Skip to content

Elaborative Reasoning

Principle · All principles
Evidence · 42 claims (21 for, 17 mixed, 4 against) · 35 studies (20 causal, 7 quant-synthesis, 4 review, 3 associational, 1 qualitative), q2–q4 · 13 of 35 report an effect size · 36 claims rest on one study

Conditional relationship

This page owns elaborating content the learner is meant to remember and understand: asking the learner to say why a stated fact or idea is true, how it connects to what they already know, or what it implies, instead of reading or rereading it. For a learner meeting statements in a text or lesson (facts, causal statements, descriptions of how something works; mostly verbal-association and concept targets) who holds relevant and accurate prior knowledge of the topic, answering a "why is this true?" prompt for each statement, or generating one's own connection or example, is expected to improve recall and recognition of that content (immediate-performance, delayed-retention) more than reading it for understanding. The expected cost is time: elaborating takes markedly longer than reading, and feels harder.

The relationship is conditional on four things, and the evidence here bears on each:

  • Prior knowledge of this content. The gain is larger in a domain the learner knows well than in one they know little about, and passages learners know more about are recalled better a week or a month later. The observation that matters is the learner's starting response on this material (what they can say about it now), not a label.
  • The kind of target. The positive experiments are about facts and prose (science facts, a biology text, passages). In one experiment where undergraduates who could not yet do the task learned to calculate a posterior probability from a written lesson, answering why-questions produced lower accuracy than simply reading the lesson twice, although it took far longer. For a procedure or principle the learner does not yet understand, an explanation or worked example comes first.
  • What the learner can generate. For facts, even inadequate or missing answers to why-questions helped relative to reading. For the probability lesson, most answers were missing, wrong or inadequate, and better answers went with better posttest scores. Whether answer quality matters therefore seems to depend on the target and the learner's knowledge; it is not settled here.
  • Who generates the elaboration. In one small classroom experiment in introductory accounting, students who generated their own elaborations scored higher than students given the instructor's elaborations, but not higher than students who did both.

Which claims test the relationship. Two randomised experiments with grade 6 and 7 students (one study, two experiments) test elaborative interrogation of science facts against reading for understanding, with recognition gains reported up to six months; a classroom study in an undergraduate biology course, reported in a review, tests interspersed why-prompts; one experiment tests it against reading twice for a probability lesson and finds the reverse. So the relationship is tested for facts and prose with learners who know something about them, and qualified for novices learning a calculation principle. No claim in this wiki tests elaboration across a whole course or term, or compares the prompt types (why, how does this relate, give an example) against each other.

How this page sits beside its converted siblings: Activation owns surfacing prior knowledge before instruction; this page owns using it while the content is studied. Analogical Reasoning owns mapping the structure of one case onto another. Cognitive Activation owns the general demand of tasks that require explaining, comparing or predicting for concept and principle targets; this page is the fact- and text-level form of that demand, plus the classroom move of unfolding a student's answer into its parts. Knowledge Organization owns the structure a learner ends up holding. Self-Explanation (not yet in this format) covers explaining the steps of worked examples and problems. The Elaboration Theory pattern is about sequencing content from simple to elaborated, a different sense of "elaboration".

Default design

The relationship is tested for facts and prose with learners who know something about them; the session design, doses and the arc across sessions below are proposals built from the earlier page and the folded pages, each step labelled with its evidence status.

