Spaced repetition: why spreading study over time beats cramming

Spaced practice is repeating the same material across separate sessions rather than in one long block, and it produces more durable memory than cramming. The finding is one of the most reliable in learning research: it has been confirmed for facts, concepts and lists, for motor and classroom skills, in science and vocabulary teaching, and across species from bees to primates. The practical version is spaced repetition, where the point of each repetition is not to reread the material but to retrieve it, at a gap long enough that the retrieval takes effort.

• Spaced repetition means studying the same material in sessions separated by gaps. Massed practice, or cramming, means studying it in one long block with no gaps between the repetitions.
• The spacing effect is the underlying finding that spaced practice beats massed practice. It holds for facts, concepts and lists, for skills, and in classroom teaching.
• Ebbinghaus documented it in 1885, and put it more precisely than most summaries do: a suitable distribution of repetitions over time is decidedly more advantageous than massing them at a single time.
• The advantage comes from how memory is built, not from the schedule itself. Spaced repetitions gather more retrieval cues, force you to retrieve the trace from the previous repetition, and leave time for consolidation.
• There is no single correct interval. The useful gap depends on how long you need to remember the material, and the evidence does not show that the gaps have to grow.
• For psychology terms, the practical version is straightforward: collect the terms into a list, retrieve each one instead of rereading it, and come back to the list on later days.

What Ebbinghaus actually found

The spacing advantage has been part of memory research since its beginning. Hermann Ebbinghaus, learning lists of nonsense syllables and testing how much faster he could relearn them, found that spreading repetitions out beat concentrating them, and summed it up himself:

It is worth being precise about what he measured. Ebbinghaus tracked savings: how much relearning time each repetition saved, not how much he could recall. Popular retellings of his forgetting curve routinely turn those savings figures into recall percentages he never reported. Our post on the forgetting curve sets out what he actually did and what the numbers do and do not mean.

The idea of spacing revision at deliberately growing intervals is usually traced to C. A. Mace's 1932 book The Psychology of Study, which suggested intervals of roughly one day, two, four and eight days. By 1939, H. F. Spitzer was testing spaced review in school classrooms. The finding has been reproduced many times since, and a 2016 review in Nature Reviews Neuroscience gathered the evidence across learning tasks, ages and species.

Why spacing beats cramming

Learning research has proposed three main explanations for the spacing advantage, and they are not mutually exclusive. A 2016 review by Paul Smolen, Yili Zhang and John Byrne set them out together and concluded that the cellular evidence lines up most closely with the third.

Spaced repetitions pick up more context

Encoding variability theory holds that repetitions spread over time are more likely to happen in different contexts: different rooms, moods and times of day. A memory formed across several contexts is attached to more of the cues that might later trigger it, which makes it easier to retrieve. Massed repetitions all happen in one context, so the memory ends up tied to fewer cues.

Each spaced repetition retrieves the previous one

Study-phase retrieval theory points at what happens inside the learner. When repetitions are far enough apart, each one makes you retrieve and reactivate the memory left by the one before it, and retrieving a memory strengthens it. When repetitions are massed, the previous trace is still active, so there is nothing to retrieve and nothing to reinforce. This is also why the gap cannot be arbitrarily long: past some point the earlier trace can no longer be retrieved at all, and the repetition stops reinforcing it.

Massed repetition leaves no time to consolidate

Deficient-processing theory says that massed practice fails because some of the processes needed to stabilise a memory never get a chance to run. A memory is not fixed the moment you learn it; it is consolidated over the hours afterwards, and that consolidation takes time and, at the cellular level, new gene transcription and protein synthesis. Massed repetitions arrive too close together to allow separate rounds of this, while spaced ones leave room for each repetition to be stabilised before the next.

The three accounts are not rivals to choose between. The 2016 review treats the consolidation account as the best supported and notes that the other two would need network-level data rather than single-neuron data to be tested fully. The phenomenon is solid; the explanation is still being worked out.

Does the schedule matter? Expanding, uniform, and irregular

If spacing works, the natural next question is whether the gaps should stay the same or grow. That is the difference between a uniform schedule, where the gap is the same each time, and an expanding one, where the gaps get longer.

The evidence does not pick a clear winner. Karpicke and Roediger found in 2007 that expanding the gaps helped retention over short delays, but that equally spaced retrieval was better over long ones. A meta-analysis by Latimier and colleagues concluded that spacing out retrieval practice episodes improves retention, without a reliable advantage for one shape of schedule over another, and a 2010 study by Karpicke and Roediger asked directly whether expanding retrieval is superior for learning from text and came away without a general yes. What does seem to matter is absolute spacing: Karpicke and Bauernschmidt found that how far apart the later repetitions sit matters more than how the gaps are distributed relative to one another.

The 2016 review reaches the same cautious place: which schedule is optimal was still an open question, and its own computational modelling suggested that irregular intervals could, in principle, beat regular ones. Spaced repetition is the method; the scheduling algorithms inside tools such as Anki and SuperMemo are one implementation of it, not the idea itself.

On the current evidence, the shape of the schedule is a detail. The gap between spacing at all and cramming is far larger than the gap between one spacing schedule and another.

How to use spaced repetition for psychology terms

Terms and definitions are close to ideal material for spacing. Each pair is a small, testable fact, and a definition you can only recognise rather than produce is exactly what an exam will expose. Here is the practical version.

1. 1.Collect the terms into one list rather than learning them chapter by chapter. An exam does not tell you which chapter a term came from, and one list is easier to return to.
2. 2.Make every pass a retrieval, not a reread. Cover the definition, produce it, then check. If you can see the answer while you are studying, you are not testing yourself.
3. 3.Spread the passes over days. The gap that works depends on how far away the exam is; for a test weeks away, sessions a few days apart are the right order of magnitude. Our guide to memorising psychology terms works through how to choose the gap in more detail.
4. 4.Mix terms from different topics once you know them. Working out which concept a question is asking about is part of the exam task, so it is worth practising.
5. 5.Keep the feedback. Reviews of retrieval practice make the point that feedback is part of what makes the method work: seeing the correct answer after an error helps you get it right next time.

There is no need to copy a rigid interval schedule from an app or a blog post. The research gives a direction, not a single timetable: space it out, retrieve it, come back later, because the best gap depends on how long you need to keep the material.

That direction is the loop Psychology World is built around. You save the terms you are studying into a list, then drill them in three modes (flashcards, multiple-choice questions and typing the answer out), and return to that list on later days. Typing is the strictest of the three, because it will not accept a definition you merely recognise. The four steps of the learning loop and what each study mode does are both set out on the home page.