Spaced Repetition

What it is

Spaced repetition is the practice of reviewing information at increasing intervals over time, instead of cramming it all at once. You study something today, see it again in a few days, then in a week or two, then in a month, and so on. Each successful recall buys you a longer interval before the next review. The technique is built directly on top of one of the oldest and most reliable findings in the science of memory: the forgetting curve.

In the 1880s, the German psychologist Hermann Ebbinghaus tested his own memory using lists of nonsense syllables and carefully tracked how quickly he forgot them. He found that forgetting is steep and fast at first, then levels off: you lose a large fraction of new material within hours and days, and what survives that early drop tends to stick longer. Crucially, he also noticed that relearning something took less effort than learning it the first time, and that spreading study sessions out across days produced better retention than packing them together. That second observation is the spacing effect, and spaced repetition is simply the deliberate, scheduled use of it.

The core idea is that memories are strengthened most when you retrieve them right as you are about to forget them. A review that comes too soon is wasted effort, because the answer is still fresh and the recall is too easy to do much work. A review that comes too late means you have already forgotten and have to relearn from scratch. Spaced repetition aims for the productive middle: review at the edge of forgetting, where pulling the answer back is effortful but still possible. Each such effortful retrieval flattens your personal forgetting curve a little more.

This chapter teaches you to do that by hand with a Leitner box and a written schedule, and then explains how software such as Anki automates the same logic. Spaced repetition does not make you smarter or raise your IQ. What it does is let you organize, encode, and recall a large body of facts far more efficiently than rereading ever could, with less total study time spread across more days.

When to use it

Use spaced repetition whenever you need to retain a large set of discrete facts in durable memory over weeks, months, or years rather than for a single afternoon. It shines for material that is item-based and clearly testable: a prompt on one side, an answer on the other. It is the wrong tool for understanding a difficult argument or developing a skill that requires practice rather than recall, though it pairs well with those once you have isolated the facts inside them.

• Learning foreign-language vocabulary and grammar, where hundreds or thousands of word pairs must stay accessible.
• Medical, legal, and other professional study where definitions, drug names, dosages, statutes, and classifications must be recalled precisely under exam pressure.
• Memorizing dates, capitals, formulas, anatomical terms, chemical symbols, or any reference facts.
• Keeping a speech, a set of customer names, or technical command syntax fresh over time.
• Studying for an exam that is weeks or months away, so reviews can be genuinely spaced rather than crammed the night before.

If your test is tomorrow, spacing has little room to operate and you should focus on active recall and sleep. Spaced repetition is the long game, and it rewards you most when you start early and review a little every day.

Step-by-step method

1. Turn the material into discrete question-and-answer items. Each card or entry should have one clear prompt and one clear answer. "What is the chemical symbol for potassium?" / "K" is good; a whole paragraph on potassium is not. Splitting big topics into atomic facts is the single most important preparation step.

1. Choose your system. For a physical method, build a Leitner box: a divided container or set of labelled envelopes with five compartments, numbered 1 through 5. For software, an app such as Anki applies the same expanding-interval logic automatically. Start with whichever you will actually keep up with daily.

1. Assign each box a review interval. A common schedule is Box 1 every day, Box 2 every 2 days, Box 3 every 4 or 5 days, Box 4 every 9 or 10 days, and Box 5 every three weeks or so. The exact numbers matter less than the principle that each box up the line is reviewed less often.

1. Put every new card into Box 1 to begin. Everything starts as daily review until it proves it can survive a gap.

1. Review by retrieval, not recognition. Read the prompt, say or write the answer from memory, then flip to check. The effortful pull is what builds the memory; peeking first wastes the rep.

1. Promote on success, demote on failure. If you recall a card correctly, move it up one box, so its next review is further away. If you miss it, send it all the way back to Box 1 to be drilled daily again. This is the engine that concentrates your time on exactly the items you find hard.

1. Review each box only on its scheduled day. Today you might do Box 1 and Box 3; tomorrow only Box 1; the day a Box 5 falls due, you add it in. Keep a simple calendar so nothing is skipped.

1. Keep cycling. Cards that survive all five boxes are essentially learned; you can retire them to an occasional refresh. New cards keep entering at Box 1, so the system runs indefinitely with a small, steady daily load.

A simple example

Suppose you are learning the chemical symbols of ten elements: hydrogen (H), helium (He), lithium (Li), sodium (Na), potassium (K), iron (Fe), gold (Au), silver (Ag), lead (Pb), and tin (Sn). The tricky ones, Na, K, Fe, Au, Ag, Pb, Sn, come from Latin names, so they resist memorization.

Make ten cards. On the front of each, write the element name; on the back, the symbol. On Monday, all ten cards sit in Box 1, and you review the whole box. You get H, He, Li, Na, K, Fe, and Au right but blank on silver (you guess Si), lead (you guess Le), and tin (you guess Ti). The seven you recalled move up to Box 2; the three you missed, Ag, Pb, Sn, stay in Box 1.

