Retrieval practice improves long-term retention¶

Claim · All claims
Evidence · 4 studies · 2 quant-synthesis r3, 1 causal r?, 1 review r3 · q3–q4 · i2–i3
Currency · newest study 2017 · not yet checked against its sources · not reviewed by a person

Actively recalling information from memory (rather than rereading or reviewing it) strengthens long-term retention of that information. This is one of the most consistently replicated effects in cognitive psychology of learning, often called the "testing effect."

Subclaims¶

q4 i2 Across 159 effect sizes from 61 studies, taking a practice test produced better retention than restudying the same material. The benefit was larger after delays of one day or more, and when the initial test used recall or gave feedback. With no feedback and initial test performance of 50% or below, the benefit was about zero. → Rowland 2014

q3 i3 With prose passages, students who recalled the material outperformed students who restudied it on tests two days and one week later. On a test five minutes later, restudying was ahead. → Roediger and Karpicke 2006

q3 i3 Repeated testing beat repeated study at a one-week delay, 61% against 40% recall, after both groups had spent the same time with the passage. → Roediger & Karpicke 2006

q3 i? The ordering reverses at a short delay: on a test taken five minutes after study, repeated study scored higher than repeated testing. The benefit is a property of the retention interval, not of the encoding session. → Roediger & Karpicke 2006

q4 i2 Meta-analysis of testing against restudy put the mean effect at g = 0.50, and the advantage grew with the length of the retention interval. → Rowland 2014

q4 i2 An independent meta-analysis of practice testing against rereading and other comparison conditions found g = 0.51, with the effect holding across education levels and item formats. → Adesope et al. 2017

q3 i2 The chapter reports that Rowland's (2014) meta-analysis found an overall retrieval practice versus repeated study effect of g = 0.50, with 81% of comparisons favoring retrieval practice. → Karpicke 2017

Evidence¶

Rowland 2014¶

Rowland, C. A. (2014). The effect of testing versus restudy on retention: A meta-analytic review of the testing effect. Psychological Bulletin, 140(6), 1432–1463. doi:10.1037/a0037559

q4 · meta-analysis · i2 · medium effect, g=0.50 · n=159 effect sizes from 61 studies · quant-synthesis · r3

This random-effects meta-analysis covers 159 effect sizes from 61 studies reported between 1975 and 2013. Each effect compares information that learners were tested on with information they restudied. The mean weighted effect was g = 0.50 (95% CI 0.42 to 0.58), and heterogeneity was high. The effect was larger at retention intervals of at least one day (g = 0.69) than below one day (g = 0.41). It was also larger when the initial test gave feedback (g = 0.73 against 0.39 without) and when the initial test was cued recall rather than recognition (0.61 against 0.29). With no feedback and initial test performance of 50% or below, the effect was about zero (g = 0.03, CI −0.21 to 0.27). Published studies showed larger effects than unpublished ones (0.58 against 0.25), so the author advises caution about publication bias.

Roediger and Karpicke 2006¶

Roediger, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science, 17(3), 249–255. doi:10.1111/j.1467-9280.2006.01693.x

q3 · peer-reviewed experiment (two experiments) · i3 · large effect, d=0.83 at 1 week (Exp. 1) · n=300 undergraduates (120 + 180) · causal · r?

Undergraduates at Washington University read short science passages. In Experiment 1 (120 students), they either restudied a passage or took a free-recall test on it, without feedback. At five minutes, restudying did better (81% against 75% recalled). At two days, testing did better (68% against 54%, d = 0.95), and at one week also (56% against 42%, d = 0.83). In Experiment 2 (180 students), one study period followed by three recall tests beat four study periods on the one-week test (61% against 40%, d = 1.26). Repeated study scored best at five minutes and made students more confident they would remember, but it did worst at one week.

Adesope et al. 2017¶

Adesope, O. O., Trevisan, D. A., & Sundararajan, N. (2017). Rethinking the use of tests: A meta-analysis of practice testing. Review of Educational Research, 87(3), 659–701. doi:10.3102/0034654316689306

q4 · i2 · quant-synthesis · r?