  1. Check what learners already know about this material. Before choosing elaboration, ask learners to write or say two or three things they know that relate to the topic (one minute per topic), or count the related sentences they produce to a short prompt, the measure one report used for familiarity. Where most learners produce little or nothing, teach or activate the background first (see Activation) or pick material closer to what they know. Basis: Elaborative interrogation benefits increase as learners' prior knowledge increases [~M]; Passage familiarity (prior related knowledge) correlates .52 with recall, and only familiarity, not imageability, affected one-week delayed recall [+M]; Subjects with a more developed schema for a body of knowledge show higher recall for materials related to that knowledge [+W] (second-hand in an essay, mostly narratives). Where the choice is between familiar material and rewriting unfamiliar material to be more cohesive, prefer familiar material: in the same report, passage cohesion (repetition of concept labels across sentences) did not affect delayed recall while familiarity did (Passage cohesion showed no effect on delayed recall; means showed greater recall from less cohesive passages [~M]). The one-minute probe is an untested proposal.
  2. Attach a why-prompt to specific statements, not to whole passages. Mark the statements worth remembering and put "Why is this true?" or "Why does this make sense, given what you know?" beside each, roughly one prompt per paragraph or per 150 words of text. Basis: the classroom biology study interspersed 21 prompts, about one per 150 words, and the prompted group outperformed controls on a true/false test (Elaborative interrogation improved true/false test performance in an introductory biology course context [+M]); attaching prompts to discrete statements is otherwise an untested proposal from the broader claim's discussion.
  3. Have each learner answer, briefly, from what they know. One or two sentences per prompt, written or spoken, drawing on the lesson, earlier reading and everyday experience. Basis: Elaborative interrogation instructions produce better recall and recognition of science facts than reading-for-understanding instructions [+M] (grade 6 and 7, science facts); Elaborative Interrogation Improves Learning [+M].
  4. Model one good answer first and look at answer quality. Before learners start, show one answer that names a cause or a link to something known, beside one that only restates the fact. While they work, read a sample of answers; where many are missing, circular or wrong, give the explanatory link and add a worked answer. Basis: the earlier folded page's recommendation, from a study where most answers were poor and answer quality went with posttest scores (Answers to elaborative interrogation questions were mostly poor in quality, and answer quality correlated positively with posttest scores [~M]); modelling and monitoring as a remedy is an untested proposal. For fact learning, do not hold learners back for imperfect answers: The quality of elaborative-interrogation answers had little effect on retention; even failed or inadequate why-answers facilitated learning relative to reading [~M].
  5. Let learners generate first, then show the expert's elaboration. Ask for the learner's own connection or example before showing the instructor's, so that both are used. Basis: In one introductory accounting experiment, self-generated elaboration scored above instructor-assisted elaboration on procedural knowledge at the p = .05 boundary, and not significantly above a combined strategy [+M] (one class, three groups of about 26; the combined strategy did as well as self-generated). The same class's other outcomes point the same way, weakly: on total posttest score self-generated beat instructor-assisted (Cognitive elaboration strategy significantly affects overall posttest performance, with self-generated elaboration highest [+M]); on case analysis the difference fell short of the 5% level and the omnibus test was not significant (In one accounting classroom experiment, self-generated elaboration scored marginally above instructor-assisted elaboration on intellectual skills (p = .06, omnibus test not significant) and no higher than a combined strategy [~M]); and attitudes to the method did not differ (Elaboration strategy does not significantly affect students' attitudes toward the learning method [~M]). The order "generate, then compare" is an untested proposal.
  6. Aim prompts at statements learners are likely to believe wrongly. Where a fact contradicts a common belief, a why-prompt is the place to put it. Basis: Elaborative interrogation produced greater pretest-to-posttest gains on items about common misbeliefs than reading control in Experiment 2 [+M]; Elaborative-interrogation students are less confident in their incorrect recognition selections than reading controls, who express great confidence in erroneous answers [+M].
  7. For a calculation, rule or procedure learners cannot yet do, explain or show a worked example first. Put why-prompts on the steps of a worked example once the learner has seen one, rather than on a lesson the learner is meeting cold; if time is short, a second reading is a safer use of it than unsupported why-questions. Basis: Elaborative interrogation produced lower posttest accuracy than reading the lesson twice in a written lesson on posterior probability [-M]; for prompts on worked examples, Prompting learners to self-explain improves understanding and problem solving on immediate tests by a moderate average amount, with little evidence yet on delayed or classroom gains. [+S].
  8. In whole-class discussion, unfold a correct-sounding answer before accepting it. When a student offers a formula or shortcut ("3 × 8"), ask them (or another student) to show the structure it stands for ("3 floors of 8 cubes"), link that structure across objects, words and symbols, then reconnect it to the original answer. Do not jump to the procedure or funnel the class to the answer. Basis: Explicitly unfolding higher-order contributions into lower-order concept elements produces deeper conceptual learning opportunities than jumping directly to procedure [+M] (eight teachers, grades 5 to 7, volume; the outcome is the depth of the learning opportunity in the talk, not measured learning).
  9. Budget the time and say it will feel harder. Plan for elaboration to take roughly two to three times as long as reading the same material, and tell learners that it feels harder because it is doing more work. Basis: Elaborative interrogation increased perceived difficulty and time spent with the lesson [~M] (much more time than reading twice, and rated harder); the "two to three times" figure is a proposal read from that claim and from the prompt-based variant in Elaborative Interrogation Improves Learning [~M], where prompted learners spent about three times as long with no retention or transfer gain.

