies: a synthesis and conceptual model. npj Science of Learning, 1(1), 16013. doi:10.1038/npjscilearn.2016.13

q3 · synthesis compiling meta-analytic results (1 meta-analysis, 24 studies, 164 effects) from a learning-strategies database · i2 · medium effect, d=0.42 · n=24 studies (2,150 participants; prorated to 15,450) · review · r?

This paper synthesizes meta-analytic evidence on learning strategies into a model of surface, deep, and transfer learning phases. Elaborative interrogation — a self-regulation strategy in which learners generate explanations for why a stated fact or idea is true — is classified under "acquiring deep learning." Its own summary table reports this strategy at an average effect size of d = 0.42 (medium), drawn from one meta-analysis covering 24 studies, 2,150 participants (prorated to 15,450), and 164 effects — notably smaller than several other deep-learning strategies in the same table (e.g., elaboration and organisation, d = 0.75; meta-cognitive strategies, d = 0.61). The paper itself does not name the underlying meta-analysis beyond attributing its strategy database to Hattie's earlier Visible Learning synthesis.

Studies bearing on this claim¶

Does the gain reach content the pre-questions did not ask about? This table gathers the studies behind this claim and the 1 narrower claims that bear on it: 3 distinct studies (2 review, 1 quant-synthesis), q2–q3, 1 of 3 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Proposed mechanisms. Question prompts are typically justified on three grounds: they induce retrieval practice (answering requires reconstructing knowledge from memory), they trigger activation of relevant prior knowledge, and they push learners toward elaborative processing similar to self-explanation. All three mechanisms are well established in the broader literature, but the strength of a prompt's effect depends heavily on whether learners actually generate a substantive answer rather than skipping or skimming the question.

Moderators and boundary conditions. Several conditions plausibly moderate the effect. Prompt type matters: questions requiring inference, application, or connection to prior knowledge are generally expected to do more work than verbatim recall questions, though recall questions may still aid retention. Timing matters: pre-questions can orient attention but may also act as advance organizers that narrow attention in ways that harm incidental learning of unprompted material — see Advance organizers improve learning; post-questions avoid this but arrive after the relevant text has been processed. Learner expertise matters: for novices, answering questions adds load to an already strained working memory, and poorly timed prompts could contribute to cognitive overload; for advanced learners, simple prompts may be redundant — a pattern consistent with the expertise reversal effect. Finally, prompts that go unanswered — because learners lack the knowledge, motivation, or feedback to answer them — likely produce little benefit and may inflate perceived difficulty.

Open questions. Two syntheses are recorded, neither a primary study: the evidence above covers adjunct questions (Hamaker 1986, abstract only, no pooled effect) and elaborative interrogation (d = 0.42, Hattie & Donoghue 2016, second-hand from one unnamed meta-analysis). Open questions: (1) do pre-questions improve learning of the questioned material at the cost of incidental material? (2) do elaborative/why-questions outperform factual recall questions? (3) does requiring a written answer matter, or does merely encountering the question suffice?

Related Claims¶

• Activation improves learning — prompts that surface prior knowledge operate partly through activation.
• Advance organizers improve learning — pre-questions function as a question-based form of advance organizer.
• Learner highlighting of text gives a small memory benefit but no reliable comprehension benefit, and evidence on other forms of annotation is not yet recorded here — answering prompts is a structured form of the generative processing annotation induces.
• Active learning improves exam performance — question prompts are one of the cheapest ways to make instruction active.
• Cognitive overload degrades learning — poorly timed or overly demanding prompts can add load for novices.
• Expertise reversal effect — prompt benefits may shrink or reverse as learner expertise grows.
• Pretend play enactment of stories facilitates narrative recall and expression over shorter time periods but not later unprompted recall — related
• Questioning Strategies Improve Learning — related
• Practicing retrieval of some portions of an educational text can enhance retention of related nontested portions (retrieval-induced facilitation), but not when relational encoding is disrupted — related
• Pretesting enhances learning — related
• Students typically receive little or no instruction in how to analyze, answer, and write answers to questions — related
• Subjects with a more developed schema for a body of knowledge show higher recall for materials related to that knowledge — related
• Promoting higher-order thinking through questions requiring analysis, explanation, and idea development is among the practices most consistently related to student growth — related
• Error correction supports learning when it requires substantive cognitive work such as explanation and reconstruction — related
• Existing theoretical accounts attributing the testing effect to enhanced semantic elaboration receive limited support — related
• Elaborative interrogation improved true/false test performance in an introductory biology course context — related
• Underlining and adjunct questions may hinder rather than help recall in young readers, reflecting a comprehension rather than a metacomprehension deficit — reports the opposite
• Adjunct aids such as learning objectives and embedded questions show limited effects and may interfere with competent readers' natural strategies — reports the opposite
• Turn-taking test review makes each member's expertise visible, enabling peers to locate and draw on community knowledge — related
• Reviews of classroom studies, read from their abstracts, associate longer teacher wait time with longer, higher-cognitive-level student responses, with no effect size recorded — related
• Answering inserted adjunct questions with the text available produces higher delayed transfer learning than answering post-reading questions — related
• Prior-knowledge activation instructions before study had little effect on retention in Experiment 2 — related
• Higher-level quiz questions engage learners in generative activities as well as retrieval — related