Meta-analysis of practice testing across education levels, comparing it against rereading, filler tasks and no treatment. Mean g = 0.51. The effect appeared for multiple-choice as well as constructed-response practice tests, which matters for a system whose selection items would otherwise be assumed inferior on this dimension.

Karpicke 2017¶

Karpicke, J. D. (2017). Retrieval-Based Learning: A Decade of Progress. Learning and Memory: A Comprehensive Reference, 2nd edition, Volume 2. https://doi.org/10.1016/B978-0-12-809324-5.21055-9

q3 · i2 · review · r3

The chapter reports Rowland's (2014) meta-analysis as its quantitative synthesis: "Across 159 studies, the overall effect size of retrieval practice relative to repeated study was g ¼ 0.50", and 81% of comparisons favored retrieval practice. The chapter did not compute this estimate itself.

"Across 159 studies, the overall effect size of retrieval practice relative to repeated study was g ¼ 0.50, and 81% of comparisons favored retrieval practice over repeated study."

Studies bearing on this claim¶

Does retrieving material (quizzes, practice tests, recall) improve later retention more than restudying it? This table gathers the studies behind this claim and the 28 narrower claims that bear on it: 14 distinct studies (5 quant-synthesis, 5 causal, 3 review, 1 qualitative), q1–q4, 9 of 14 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¶

Mechanism. Retrieval practice is typically explained as a memory-modification effect: each successful retrieval attempt strengthens and diversifies the retrieval routes to a memory, making future access more likely. This contrasts with restudying, which re-exposes learners to material but does not exercise the retrieval pathway itself. The claim connects to broader information processing theory accounts of encoding and consolidation, and it is a core component of many active learning designs.

Boundary conditions. The benefit depends on retrieval being successful or quickly corrected — repeated failed retrieval with no feedback can entrench errors [-M]. Benefits also vary with material type (stronger for factual and paired-associate content than for complex skill performance) [~M] and with learner ability, and the added difficulty of retrieval must be managed so it does not induce overload in the sense described by cognitive load theory [~M]. As with other desirable-difficulty effects, an expertise-reversal pattern is plausible: very novice learners may lack enough knowledge for retrieval attempts to succeed productively, echoing the moderation documented for expertise reversal [~W].

Design implications. Low-stakes quizzing, closed-book free recall, and brief write-to-recall prompts are the most direct implementations; rereading and highlighting are the common but weaker alternatives they displace. Retrieval attempts pair naturally with spacing of practice and with feedback that corrects unsuccessful attempts quickly. In course designs, retrieval works best as a routine element — brief recall prompts at the start of sessions, exit tickets, or cumulative low-stakes quizzes — rather than as a one-off review activity before an exam. Because retrieval is effortful, learners often misjudge its value relative to the fluency of rereading; designers should expect resistance and communicate the counterintuitive benefit explicitly.

Open questions. The durability of effects across long retention intervals, transfer to applied tasks beyond the retrieved content, and optimal scheduling (spacing of retrieval attempts) all require further evidence before strong design prescriptions can be made.

Evidence status. The Evidence section records three studies (Rowland 2014; Roediger & Karpicke 2006; Adesope et al. 2017): two meta-analyses of testing against restudy and one two-experiment laboratory study; the subclaims above read from those entries.

Merged from “Retrieval Practice Enhances Long Term Retention” (retrieval-practice-enhances-long-term-retention): The proposed mechanism is that retrieval is itself a learning event: successfully reconstructing information from memory modifies the memory trace and makes future retrieval more likely, whereas restudying only re-exposes learners to material without requiring reconstruction. This distinguishes retrieval practice from mere re-exposure and places it within cognitive load theory and broader accounts of how chunking reduces working memory load during encoding and consolidation.

Several boundary conditions are worth noting. Retrieval practice is effortful, and learners often misjudge its value because successful retrieval feels harder and less productive than rereading — a fluency illusion that can lead learners to prefer less effective strategies [-M]. Retrieval attempts that fail entirely, without feedback or subsequent successful retrieval, may contribute little; the benefit is generally tied to successful (or eventually successful) retrieval [~M]. The effect also depends on the material being retrievable and on learners having sufficient prior knowledge to attempt retrieval meaningfully — for very novice learners, unsupported retrieval attempts can impose extraneous load and contribute to conditions where cognitive overload degrades learning [~M].