Dose and progression across sessions (all untested proposals). Within one session: about 5 minutes to probe what learners know and model one answer, then 15 to 25 minutes of reading with why-prompts (around 10 to 20 prompts), then 5 minutes comparing a few answers with a reference answer. Across a unit of several sessions:

  • Early sessions: the teacher or text supplies every prompt, one modelled answer is shown before each set, and the teacher reads a sample of answers and supplies the explanatory link where answers fail.
  • Middle sessions: prompts are still supplied, but learners check their own answers against a reference answer or a partner's, and the teacher only samples.
  • Late sessions: learners mark the statements they find surprising and write their own why-questions for them, then answer each other's; the teacher checks that the questions target causes or links rather than definitions.

Move a learner from supplied to self-generated prompts once most of their answers in a set name a reason or a link to known material, for two sets in a row. Move back to explanation or a worked example when most answers in a set are missing or wrong. If retention matters, retest the key statements at a delay (a week or more) rather than inferring it from answers during study; the six-month result in Elaborative-interrogation learning gains are durable, with performance maintained up to six months after study [+M] is from one study's recognition tests.

When a condition is not met: if learners know little about the topic, teach it first and use elaboration later in the unit; if the content is arbitrary (labels, foreign vocabulary, symbols) and there is nothing meaningful to connect it to, use a different technique (mnemonics or retrieval practice) rather than why-prompts.

Fitting the design to a situation

The three facts that most change the decision:

