Written by Dr. Roupen Odabashian MD, FRCPC, FASC
Hematologist-Oncologist | Founder, MeDucation AI | Updated November 2026
The short answer: yes, spaced repetition is worth building into exam prep, and the evidence for it is unusually strong for a study technique. A review covering 839 assessments across 317 experiments and 184 articles found that spacing out study sessions beats massing them into one sitting, with the ideal gap between sessions actually growing as the time until your exam grows[1]. This is not a new idea — it traces back to Hermann Ebbinghaus's 1885 work on memory and forgetting[2] — but it is one of the few study habits with a century-plus of replication behind it. Before going further: I am a practicing hematologist-oncologist and the founder of MeDucation AI, which builds spaced-repetition flashcards into its Notebook and Learning Hub tools for fellows and students. I have a direct financial interest in more people valuing spaced repetition. I have tried to write this piece so that it holds up even if you never touch our product — including recommending Anki, a free, non-MeDucation tool, in several places below.
Rank | Approach | Best for | Where it falls short |
|---|---|---|---|
1 | Spaced repetition flashcards (FSRS or SM-2 algorithm) | Long-haul retention for board exams, in-service exams, anything cumulative | Requires upfront deck-building or a tool that generates cards for you |
2 | Massed review ("cramming") close to exam day | Short-term recognition on a test happening the next day | Poorer durability over time compared with spaced study[2] |
3 | Re-reading notes or slides passively | Initial exposure to new material | Weakest evidence of the three for actual retention over time |
What is spaced repetition, exactly?
Spaced repetition is reviewing material at increasing intervals instead of all at once, timed so each review happens just as you're about to forget. The underlying idea, the spacing effect, shows that more information gets encoded into long-term memory through spaced study sessions than through cramming[2]. Ebbinghaus's original 1885 work argued that active recall with increasing time intervals reduces the probability of forgetting information[2], and that basic finding has since been replicated across free recall, recognition, cued-recall, and frequency-estimation tasks[2].
The practical mechanism usually used today is the flashcard: a question on one side, the answer on the other, reviewed on a schedule that an algorithm adjusts based on how well you remembered it last time[3]. Cards you get wrong come back soon; cards you get right drift further out. This is sometimes combined with the testing effect — the act of retrieving an answer, not just re-reading it — which further strengthens long-term memory[3].
Does the research actually back this up, or is it folklore?
It backs it up more thoroughly than almost any other study technique in the literature. The Cepeda, Pashler, Vul, Wixted, and Rohrer review is a large-scale synthesis: 839 assessments in 317 experiments located in 184 articles, examining spacing and lag effects[1]. A separate tally found that in a related analysis, participants using spaced practice outperformed those using massed practice in 259 out of 271 cases[2].
One detail that surprises people who assume "more spacing is always better": the ideal interval depends on how long you need to remember the material. The Cepeda review found that the inter-study interval producing maximal retention increased as the retention interval increased[1] — in other words, a study plan that needs the material fresh for a two-week in-service exam should be paced differently than one aimed at a board exam eight months out.
Study detail | Finding |
|---|---|
Comparisons in Cepeda review | 839 assessments across 317 experiments and 184 articles[1] |
Spaced vs. massed practice outcomes (separate tally) | 259 out of 271 favored spaced practice[2] |
Foreign-vocabulary recall interval study | 13 sessions spaced 56 days apart matched the retention of 26 sessions spaced 14 days apart[2] |
There's also a related finding worth knowing for board prep specifically: cumulative final exams, which force students to retrieve older material repeatedly rather than studying each unit once and moving on, produce better long-term retention — and the literature calls this evidence "conclusive"[2]. That maps directly onto board exams, which test everything you learned over a multi-year fellowship at once.
How is this different from just rereading my notes a few times?
What tool should I actually use — Anki, or something else?