Design implications follow directly. Low-stakes quizzing, closed-book recall prompts, and assessment for learning designs all operationalize retrieval practice; the key is that learners must generate answers from memory rather than recognize or reread them. Because retrieval is demanding, it should be introduced with feedback and adequate scaffolding, consistent with cognitive load reduction principles, and paired with activation of relevant prior knowledge so that retrieval attempts are meaningful rather than guesswork.

Open questions that evidence entries should address include: how retrieval practice compares with restudying across retention intervals (the effect is expected to grow as the retention interval lengthens), how feedback moderates the effect, whether benefits transfer beyond memorization to inference and application, and how effect sizes vary across domains, learner ages, and formats (free recall, cued recall, multiple choice).

Merged from “Retrieval Practice Improves Learning” (retrieval-practice-improves-learning): Mechanism. Retrieval practice is generally explained as a memory-modification effect: the act of successfully reconstructing information from memory changes that memory's accessibility, making it easier to retrieve again later. This distinguishes it from restudying, which re-exposes learners to material but does not exercise the retrieval route itself. It also explains why retrieval works best when embedded in practice rather than passive review.

Moderators to expect. The size of the benefit typically depends on retrieval success (learners must actually retrieve, not fail repeatedly), feedback availability after retrieval, the interval between learning and test, and the match between retrieval format and the criterion task. Designers should treat these as open design parameters rather than assuming a uniform effect across all implementations. Spacing repeated retrieval attempts over time, consistent with spaced practice, generally amplifies the effect; massed retrieval yields smaller durable gains.

Boundary conditions. Retrieval practice is not a substitute for initial encoding: if learners have not formed a retrievable memory in the first place, retrieval attempts can be unproductive or discouraging. The claim should be scoped to material that has been at least minimally learned. Where retrieval demands exceed working-memory capacity, chunking the target material first improves the odds of successful retrieval.

Open questions. How retrieval practice effects scale from laboratory materials to complex, higher-order learning outcomes in authentic classrooms remains an active area of investigation, as does the optimal scheduling of repeated retrieval attempts. Low-stakes quizzing in real courses offers a natural bridge between the laboratory finding and assessment for learning.

Merged from “Testing Effect Improves Learning” (testing-effect-improves-learning): Mechanism. The dominant explanation is that successful retrieval modifies the underlying memory trace, making it more accessible later, and that the effortful, error-prone character of retrieval — unlike passive rereading — is what drives the benefit. Retrieval also functions as a diagnostic: it exposes gaps that restudying conceals, which supports subsequent study decisions and connects to Assessment for Learning.

Boundary conditions. The benefit depends on retrieval success and feedback. Tests that learners largely fail, or that provide no corrective feedback, can entrench errors rather than strengthen correct knowledge [-M]. The effect is strongest when retrieval is effortful but successful — the "desirable difficulty" framing — and diminishes when the task is so hard that retrieval fails entirely [~M]. The benefit also grows over retention intervals: much of the advantage appears at delayed test rather than immediately, which means immediate post-tests can understate the effect [~M].

Design implications. Low-stakes quizzing, free recall prompts, and flashcard-style practice are the standard delivery vehicles; the key design variable is requiring actual retrieval rather than recognition or rereading. Feedback after retrieval attempts preserves the diagnostic function and prevents error entrenchment. Spacing retrieval attempts across time compounds the benefit, linking this claim to spacing and Chunking considerations about how much material a single retrieval attempt can reasonably cover. Because retrieval practice is effortful, it pairs poorly with designs that simultaneously impose heavy extraneous load — see Cognitive load reduction improves learning. Learners' preference for rereading over testing makes explicit use of low-stakes retrieval important; without it, learners default to less effective study strategies [~M].

Open questions. How retrieval practice transfers beyond the exact retrieved items (e.g., to inference and application) is less settled than its effect on verbatim retention, and optimal spacing of retrieval attempts remains an active design question.