  • What learners already know about this material. Elaboration builds on knowledge the learner has; with little of it, answers are poor and the gain shrinks. If unstated, ask: can a typical learner say anything accurate about why these statements might be true before the lesson?
  • What kind of target it is. Facts and prose content suit why-prompts; a calculation, rule or procedure the learner cannot yet do needs an explanation or worked example first; arbitrary associations have nothing to explain. If unstated, ask: what must learners be able to do afterwards: recall and recognise statements, explain them, or carry out a procedure?
  • How much time there is, and who will look at answers. Elaboration costs time, and its quality depends on someone or something checking answers when learners are novices. If unstated, ask: how many minutes are there for this material, and who will see learners' answers, and when?
If the brief says… Then change… Basis
Learners are advanced or intermediate on this topic Use why-prompts freely, on denser material, with fewer modelled answers; move early to learner-written questions Elaborative interrogation benefits increase as learners' prior knowledge increases [~M] (one study reported in a review: larger gain in the high-knowledge domain)
Learners are novice on a calculation, rule or procedure Explain it or show a worked example first; put prompts on the steps of the example; do not replace a second reading with why-prompts on a lesson met cold Elaborative interrogation produced lower posttest accuracy than reading the lesson twice in a written lesson on posterior probability [-M] (undergraduates, one written lesson)
Learners are adolescent (lower secondary) reading science or social-studies text Why-prompts on each fact, answered in writing or aloud; target facts that contradict common beliefs Elaborative interrogation instructions produce better recall and recognition of science facts than reading-for-understanding instructions [+M] (grades 6 and 7)
Learners are child (under 12) Ask why-questions aloud in a small group, one statement at a time, accept spoken answers, and give the reason after each answer untested proposal; the experiments recorded here are with grade 6 and 7 students and older
Learners are adult-workplace Have them connect each statement to a case from their own work ("where have you seen this, and why did it happen?") rather than to textbook examples Readers given a perspective or schema recall more text information than readers instructed to carefully read the story [+W] (one story-recall experiment reported second-hand in an essay; not tested with workplace learners); the work-case prompt is an untested proposal
Goal is to correct misconceptions Put why-prompts on the statements that contradict the belief, and check confidence as well as accuracy Elaborative interrogation produced greater pretest-to-posttest gains on items about common misbeliefs than reading control in Experiment 2 [+M]
Goal is arbitrary associations (vocabulary, labels, symbols) Do not use why-prompts; use mnemonics or retrieval practice untested; from the earlier page's constraint (no claim here tests elaboration on arbitrary material)
Goal is a whole-task performance the brief wants kept whole Keep the task whole; ask "why that step?" at two or three decision points during or after the performance, not as separate paper items untested proposal
Setting is online-self-paced, no teacher Embed the prompts in the text with a text box; after each answer, reveal a short reference answer to compare against untested proposal; carried from the classroom biology study (Elaborative interrogation improved true/false test performance in an introductory biology course context [+M]), which did not test the self-check
Mixed modes: some learners remote, some in the room, or phone-only Use short typed answers (one sentence) in a shared chat or form so remote and in-room answers are seen together; read two or three aloud, from both groups untested proposal
One teacher with a large group Everyone writes an answer; the teacher reads a sample of five, models one strong answer to the class, and gives the missing reason for the commonest failure untested proposal; answer quality was poor in Answers to elaborative interrogation questions were mostly poor in quality, and answer quality correlated positively with posttest scores [~M]
Scarce facilitator time (volunteers, a busy shift, a tutor who checks in briefly) For fact learning, skip marking individual answers; give a reference answer for self-checking and use the check-in to model one answer The quality of elaborative-interrogation answers had little effect on retention; even failed or inadequate why-answers facilitated learning relative to reading [~M] (science facts; not shown for procedures)
A fixed short time budget, or a single session only Elaborate only the five to ten statements that matter most; plan about 20 minutes for 10 prompts with one modelled answer; do not claim retention from one session Elaborative interrogation increased perceived difficulty and time spent with the lesson [~M]; minutes are an untested proposal
High stakes, with an unaided test of the same content soon Pair elaboration with retrieval: answer the why-prompt, then later recall the statement without the text; prefer retrieval over map-building when time is matched Concept mapping improves learning over most comparison conditions by a moderate average amount, but not over time-matched retrieval practice [~M]; Whether prompting students to construct explanations during retrieval adds benefit beyond initial free recall is mixed across studies [~M]
Learners with low confidence or low literacy Choose familiar passages; let answers be spoken; give a sentence frame ("This is true because…"); tell them a partial answer still helps Elaborative processing of prose enhances later retrievability, accounting for roughly 30% of variance in one-month recall when original learning is controlled [+M] (replicated with less able middle-school students); frames are an untested proposal
Whole-class mathematics discussion When a student offers a formula, ask them to show the structure it stands for, then reconnect it to the formula Explicitly unfolding higher-order contributions into lower-order concept elements produces deeper conceptual learning opportunities than jumping directly to procedure [+M] (qualitative; learning opportunity, not measured learning)
The brief names its own routine, prompt, rubric or categories (its own question stems, a reading log, a check-in) Keep the named routine and its categories; change only what the evidence says matters: attach one why-prompt to specific statements, and add one modelled answer untested proposal (wave lessons)
The brief reports an error learners actually made Use the wrong answer as a contrast: put a why-prompt on the correct statement beside the common wrong one untested proposal
Private or sensitive material (personal experiences, health, family) Learners connect content to their own experience privately; only answers they choose to share are read untested proposal
A product or performance for a real audience Keep elaboration in the learner's drafting notes or a debrief, not in the product the audience sees untested proposal
The brief asks for no test or quiz Judge from the answers themselves and from later use; add no quiz untested proposal

Twelve of the twenty-one rows cite a claim; five of those carry a finding to a setting it did not test and say so (the workplace, online, large-group, low-literacy and high-stakes rows). The age evidence is from grade 6 and 7 students and undergraduates only.

Observation, state and explanation

What can be observed is an answer to a prompt under stated conditions: the statement, the prompt, the time taken, what the learner wrote or said, whether it names a cause or link to known material, and later recall or recognition of that statement. "Has integrated it with prior knowledge" and "lacks the background" are inferred states.

