Spaced repetition vs massed practice: which one actually works?

Spaced repetition spreads study sessions over days or weeks. Massed practice (cramming) packs them back-to-back in a single sitting. For anything you need to remember beyond tomorrow, spacing wins — meta-analyses show effect sizes from d = 0.46 (Donovan & Radosevich, 1999) to g = 1.15 for vocabulary on delayed tests (Kim & Webb, 2022). Cramming only beats spacing when the test is hours away.

What is the difference between spaced repetition and massed practice?

Massed practice is the default study mode almost every student uses without thinking about it: read the chapter twice, highlight, reread the night before the exam. All the contact with the material is concentrated in one block.

Spaced repetition does the opposite. The same total study time gets distributed across multiple sessions separated by gaps — hours, days, or weeks. Each session is shorter, but the material is revisited just before you would otherwise forget it. The technique was first formalized by Hermann Ebbinghaus in 1885 when he plotted his own forgetting curve, and it has been refined into modern flashcard algorithms like SM-2 and FSRS.

The two approaches feel very different. Massed practice feels productive — you "get a lot done" in one sitting and the material feels familiar by the end. Spaced repetition feels harder, because every session begins with material that has partially decayed and has to be retrieved from a colder memory. That difficulty is not a bug. It is the mechanism.

Which one wins for long-term retention?

The evidence on this is not subtle. Cepeda, Pashler, Vul, Wixted, and Rohrer (2006) published the largest review of distributed practice ever attempted in Psychological Bulletin — 839 assessments from 317 experiments across 184 articles. Spaced practice outperformed massed practice in nearly every comparison.

The numbers across newer meta-analyses are remarkably consistent:

Donovan & Radosevich (1999), 63 studies: d = 0.46 favoring spacing (Journal of Applied Psychology)
Latimier, Peyre, & Ramus (2021), 29 studies: g = 0.74 for spaced retrieval (Educational Psychology Review)
Kim & Webb (2022), 48 experiments with 3,411 learners: g = 1.15 for vocabulary on delayed tests (Language Learning)
Mawson & Kang (2025), 22 reports across 3,000+ classroom students: d = 0.54 (Behavioral Sciences)

For perspective, an effect size of d = 0.5 is what educational researchers consider a "moderate to large" intervention — comparable to the difference between an average teacher and a top-decile one. Spacing produces this kind of advantage for free, without changing what you study or how long.

When does cramming actually beat spaced practice?

Cramming has one situation where it genuinely wins: an exam in the next few hours.

Roediger and Karpicke (2006) demonstrated this directly. Students who massed their studying outperformed students who spaced it on a test taken 5 minutes after the final session. But on a test taken 1 week later, the spaced group recalled roughly 50% more material (Psychological Science, 17(3)). Rawson and Kintsch (2005) showed the same pattern with text learning — massed practice won on same-day tests; spaced practice won at every longer interval (Journal of Educational Psychology, 97(1)).

Robert Bjork's framework of "learning versus performance" captures the trap (Soderstrom & Bjork, 2015, Perspectives on Psychological Science). Cramming maximizes performance in the moment — what you can demonstrate right now. Spaced repetition maximizes learning — what is durably encoded and retrievable later. The two are often mistaken for each other because cramming feels productive while it is happening. The fluency illusion that follows a long study block is real, but it usually does not survive a week's sleep.

So the honest rule: if your only goal is a quiz in two hours and you will never need the material again, cram. For everything else — board exams, languages, professional certifications, anything you actually want to know — spacing is the dominant strategy.

Why does spacing work?

Three mechanisms, each independently supported, combine to produce the spacing effect.

Retrieval strength rebuilds memories. Each time you re-encounter material that has partially faded, your brain has to reconstruct it from cues. That reconstruction strengthens the memory trace more than the original encoding did. This is why active recall compounds with spacing — the gap forces the retrieval to be effortful, and effortful retrieval is what produces durable memory.

Encoding variability. When you study the same material in different sessions, you are also encoding it in slightly different mental contexts — different moods, locations, times of day. Each new context becomes a retrieval cue, and a memory with many cues is easier to retrieve later than a memory with one. Massed practice gives you one rich context. Spaced practice gives you several thinner ones, which together work better.

Consolidation. Memory is not stable the moment you encode it. It restabilizes over hours and days, partly during sleep. Spacing study sessions across multiple sleep cycles gives your brain repeated opportunities to consolidate, while massed practice tries to do all the encoding before any consolidation has occurred.

Kornell and Bjork (2008) demonstrated something especially counterintuitive: even for inductive learning — extracting general patterns from examples, the kind of task you would assume benefits from concentration — interleaved, spaced exposure beat massed exposure (Psychological Science, 19(6)). The benefit is not limited to rote memorization.

How should you actually choose between them?

Cepeda, Vul, Rohrer, Wixted, and Pashler (2008) analyzed retention in 1,350+ participants across many gap lengths and produced a usable rule: the optimal gap between study sessions is roughly 10–20% of how long you need to remember the material (Psychological Science, 19(11)).

A practical decoder:

| Retention goal | Suggested gap | |---|---| | Test in 1 week | Review 1 day later | | Test in 1 month | Review 3–7 days later | | Test in 6 months | Review 3–4 weeks later | | Test in 5 years | Review every 6–12 months |

If you have time before the exam, never mass. If you do not, accept that you are optimizing for performance, not learning. The Bahrick family study (1993) showed that just 13 sessions at 56-day spacing produced the same retention as 26 sessions at 14-day spacing (Psychological Science, 4(5)) — when you space correctly, you can study less and remember more.

How does Memor More handle the spaced repetition vs cramming tradeoff?

Memor More schedules every card based on your past performance, so each review lands close to the optimal gap rather than wherever your motivation happens to peak. Cards you struggled with come back sooner. Cards you know cold get pushed weeks or months out — exactly the pattern the Cepeda 10–20% rule recommends.

There is no cram mode by design. If your exam is in two hours, any flashcard app will do. For anything beyond that, the schedule is the product, and it is built around the research above.