Merged from “Testing Effect Improves Retention” (testing-effect-improves-retention): Mechanism. The testing effect (also called retrieval practice) is generally explained by the idea that the act of successful retrieval itself modifies the underlying memory trace, making it more accessible later — a "desirable difficulty" in Bjork's sense. Restudying, by contrast, can create fluency illusions: material feels familiar, but that familiarity does not predict durable retention. This connects to broader claims that cognitive overload degrades learning when effort is misallocated, and that effortful processing — like chunking — shapes what survives in information processing.

Boundary conditions to watch for. The benefit depends on successful retrieval: if learners fail the test and receive no feedback, the effect shrinks or reverses [-M]. It is strongest when retrieval is effortful but successful (e.g., after a short delay rather than immediately), when feedback follows errors, and when tests are low-stakes so the effect is not confounded with test anxiety [~M]. Repeated spaced retrieval outperforms a single retrieval [+M]. These moderators should be confirmed against specific studies once Evidence entries are added.

Design implications. Because the effect requires retrieval to be effortful but successful, designers should schedule low-stakes quizzes after a short delay rather than immediately after study, provide corrective feedback, and space repeated retrieval opportunities across sessions. In classroom settings, frequent low-stakes quizzing (rather than a few high-stakes exams) captures the retention benefit while avoiding anxiety confounds. Retrieval practice also pairs naturally with active learning formats such as clicker questions and free recall, and functions as a form of assessment for learning when quiz results feed back into instruction.

Open questions. How the effect scales from lab materials (word lists, prose passages) to complex classroom content, and how it interacts with active learning formats, remain active areas of replication and meta-analysis.

Merged from “Retrieval practice improves long-term retention more than restudying” (testing-effect-retrieval-practice-improves-retention): The testing effect is widely regarded as one of the most robust findings in memory research; the Evidence section above records three studies (Rowland 2014; Roediger & Karpicke 2006; Adesope et al. 2017).

Boundary conditions. Retrieval practice benefits depend on successful retrieval attempts — learners must actually retrieve, not merely reread — and feedback after retrieval is important when initial attempts fail. Benefits are typically larger for delayed retention than for immediate test performance, which means short-term assessments can understate the effect. The effect also interacts with learner expertise and material complexity: for complex, high-element-interactivity material, the added demands of retrieval may impose load, connecting this claim to Cognitive Load Theory and potential expertise-reversal dynamics (see Expertise Reversal Effect).

Design implications. Because the benefit accrues from the act of retrieval rather than the presence of a test, low-stakes quizzing, free recall prompts, and flashcard routines can all serve as retrieval practice — see Practice and Assessment. Retrieval attempts should be spaced and followed by feedback when retrieval fails, and designers should avoid substituting recognition formats (multiple choice) for recall formats when the goal is durable retention, since recognition can often be performed without productive retrieval. Well-designed Assessment for Learning can double as retrieval practice when treated as a learning event rather than purely a measurement event.

Open questions. How retrieval practice combines with other strategies such as Spaced Practice and interleaving, and how it transfers to complex or applied outcomes rather than retention of discrete facts, remain active areas of investigation. Whether retrieval practice confers benefits beyond retention — for organizing knowledge or supporting transfer — is contested in the literature and should be resolved with cited evidence before being asserted on this page.

Merged from “Testing Improves Retention” (testing-improves-retention): Mechanism. The retrieval-effort account holds that the act of successfully reconstructing knowledge from memory changes its storage — each successful retrieval makes the memory more accessible on future attempts, whereas rereading primarily increases fluency without strengthening retrieval pathways. This is why testing is best understood as a learning event, not merely an assessment event, and why it pairs naturally with assessment for learning approaches. The mechanism also connects to broader active learning accounts: retrieval is an active, generative act, in contrast to the passive reception of restudy.

Boundary conditions. The benefit depends on successful retrieval: tests that learners largely fail, or that occur before the material has been encoded at all, produce far weaker effects. Feedback after retrieval matters most when initial retrieval fails. The effect also varies with test format — free recall and cued recall tend to produce larger retention benefits than multiple-choice recognition, though recognition-based retrieval can still help. Spacing repeated retrievals over time amplifies the benefit relative to massed testing, consistent with distributed practice findings. Retrieval that exceeds working-memory capacity — for example, recalling material that was never chunked into manageable units — can impose cognitive overload that undermines the benefit, a concern that falls under cognitive load management.

