Self-questioning improves comprehension¶

Claim · All claims
Evidence · 2 studies · 1 quant-synthesis r?, 1 review r? · q3 · i1 small
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

When learners generate and answer their own questions during reading or study, their comprehension of the material improves relative to reading alone. The mechanism is generally understood as generative processing: posing questions forces learners to identify main ideas, connect them to prior knowledge, and monitor their own understanding.

Subclaims¶

q3 i1 Across the intervention-study literature, teaching students the cognitive strategy of generating their own questions during reading produces gains in comprehension on posttests, whether taught through skill-based instruction or reciprocal teaching. → Rosenshine et al 1996

q3 i? A review of 35 experimental studies teaching self-questioning to K–12 students with and without disabilities finds the strategy effective for improving reading-comprehension performance across a range of diverse learners and educational settings. → Joseph et al. 2016

Evidence¶

Rosenshine et al 1996¶

Rosenshine, B., Meister, C., & Chapman, S. (1996). Teaching Students to Generate Questions: A Review of the Intervention Studies. Review of Educational Research, 66(2), 181–221. doi:10.3102/00346543066002181

q3 · review/synthesis of intervention studies (abstract only) · i1 · small–medium effect, median d=0.36 (standardized tests) · n=studies not counted in abstract · quant-synthesis · r?

Reviewed intervention studies in which students were taught the cognitive strategy of generating their own questions about text as a comprehension aid (a self-questioning/text-embedded-questioning strategy, distinct from teacher-posed questions). Comprehension was measured on new material at the end of training. Across studies, question-generation training improved comprehension, with a median effect size of 0.36 (64th percentile) on standardized comprehension tests and 0.86 (81st percentile) on experimenter-developed tests; a traditional skill-based approach and a reciprocal-teaching approach produced similar results. Only the abstract was read (paywalled full text; not found via Unpaywall or Semantic Scholar), so the number of contributing studies is not confirmed and the i code reflects only the standardized-test figure the abstract states as a reported effect size.

Joseph et al. 2016¶

Joseph, L. M., Alber-Morgan, S., Cullen, J., & Rouse, C. (2016). The Effects of Self-Questioning on Reading Comprehension: A Literature Review. Reading & Writing Quarterly, 32(2), 152–173. doi:10.1080/10573569.2014.891449

q3 · systematic literature review of experimental studies · i? · no effect size reported in what was read · n=35 studies · review · r?

This review examined 35 experimental research studies that taught self-questioning to K–12 students, with and without disabilities, to determine whether self-questioning is an evidence-based practice for reading comprehension. The authors report that a variety of specific self-questioning strategies have been used across these studies, and that these strategies were effective for improving reading-comprehension performance across a range of diverse learners and across various educational settings. The publisher's full text was not accessible from this environment (Taylor & Francis returned an access-denied response); the finding above is drawn from the ERIC abstract, which describes the strategies as effective but reports no pooled or per-study effect size, so impact is coded i? rather than inferred.

Studies bearing on this claim¶

Do annotations that make the reader generate questions, explanations or summaries improve comprehension? This table gathers the studies behind this claim and the 4 narrower claims that bear on it: 12 distinct studies (5 quant-synthesis, 4 causal, 2 review, 1 associational), q2–q4, 3 of 12 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¶

Why it works. Self-questioning is a generative strategy: constructing a question requires the learner to segment text into idea units, judge which content is central, and formulate the relationship explicitly. Answering one's own question then provides immediate feedback on whether the idea was understood, supporting the comprehension-monitoring component of self-regulated learning. This places self-questioning alongside other generative strategies such as 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 and Activation improves learning, which similarly require learners to produce content rather than passively receive it.

Moderators and boundary conditions. The quality of the questions matters. Shallow, fact-recall questions ("what year did...?") are likely to yield smaller comprehension gains than questions targeting causal relations, main ideas, or explanations ("why did...?", "how does X relate to Y?"). Novice learners may need explicit training and question prompts or stems before they can generate productive questions independently; without scaffolding, some learners generate questions about trivial details or skip question generation entirely. Self-questioning also imposes working-memory demands, so for very difficult or high-load material it may compete with comprehension itself — consistent with the conditions described in Cognitive overload degrades learning and Chunking reduces working memory load.

Instructional design implications. Effective implementations typically (a) model question generation before asking learners to do it independently — the core move in reciprocal teaching, where the "questioner" role is one of several strategies modeled by the teacher; (b) provide question stems or prompts early and fade them as learners internalize the strategy; and (c) build question generation into the reading cycle itself (before, during, and after reading) rather than treating it as a post-hoc quiz. Self-questioning also functions as a comprehension-monitoring device: a learner who cannot answer their own question has located a gap, which can trigger rereading or help-seeking — a link to the broader self-regulated learning literature.

Open questions. The two reviews recorded above support the claim; only one reports an effect size (median d = 0.36 on standardized tests, 0.86 on experimenter-made tests), and both were read at abstract level. Until the full texts or a pooled estimate across studies are recorded, the strength of this claim should be treated as provisional despite its long history in the reading-comprehension literature.

Related Claims¶

• 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 — another generative strategy that forces active processing of text.
• Activation improves learning — self-questioning often serves as an activation mechanism, connecting new text to prior knowledge.
• Cognitive overload degrades learning — a boundary condition: questioning during very high-load material can compete with comprehension.
• Chunking reduces working memory load — segmenting text into question-sized units supports the strategy.
• Reciprocal Teaching Improves Reading Comprehension — related
• Scaffolded Close Reading Improves Comprehension For Struggling Readers — related
• Metacognitive strategy instruction, which includes comprehension monitoring, improves academic performance by a medium amount, and in one non-randomised study comprehension-monitoring training raised low achievers' reading scores — a broader claim this one bears on
• Questioning Strategies Improve Learning — related
• Reading Strategy Instruction Improves Comprehension — a broader claim this one bears on
• Students typically receive little or no instruction in how to analyze, answer, and write answers to questions — related
• Teachers reported redesigning instruction toward learner question formulation, comparison, and inquiry when using AI — related
• Adjunct aids such as learning objectives and embedded questions show limited effects and may interfere with competent readers' natural strategies — related
• Comprehension instruction in resource rooms consisted mostly of teacher question-and-answer monitoring rather than explicit comprehension-strategy instruction — related
• Explicit training in constraint-seeking questions showed only moderate, non-generalizing benefits, whereas prompting explanation supports better question asking — related
• Trained students use more diverse higher-order prompting strategies, including generating new questions absent among controls — related
• Whether strategy instruction works better in authentic content-area tasks is untested, and the one meta-analytic comparison recorded found larger gains in strategic ability in language-arts than in content-area classes — related
• Self-regulated learning improves achievement — a broader claim this one bears on
• Metacognitive Strategies Improve Learning — a broader claim this one bears on
• The review reports reciprocal teaching of reading, an approach exemplifying cognitive apprenticeship, was remarkably effective in raising reading comprehension scores, especially for poor readers — related
• Reading comprehension strategy programs in grades 3–6 yield a medium average effect size of 0.34 — related
• Pedagogical interventions and scaffolds increase learners' use of self-regulated reading strategies — related