Your schedule is Box 1 daily, Box 2 every 2 days. On Tuesday, Box 2 is not due yet, so you review only Box 1: silver, lead, tin. You now remember silver is Ag and lead is Pb, but tin still escapes you. Ag and Pb move to Box 2; Sn stays in Box 1.

On Wednesday, both Box 1 (just Sn) and Box 2 (the eight that graduated) come due. You finally nail tin as Sn, so it joins Box 2. In Box 2 you recall all eight correctly, so they advance to Box 3, whose interval is every four days. Notice what happened: the easy symbols are already three boxes deep and will only resurface next week, while tin, the one you kept missing, got drilled three days running until it stuck. The system spent your attention precisely where it was needed, and within a week all ten elements are climbing toward Box 5.

An advanced example

Now scale up to a realistic exam case. You are preparing for a Spanish proficiency test ten weeks away and need to learn 600 vocabulary items plus 40 irregular verb conjugations. Cramming 640 cards is hopeless; spacing makes it routine.

Set a sustainable intake of roughly 15 new cards per day. With a five-box Leitner schedule (Box 1 daily, Box 2 every 2 days, Box 3 every 4 days, Box 4 every 8 days, Box 5 every 16 days), your daily workload stays small because most cards quickly climb out of the daily and every-other-day boxes. A worked schedule for one card, the verb "tener" (to have), might run: introduced Day 1 in Box 1; recalled Day 2, up to Box 2; recalled Day 4, up to Box 3; recalled Day 8, up to Box 4; recalled Day 16, up to Box 5; recalled Day 32, learned. Five successful reviews across a month, then it is effectively yours.

Contrast a stubborn card, the false friend "embarazada" (which means pregnant, not embarrassed). Day 1 Box 1; Day 2 you misremember it as "embarrassed," so it stays in Box 1; Day 3 wrong again, still Box 1; Day 4 correct, up to Box 2; Day 6 wrong, demoted back to Box 1. This card keeps surfacing almost daily until the correct meaning finally wins, which is exactly right: it is your weakest item and deserves the most exposure.

This is also where Anki's logic improves on rigid boxes. Instead of fixed box intervals, Anki computes a per-card interval and rates each answer Again, Hard, Good, or Easy. A "Good" answer multiplies the previous interval by an ease factor (around 2.5 by default), so intervals grow smoothly: 1 day, then 3, then 8, then about 20, then 50, then 120. "Again" collapses the interval and reschedules the card soon, like demotion to Box 1; "Easy" jumps the interval further ahead. The conceptual machinery is identical to the Leitner box, just continuous rather than five discrete steps. The research backing this approach is robust: Cepeda and colleagues, in a 2006 review of over 300 spacing experiments, found that spaced study reliably beat massed study, and that wider gaps favor longer-term retention, while Roediger and Karpicke showed in 2006 that the act of testing yourself, the retrieval each review demands, is itself a powerful memory builder.

Common mistakes

• Reviewing by rereading instead of retrieving. Looking at the front and back together feels productive but skips the effortful recall that actually strengthens memory. Always answer from memory first, then check. The struggle is the mechanism, not a bug.

• Spacing too tightly so it becomes cramming in disguise. If you review every card every day, you never let the productive forgetting happen and you waste time on items you already know. Trust the expanding intervals and let easy cards drift away.

• Making cards too big. A card asking you to "explain the causes of World War I" cannot be cleanly scored as right or wrong, so it breaks promotion and demotion. Split fat cards into atomic one-fact prompts that have a single correct answer.

• Letting the backlog explode. If you skip several days, due cards pile up and the session becomes punishing, which makes you skip more. Keep a small, steady daily intake, and if you fall behind, prioritize the lowest boxes first rather than trying to clear everything at once.

• Not demoting honestly. If you half-remember a card and still promote it, you smuggle weak items into long intervals where they quietly rot. Be strict: a hesitant or partial answer counts as a miss and goes back to Box 1.

• Starting too late before an exam. Spacing needs calendar room. Beginning two days out gives the intervals nowhere to expand, so you lose the technique's main advantage and are effectively back to cramming.

Related techniques

Spaced repetition is the scheduling engine that makes every other technique in this book durable. It depends on active-recall, since each review must be a real retrieval attempt, not a reread; the two are best understood as partners, and the chapter on active recall explains why the effortful pull works. The cards you space are far stronger when their answers are encoded with the method-of-loci memory palace, the peg system, the major-system for numbers, or the story method, so that even a hard card has a vivid hook to grab. Chunking helps you keep each card atomic and within the limits described in how-memory-works. When you reach studying-for-exams, you will see spaced repetition as the backbone of a multi-week study plan, and remembering-vocabulary and remembering-numbers both lean on it directly to move material into long-term memory.