Open questions. How retrieval practice transfers to complex, application-level outcomes (as opposed to retention of discrete facts) remains actively debated, and the optimal scheduling of retrievals relative to forgetting points is not fully settled.

Note on evidence. The evidence recorded above supports the claim, with one condition worth stating: in Roediger and Karpicke (2006), restudying was ahead on a test five minutes later, and testing won only at two days and one week. The retrieval-practice literature is large and well-established, so populating this section — ideally with the major meta-analyses and the canonical restudy-comparison experiments — should be a high priority for a future enrichment pass.

Merged from “Retrieval Practice Improves Long Term Retention” (retrieval-practice-improves-long-term-retention): Mechanism. Retrieval is assumed to strengthen memory by requiring learners to reconstruct knowledge from memory rather than re-encode it passively — consistent with accounts in Information Processing Theory. Successful retrieval is thought to both consolidate the retrieved item and render it more accessible on future attempts. A second proposed mechanism is that retrieval attempts make learners aware of what they do not yet know, supporting better allocation of subsequent study — a diagnostic function related to Assessment for learning improves achievement [+M].

Relation to other effects. Retrieval practice is a form of active learning [+M], and its benefits are often discussed alongside spacing and interleaving, which manipulate when and how varied retrieval attempts are rather than whether they occur. Retrieval also interacts with working-memory demands: overly complex retrieval tasks can overload learners, echoing concerns in Cognitive overload degrades learning [~M] and Chunking reduces working memory load [+S].

Boundary conditions to document. The size of the testing effect is expected to depend on several moderators, which the entries above report only in part: whether retrieval is successful or failed on the first attempt, whether feedback follows the attempt, the complexity of the material, and learner expertise. For very complex tasks, retrieval demands may exceed working-memory capacity, an expertise-reversal-style boundary condition analogous to Worked examples can become redundant or counterproductive for advanced learners. [~M]. Low-stakes or ungraded formats are commonly recommended so that retrieval practice functions as learning rather than as high-pressure evaluation.

Measurement artifact risk. A recurring methodological caution: on an immediate final test, restudy can match or beat retrieval, so claims about the testing effect must specify delayed outcomes. Studies that only measure immediate performance cannot support this claim, and designers should be wary of implementations (e.g., flashcard apps) evaluated solely on short-term gains.

Design implications pending evidence. Until evidence entries are verified, practitioners should treat the strong rating as provisional. The most defensible design heuristics consistent with the claim's scope are: require learners to generate answers before re-exposing them to material; schedule retrieval attempts after a delay rather than immediately after study; and keep retrieval tasks low-stakes so errors carry diagnostic rather than evaluative weight. These follow from the claim's stated mechanism and boundary conditions, not from cited experiments on this page.

