Grep · sample report

Literature Review — You get a cited review of the scholarly evidence on your question: key studies, methods, effect sizes, evidence strength, consensus, disagreements and gaps.

workon demandinsight
Inputs used for this sample
Literature Review

Research question: Does spaced repetition improve long-term retention in adult learners?

Setting or decision context (optional): corporate onboarding programs, mid-size firms

Evidence scope: years, study designs, populations (optional): 2010 onward; RCTs and meta-analyses; working adults

Spaced Repetition and Long-Term Retention in Adult Learners: Moderate Evidence of Benefit, Thin Evidence for Corporate Onboarding

1. Bottom line

Spaced repetition very likely improves long-term retention of knowledge in adults, but the evidence that applies directly to corporate onboarding is thin. Meta-analyses of spaced retrieval and spaced digital education report effects in the small-to-large range, for example g = 0.74 for spaced versus massed retrieval practice and SMD 0.32 (95% CI 0.13 to 0.51) for knowledge after spaced online education. I rate certainty as moderate that spacing beats massed training for knowledge retention in adult professionals, and low that it does so for workers in non-clinical corporate settings. The main caveat: nearly all trials with working adults come from health care, and 83% of the trials in the largest relevant meta-analysis had unclear or high risk of bias. Findings are my reading of study abstracts. I could not open several full texts (see section 2).

2. Search strategy

This was a time-limited review, not a full systematic review. It was run on 5 October 2026.

3. Key studies

Authors, year Design Sample Population and setting Main finding Effect size Flags
Latimier, Peyre, Ramus 2021 Meta-analysis, 29 studies 39 effect sizes (subset 1) Mixed learners, spaced vs massed retrieval practice Strong benefit of spacing; expanding vs uniform schedules did not differ g = 0.74; expanding vs uniform g = 0.034 (CIs not in abstract) Population not restricted to working adults
Martinengo et al. 2024 Cochrane-style SR and meta-analysis, 23 RCTs/quasi-RCTs 9 studies in knowledge outcome Pre- and post-registration health professionals, spaced digital education Spaced beat massed online education for knowledge SMD 0.32 (95% CI 0.13 to 0.51), I2 66%, moderate certainty; clinical behaviour SMD 0.67 (0.43 to 0.91); simulation skills SMD 1.15 (0.34 to 1.96), low certainty 83% unclear/high risk of bias; includes pre-registration students
Sezgin and Bektas 2026 Meta-analysis, 9 RCTs Not stated in abstract Health professions students Spaced learning improved retention Retention Hedges' g = 0.62 (95% CI 0.43 to 0.82) Students, not workers; authors rate overall risk of bias high; recent, 0 citations at indexing
Trumble et al. 2024 Systematic review, no pooled estimate 56 studies, 63 experiments Health professions education 43 of the 56 studies showed significant benefit of distributed and/or retrieval practice Vote count only Mixes spacing with retrieval practice; heterogeneous designs
Lam and Lam 2024 Cluster RCT, unblinded, wait-list 456 employees; 236 (51.8%) answered follow-up Employees in high-stress industries, mental health literacy Intervention beat control immediately and at 3 months on most outcomes Weighted mean differences 0.2, 0.9, 1.8 on three scales (P = .003 to <.001) Self-report outcomes; about 48% follow-up non-response (my calculation); no massed-training comparator
Kerfoot et al. 2010 (J Am Coll Surg) RCT 724 urology residents US residents, histopathology diagnosis Spaced education beat bolus web modules at long-term retest (weeks 18 to 45) Mean score gain 15.2% (SD 15.3) vs 3.4% (SD 16.3), p < 0.01 Author-built platform; trainees, not onboarding
Kerfoot et al. 2010 (Am J Prev Med) RCT 95 primary care clinicians VA clinicians, PSA ordering behaviour Fewer inappropriate tests, sustained over 72 weeks after the intervention 7.8% vs 13.1% inappropriate tests, p = 0.011 (40% relative reduction) Pharmaceutical funders listed among grants (Astellas, Wyeth); no-intervention control, so not a spacing test
Kerfoot et al. 2014 RCT, 52 weeks 111 clinicians; 14,336 patients Primary care clinicians, hypertension Modest faster time to blood-pressure target Hazard ratio 1.043 (95% CI 1.007 to 1.081) Patient outcome, not retention; control got same content as a posting
Kauffeld and Lehmann-Willenbrock 2010 Quasi-experimental field study 64 bank employees (32 per arm) Sales training, bank Spaced beat massed on transfer quality, self-rated competence and key figures; not on transfer quantity Not given in abstract Not randomised; small sample; the only corporate-trainee study surfaced by my limited search

Out-of-scope seminal work: Cepeda et al. 2006 pooled 839 assessments from 317 experiments in 184 articles and found the best gap between study sessions grows with the delay before the final test. It is a 2006 verbal-recall synthesis, mostly lab studies.

4. Synthesis

Where studies agree. Every synthesis found points the same way. Spaced formats beat massed formats on knowledge tests (Martinengo 2024; Sezgin 2026). The largest effect sizes in the retrieval meta-analysis come from comparing spaced to massed practice (g = 0.74). Individual RCTs in working clinicians show durable gains weeks to months after training (Kerfoot 2010).

Where they differ. Effects vary a lot: I2 of 66% for knowledge, and estimates from 0.32 to 1.15 SMD across outcome types. Likely reasons (inference, not tested in the studies): different gap lengths, different test delays, student versus worker samples, and whether the control was a different delivery of the same content. The expanding-schedule debate is settled in this data only to the extent that no difference was found between expanding and uniform spacing (g = 0.034), so a simple uniform schedule looks adequate.

Quality and bias. Most trials are weak. 83% had unclear or high risk of bias, and the 2026 meta-analysis authors rated the overall risk of bias high and urged caution. The 2024 review of 56 studies found mixed designs and unvalidated assessments. Several trials also bundle spacing with quizzing and feedback, so the benefit of spacing alone is hard to isolate (inference).

Counter-evidence and limits. The workplace cluster RCT had about 48% follow-up survey non-response (my calculation from 236 of 456 responding) and no massed comparator. The sales-training study found no advantage for spacing on transfer quantity. The blood-pressure trial's effect was modest (HR 1.043).

5. Evidence strength (my GRADE-style ratings)

These ratings are my own judgement from abstracts, not published GRADE ratings except where noted.

Outcome Rating Reason
Knowledge retention, spaced vs massed, adult professionals (mostly health care) Moderate Consistent direction across three syntheses; moderate certainty rated by Martinengo et al. for knowledge; downgraded for risk of bias and inconsistency (I2 66%)
Long-term retention at delayed follow-up Low Few trials test beyond weeks; Kerfoot 2010 is strong but author-built platform, single research group
Workplace behaviour change or training transfer Low Clinical behaviour pooled SMD 0.67, which the review authors rated moderate certainty, from only three spaced-online trials (my downgrade to low reflects indirectness to non-clinical firms); the only corporate study is quasi-experimental, n = 64
Surgical or procedural skills Low Rated low certainty by the review authors; two trials, wide CI
Non-health corporate onboarding (any outcome) Very low No RCT found in this search; indirect evidence only

6. Common measures and indicators

7. Gaps and future directions

8. Implications for corporate onboarding at mid-size firms

This is evidence interpretation, not a recommendation.

9. Reference list

Sources