Anki historically used the SM-2 algorithm from 1980s SuperMemo research, but newer versions (23.10 and later) include a native implementation of FSRS — the Free Spaced Repetition Scheduler — whose default parameters are more accurate than the standard SM-2 algorithm according to benchmarks, leading to fewer necessary reviews for the same retention rate[4]. The American Medical Association describes Anki's core loop plainly: you rate your own recall after each card, and the algorithm pushes difficult or missed cards back sooner while easier ones get pushed further out, focusing your time on your actual weak spots[5]. Anki also supports embedded images, audio, video, and even scientific notation via LaTeX, and some residents in fields like radiology, dermatology, and urology have built specialty-specific decks for in-service and licensing exam prep[5].
This is genuinely a case where I'll point you to a competitor: if all you want is a flashcard engine and you're willing to build or download a deck yourself, Anki is free, proven, and more than capable[6]. Where MeDucation's Notebook differs isn't the underlying spaced-repetition math — our flashcards run on the same FSRS algorithm Anki uses — it's that cards are generated automatically from a fellow's own notes rather than requiring manual deck-building, and they sit alongside auto-regenerating mind maps and an "Ask My Notes" chat grounded in those same notes, with decks shareable org-wide or by email and independent FSRS schedules tracked per recipient.
How far out from boards should I start, and how should I pace it?
Start earlier than you think, because the evidence suggests the ideal gap between reviews grows as your retention window grows[1]. Studies have shown that long-term spacing effects are most pronounced when the gaps between study sessions are on the order of days or weeks[2], and spaced practice has been shown to produce significant improvement over massed practice when tested after a delay[2] — meaning spacing's advantage often doesn't even show up if you only test yourself immediately after studying, which is exactly the trap cramming exploits.
There's a second lesson buried in the same literature that board-exam candidates routinely ignore: the same research that supports spacing in time also supports mixing content types within a session rather than grouping by topic. In one experiment, practice problems randomly mixed by type produced vastly superior performance compared to problems grouped by type, despite (or because of) feeling harder in the moment — a "desirable difficulty"[2]. If you're building a study plan around a six-month runway to the ABIM hematology and oncology boards, this argues for interleaving disease topics across weeks rather than finishing leukemia before starting lymphoma.
Does spaced repetition replace question banks, or work alongside them?
It works alongside them, not instead of them — the two are testing different things. Spaced-repetition flashcards are built for durable recall of discrete facts (a cytogenetic abnormality, a drug's mechanism, a staging cutoff), while board-style question banks test applied clinical reasoning across a full vignette. Our own question bank is physician-reviewed by board-certified hematologist-oncologists, with explanations covering why the correct answer is right and why each distractor is wrong, embedded images and tables, and direct PubMed links — that's a different job than a flashcard, and fellows generally need both. If you're deciding how to split limited study hours between the two, how to pair recall tools with applied-reasoning tools is worth thinking through deliberately rather than defaulting to whichever feels more comfortable.
Is spacing worth it for something as short as a two-week in-service exam cram?
Marginally, but it's not where spacing earns its keep. The core finding is that inter-study interval and retention interval operate jointly — the gap that maximizes retention scales with how long you need to remember the material[1]. For a two-week runway, you simply don't have room for the long gaps that produce spacing's biggest gains; in a vocabulary-recall study, 13 sessions spaced 56 days apart matched the retention of 26 sessions spaced only 14 days apart[2], which tells you the effect compounds over months, not days. If your in-service exam is two weeks out, dense near-term review still beats pure rereading, but don't expect the same payoff you'd get from a flashcard habit that started months earlier.
FAQ
Does spaced repetition work for clinical reasoning, or only for rote facts?
The strongest, most direct evidence is for explicit memory tasks — free recall, recognition, cued-recall[2] — which maps best onto facts, mechanisms, and classifications rather than multi-step clinical judgment. A 2015 analysis did find a positive relationship between Anki card volume and USMLE Step 1 performance[4], a test that blends factual recall with applied reasoning, which suggests spillover benefit, but flashcards are still best understood as the recall layer underneath your clinical reasoning, not a replacement for working through full vignettes.
How much does Anki cost compared to building flashcards inside MeDucation?
Check Anki's site directly for current pricing across platforms. For what MeDucation's Notebook and Learning Hub flashcard tools cost as part of a subscription, see our pricing breakdown.