Response Plausible explanations Observation that could change the interpretation
Answers restate the fact ("because it is") No relevant knowledge to draw on; does not understand what a why-answer is; low effort or time pressure Ask what they know about the topic in general; show one modelled answer and give one more prompt; compare answers given with and without a time limit.
Answers are confident and wrong A misconception being recruited; an analogy carried too far Ask for the reason behind the reason; give a statement where the misconception and the correct idea predict different outcomes.
Good answers during study, poor recall a week later Answers came from the text, not from the learner's own knowledge; little review; interference Retest with a new prompt at the delay; check whether answers repeated the text's own words.
Fast answers, no gain over reading Prompts on statements learners already knew; prompts answered superficially Probe which statements were already known at the start; vary which statements carry prompts.
Learner complains it is slow and hard Elaboration is effortful and feels harder than reading; or the material is beyond them Compare their answer quality with their time: poor answers and long time suggest missing knowledge; good answers and long time suggest normal cost.

These rows are proposals, untested with learners as diagnostics. A probe is itself elaboration and changes what it measures; record each.

Evidence and qualifications

Learners, materials and outcomes differ across these (grade 6 and 7 science facts, an undergraduate biology text, an undergraduate probability lesson, an accounting class), so do not rank them by effect label. The only delayed evidence is one study's recognition tests and one report's one-week and one-month recall.

The evidence by question

  • Does answering "why is this true?" about each statement improve memory and understanding of it over reading it, and when does it not? Elaborative Interrogation Improves Learning [+M]
  • Does having learners generate their own elaborations improve learning, compared with no elaboration or with elaborations supplied to them? Elaborative Encoding Improves Retention [+M]
  • Does knowing more about the material beforehand mean learning more from it? Prior Knowledge Determines New Learning [~M] (its table counts 3 distinct studies before this page's additions; a meta-analysis there finds prior knowledge strongly predicts later knowledge but barely predicts normalised gains)

Further evidence, not yet read against this model

Claims the earlier page (and the pages folded into it) cited, not used above. Each says what it is and how far it bears on this page; none tests elaborating content against reading it.

Objective and learner-valued goal

The designer's objective is usually recall, recognition or explanation of specific content at a stated delay, with the aids stated. Elicit separately what the learner wants from the material: to pass Friday's test, to understand something that puzzled them, to be able to use it at work. Agreement and divergence are both observations. A learner who wants only to pass a recognition test may see why-prompts as slow and skip them; one who wants to understand may value them and still answer poorly if the background is missing.

For example, a nursing assistant reading about infection control may want to know why a step matters so they can explain it to a patient's family, while the course objective is passing a multiple-choice module test. Why-prompts serve both if the answers draw on the learner's own ward experience; record whether they can explain the step to someone, not only the test score.

What would revise this model?

The expectation should weaken if further comparisons of why-prompts against reading or rereading, with comparable learners and aligned delayed tests, show no advantage when study time is matched, or if the six-month result does not replicate. If answer quality turns out to matter for facts as well as for principles, the design should require checked answers, not just answering. If the probability-lesson result recurs for other principle and procedure targets with novices, the page should restrict why-prompts to content learners can already partly explain. If self-generated and supplied elaborations come out equal in larger studies, step 5 should go. Do not protect the model by calling every failure "not enough prior knowledge" after the fact; record the prior-knowledge probe before the lesson.

Immediate recall, delayed recall, explanation and transfer are separate claims; the present evidence establishes no number of prompts, no answer-quality threshold and no rule for when to move from supplied to self-generated prompts.

Claims

  • Analogical Reasoning — analogy generation is one of the most powerful forms of elaboration, linking new material to known structures
  • Annotating — margin notes and annotations are a practical vehicle for learner-generated elaboration during reading
  • Chunking — elaboration organizes information into meaningful chunks, reducing working-memory load
  • Activation — prior knowledge must be retrieved before it can be connected to new content
  • Asking Questions — question generation is an elaborative act that directs attention to explanatory gaps

Examples

Validated

  • Comparing contrasting cases improves learning. [+M] — Schwartz & Bransford's (1998) "time for telling" studies showed that learners who first compared and elaborated on contrasting cases of bird-adaptation data learned far more from a subsequent lecture than lecture-only controls.