Learner Variables¶

• Long-Term Memory — outcome: instruction changes it

Related Claims¶

• Active learning improves exam performance. — retrieval-oriented activity is a central ingredient of active-learning effects on exams
• Chunking reduces working memory load. — well-chunked material makes retrieval attempts more likely to succeed
• Cognitive overload degrades learning. — overly difficult retrieval can cross into counterproductive load
• Activation improves learning. — pre-instruction retrieval of prior knowledge is a related, forward-facing use of recall
• Assessment for learning improves achievement — low-stakes formative quizzing embeds retrieval practice into instruction.
• Cognitive load reduction improves learning — scaffolding retrieval attempts keeps them productive rather than overwhelming.
• Cognitive load management. — retrieval demands must fit within working-memory limits
• Cognitive Load Theory — frames when retrieval demands help versus hinder learning
• Information Processing Theory — explains how retrieval strengthens memory traces over time
• Cognitive Load Reduction Improves Learning — retrieval imposes load; benefits depend on managing it
• Chunking Reduces Working Memory Load — chunked material is easier to retrieve successfully
• Advance organizers aided learning in a 1980 meta-analysis, but activating prior topic knowledge did not improve primary pupils' text comprehension in one experiment — prior retrieval of related knowledge primes new learning
• Assessment for Learning Improves Achievement — assessment used as a learning event overlaps with retrieval practice
• Active Learning Improves Exam Performance — retrieval-based classroom activity is a core active-learning mechanism
• Distributed practice improves retention. — spaced retrievals amplify the testing effect relative to massed testing
• Retrieval practice and spaced retrieval improve achievement and retention by medium amounts, while interleaving helps category learning only for some materials and blocking beats it for words — possibly the same claim (merge candidate)
• Feedback Enhances Retrieval Practice — related
• High-confidence errors are more likely than low-confidence errors to be corrected on a later test after feedback. — a narrower finding that bears on this claim
• Gains from an intensive period of second-language study persist for weeks without rehearsal. — a narrower finding that bears on this claim
• Faster rate of learning may be negatively related to long-term retention (efficiency-effectiveness trade-off) — related
• Repeatedly recalling a text without feedback or rereading produces the best one-week retention, while repeated studying leads on a final test at the end of the session — a narrower finding that bears on this claim
• Learning should be assessed with retention and transfer tests rather than current practice performance — related
• Retrieval Fails Without Encoding — related
• Retrieval Failure Reduces Benefit — a narrower finding that bears on this claim
• Retrieval practice instructions improved final recall of translations in Experiment 1 — a narrower finding that bears on this claim
• Retrieval practice effects are larger at retention intervals greater than 1 day (g = 0.69) than at intervals less than 1 day (g = 0.41) in Rowland's (2014) meta-analysis — related
• A review reports that retrieval practice effects grow more robust as initial retrieval success rises, especially above 75%, and separately that giving more retrieval cues yields smaller effects — related
• Retrieval practice produces reliable positive effects at short delays, although some forgetting must set in so that the final test is not at ceiling — a narrower finding that bears on this claim
• Retrieval Practice Improves Transfer — related
• Spaced retrieval practice produces better final retention than massed retrieval even though spacing lowers initial retrieval success, and more absolute spacing enhances long-term retention — related
• Desirable difficulty: harder-to-learn conditions yield enhanced long-term retention — related
• Unsuccessful retrieval attempts confer a smaller benefit to future performance than successful retrievals, whose advantage grows with item difficulty — a narrower finding that bears on this claim
• A math teacher whose students had typically prepared for tests by rereading notes implemented three retrieval-practice strategies after learning that rereading is not effective for long-term retention — a narrower finding that bears on this claim
• In a Terracotta study across classes, retrieval practice review outperformed restudy review on subsequent exams — a narrower finding that bears on this claim
• After training, d.tech teachers adopted low-stakes practice tests and spaced, interleaved practice in their classrooms — a narrower finding that bears on this claim
• Retrieval practice strengthens memory and produces stronger, more enduring learning — a broader claim this one bears on
• Spaced and interleaved homework practice improves learning and retention — related
• Cognitive overload degrades learning — a boundary condition: retrieval that exceeds working-memory capacity may not help
• Repeated successful retrieval during learning predicted final recall in both experiments — related
• Final-test reading comprehension after retrieval practice does not differ significantly between an immediate test and a test delayed by three weeks — related
• Evidence tracing links 96 of 114 generated claims to source chunks, but three claims with retrieval scores of 0.50 or below show retrieval similarity alone is insufficient — related
• Error correction supports learning when it requires substantive cognitive work such as explanation and reconstruction — related
• VeriForge's integrated workflow is associated with higher one-week delayed recall of domain terms and a higher retention rate (exploratory) — a narrower finding that bears on this claim
• Struggling to encode, retrieve, and reorganize information produces deeper comprehension and retention, and AI-supplied ready-made solutions short-circuit these processes — a narrower finding that bears on this claim