Illustrative

Elaborative interrogation prompts — After each paragraph of a text, learners answer "Why does this make sense?" or "Why is this true?" rather than summarizing. A widely studied classroom variant of articulation; effective mainly when learners have background knowledge to draw on.

Concept mapping — Learners construct node-and-link diagrams connecting new concepts to each other and to prior knowledge, labeling each link with a relational phrase. Tools such as CmapTools support this; the mapping task forces explicit elaboration of relationships rather than listing facts.

"Why?" chains in science discussion — A teacher repeatedly asks "Why do you think that happens?" during class discussion, pushing students past restating a phenomenon to proposing mechanisms. Common in CGI-style math and science discourse routines.

Self-generated examples — After studying a definition or principle, learners must produce their own example that fits it and explain why it fits. Stronger retention than studying provided examples, because generation forces elaborative encoding.

Elaboration through analogy writing — Learners write their own analogy for a new concept (e.g., "an electrical circuit is like…") and then evaluate where the analogy breaks down, combining generative elaboration with critical comparison.

Key Sources

  • Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., & Willingham, D. T. (2013). Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology. Psychological Science in the Public Interest, 14(1), 4–58. doi:10.1177/1529100612453266
  • Craik, F. I. M., & Lockhart, R. S. (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11(6), 671–684. doi:10.1016/S0022-5371(72)80001-X
  • Pressley, M., McDaniel, M. A., Turnure, J. E., Wood, E., & Ahmad, M. (1987). Generation and precision of elaboration: Effects on intentional and incidental learning. Journal of Experimental Psychology: Learning, Memory, and Cognition, 13(2), 291–300. doi:10.1037/0278-7393.13.2.291
  • Ozgungor, S., & Guthrie, J. T. (2004). Interactions between elaboration and knowledge-building processes in learning from text. American Educational Research Journal, 41(2), 403–439.
  • Schwartz, D. L., & Bransford, J. D. (1998). A time for telling. Cognition and Instruction, 16(4), 475–5223. doi:10.1207/s1532690xci1604_4
  • Clinton, V., Alibali, M. W., & Nathan, M. J. (2016). Learning About Posterior Probability: Do Diagrams and Elaborative Interrogation Help? The Journal of Experimental Education, 84(3), 579-599. https://doi.org/10.1080/00220973.2015.1048847
  • Bigenho, Frederick W., Jr. (1992). Conceptual Developments in Schema Theory. https://eric.ed.gov/?id=ED351392
  • Xiong, Y., Zhou, H., & Ogilby, S. M. (2014). Experimental Investigation of the Effects of Cognitive Elaboration on Accounting Learning Outcomes. Journal of Education and Learning, 3(4). https://doi.org/10.5539/jel.v3n4p1
  • Ademmer, C., & Prediger, S. (2025). How Can Ideas Be Connected Afterwards? Decomposing Teachers' Facilitation Practices for Conceptual Learning in a Case of Formal Volume Calculation. Cognition and Instruction, 43(4), 355-388. https://doi.org/10.1080/07370008.2025.2527688
  • Gagne, Ellen D. (1981). Comprehension and the Long-Term Recall of Information. Final Report. https://eric.ed.gov/?id=ED221831
Page metadata
Field Value
type principle
id elaborative-reasoning
aliases model-and-monitor-elaborative-interrogation-answer-quality, precise-meaningful-elaboration-facilitates-encoding, promote-self-generated-elaboration-for-expertise, teach-information-with-prior-related-knowledge
title Elaborative Reasoning
description For a learner with relevant prior knowledge studying facts or prose, answering why-questions about each statement or generating their own connections is expected to improve recall and recognition over reading it, at a cost in time; randomised experiments with school students test this for science facts, and one experiment finds the reverse for novice undergraduates learning a probability calculation.
status review
moved_from pattern/grounding-and-unfolding-facilitation-moves
canonical True
generated by claude/unspecified on 2026-10-15
trust unverified — no one has confirmed this page's content
sources 5 — clinton-2016, bigenho-1992, xiong-2014, ademmer-prediger-2025, gagne-1981

Frontmatter is the page's machine-readable record — see the Schema & Guide.