• The pretesting effect on text memory grows with retention interval, being largest at one week — a narrower finding that bears on this claim
• Quizzing outperforms no/filler activity, fewer-question testing and restudying, with a larger advantage over no/filler activity (low reexposure) than over restudying (high reexposure); against other elaborative strategies its advantage is small and not significant. — related
• Retention interval moderates the testing effect: effects are larger after a day or more than at shorter intervals, where individual studies vary from reliable to null or reversed — related
• The meta-analysis supports retrieval effort as a contributor to the testing effect — a narrower finding that bears on this claim
• Rereading Less Effective Than Retrieval Practice
• Learners misjudge which learning strategies are effective — related
• A significant but negligible benefit of peer assessment (d = .122) persisted to the final exam in one college statistics course — related
• Repeated test-restudy trials produced higher final cued recall than study-only or single test-restudy controls across both experiments — a narrower finding that bears on this claim
• Active control over the learning experience enhances recognition memory from age 7 into adulthood — related
• Retrieval practice conditions produced significantly higher cognitive load than rereading — related
• Retrieval practice produces better retention than restudying, with meta-analytic reviews reporting people 2.5 times as likely to remember tested information and medium-to-large classroom effects — related
• Retrieval practice enhances learning across domains and the lifespan in applied educational settings — a broader claim this one bears on
• Targeted retrieval via specific short-answer tasks improves retention of directly targeted information more than holistic retrieval via free-recall tasks — a narrower finding that bears on this claim
• Students rarely open transfer-practice cards immediately after solving; they want such problems organized into delayed, spaced wrong-book review instead — related
• The authors attribute the spaced-practice knowledge benefit to deeper learning or transfer, because quiz and final-test items were never repeated — related
• The same quizzing or concept-mapping activity produces more learning when done without viewing the material, as retrieval, than while viewing it — a narrower finding that bears on this claim
• In authentic undergraduate courses, taking more voluntary practice tests predicts higher final-exam performance, with and without corrective feedback — a narrower finding that bears on this claim
• Spaced recall quizzing significantly improves overall post-test performance on fundamental mathematics without harming learning of current material — a narrower finding that bears on this claim
• Spaced recall quizzing improves long-term retention even when immediate feedback is not provided — related
• Recent research identifies time intervals during which spaced study should occur to maximize memory retention — related
• Weekly spaced recall quizzes on fundamental concepts significantly improve 6-month retention of those concepts in a precalculus course — a narrower finding that bears on this claim
• Generation as the sole desirable difficulty produced no detrimental effect on L2 vocabulary recall, in contrast to prior short-term findings — a narrower finding that bears on this claim
• Testing benefits retention over a conservative restudy control, establishing the testing effect as robust — a broader claim this one bears on
• Spacing out retrieval practice produces better long-term retention than massing it — related
• The retrieval benefit occurs at the level of individual items, not through general practice at retrieval — a narrower finding that bears on this claim
• Failure to obtain a testing effect with immediate tests may be partly due to working memory depletion — related
• Testing effect disappears for a coherent complex text but reappears, as reduced forgetting, when the same sentences are scrambled — reports the opposite
• Taking an immediate plus delayed final test yields higher delayed test performance with less invested effort than taking only a delayed test (Experiment 2) — a narrower finding that bears on this claim
• Across eight delayed-test comparisons, testing showed a small numerical advantage over restudy, but confidence intervals were wide and the combined effect imprecise — a narrower finding that bears on this claim
• The testing effect has been demonstrated across a variety of mostly low-element-interactivity materials, but problem-solving tasks had received very little testing-effect research attention — a narrower finding that bears on this claim
• Mental effort did not explain the testing effect: retrieval practice effort did not predict delayed test scores — a narrower finding that bears on this claim
• Retrieval practice yields better delayed test performance than restudy (the testing effect) in both experiments — a narrower finding that bears on this claim
• In the author's cited longitudinal classroom experiments, lessons with retrieval practice raised course grades from a C to an A level, and three spaced retrieval opportunities were optimal even nine months later — a narrower finding that bears on this claim
• In spaced practice with post-feedback memory ratings, restudied Swahili–English pairs were recalled better than covertly or overtly retrieved pairs on a delayed final cued-recall test. — related
• When the memory rating was removed entirely, retrieval practice was superior to restudying regardless of whether items were restudied or retrieved once, twice, or thrice. — a narrower finding that bears on this claim
• Closed-book generative tasks can outperform open-book ones when learners can later reinspect the material — a narrower finding that bears on this claim
• Retrieval practice improves delayed but not immediate performance, consistent with a desirable-difficulties explanation — possibly the same claim (merge candidate)