The Science of Spelling: How Spaced Repetition Boosts Memory

Weekly spelling lists have a poor track record. A child crams on Thursday night, scores well on Friday, and by the next Wednesday cannot spell half the words. Adults are no better: most of us have looked up "accommodate" a dozen times over the years, and it has never quite stayed.

The usual explanation is that some people are just bad at spelling. The better explanation is about timing. Traditional practice is built on cramming, which packs all the effort into one or two sittings. Human memory, by contrast, is built to hold on to things it meets repeatedly, with gaps in between. Spaced repetition is the practical method that follows from that fact, and it rests on well over a century of research.

This post explains what that research found, why it applies with particular force to spelling, and how to use it with nothing more than a pencil and a few minutes a day.

Why we forget words we have just learned

In the 1880s, the German psychologist Hermann Ebbinghaus spent months memorizing lists of meaningless syllables and testing himself at intervals to see how much survived. He found that forgetting is fast at first and then slows: a large share of a new list was gone within a day, and what remained faded far more gently afterwards. That shape is the forgetting curve. In 2015, Jaap Murre and Joeri Dros repeated his experiment with modern methods and got a strikingly similar result.

Spelling follows the same curve. A word learned on Monday is fragile. If nothing brings it back, the fragile version is what you have when you need it next month, which is often nothing at all.

None of this is a character flaw. A memory system that kept everything would drown in detail. Your brain is making a sensible bet that information you met once and never again was not important. Our post on the forgetting curve goes into this in more depth.

What the spacing effect is

The spacing effect is the finding that practice spread across several sessions produces much more durable memories than the same amount of practice in one session. It is among the most replicated results in psychology.

In 2006, Nicholas Cepeda, Harold Pashler and colleagues reviewed hundreds of experiments on distributed practice and found the advantage for spacing held across materials, ages and tests. An earlier meta-analysis by John Donovan and David Radosevich, published in 1999 and covering 63 studies, found a solid, medium-sized benefit. And in one of the most striking studies of all, Harry Bahrick and his family learned foreign-language vocabulary over several years in the 1980s and 1990s. They found that practice sessions spaced about two months apart produced better recall years later than sessions two weeks apart, even though the total practice was the same.

The effect holds for learning another language too. Joseph Kim and Stuart Webb combined 48 experiments in a 2022 meta-analysis and found spaced practice clearly beat massed practice, and longer gaps beat shorter ones on delayed tests.

That last detail matters. Short gaps win if the test is tomorrow. Long gaps win if the test is next year. Life is the test next year.

How a well-timed review changes the curve

Each time you successfully recall a word after a gap, two things happen. The memory is brought back to full strength, and it fades more slowly afterwards than it did before. The curve flattens. After three or four well-spaced reviews, a word that would have been gone in a week can hold for months.

This explains the characteristic shape of a good schedule: short gaps at first, then longer ones. Review a new word the next day, then after a few days, then a week, then a few weeks, then a few months. Every successful recall earns a longer rest.

There is a sweet spot in the timing. Review too soon, while the word is still fresh, and the recall is so easy it does little. Wait too long and the word is gone, so you are learning it again from scratch. The best moment is when recall takes a bit of effort but still succeeds. Robert Bjork calls this kind of productive struggle a desirable difficulty: effort that feels harder in the moment and produces stronger learning.

Why spacing works so well for spelling

Spelling is an unusually demanding kind of memory. To recognize a word on a page, you need only a rough impression: the first letter, the length, a familiar chunk. To spell it, you need every letter in the right order, pulled from nothing.

The reading researcher Linnea Ehri describes how words become spellable through orthographic mapping, the process of binding each sound in a word to the letters that spell it and attaching the whole to its meaning. Spelling is not a visual snapshot. It is a set of sound-to-letter connections, and like any connection it strengthens each time it is used and weakens when it is not.

Spacing helps in three specific ways. It strengthens the exact letter sequence, including the one tricky spot (the double m, the silent h) where errors happen. It sharpens the difference between look-alike pairs such as lose and loose, because each recall after a gap forces a real choice. And it makes spellings more robust under pressure, in an exam hall or at the end of a tiring day, because a memory recalled many times over weeks is easier to reach.

Spacing plus recall: the combination that does the work

Spacing decides when you practice. The second ingredient decides how.

Retrieval practice, also called the testing effect, is the finding that pulling information out of memory strengthens it more than taking it in again. Henry Roediger and Jeffrey Karpicke showed this in 2006 with students learning prose passages: rereading did slightly better on a test five minutes later, but practicing recall did much better a week later. John Dunlosky and colleagues reviewed ten popular study techniques in 2013 and rated practice testing and spaced practice as the two most effective.

For spelling, this has been tested directly. Eric Jones and colleagues found that children who practiced by recalling spellings remembered them better than children who traced the same words in colored "rainbow writing". In a 2023 study with fifth graders in Brazil, spelling from memory beat copying four days later, but only when the children saw the correct spelling straight after each attempt.

So the recipe is: hear or think of the word, spell it from memory, check it at once, and come back to it on a widening schedule. Copying, highlighting and staring at lists feel like study. They leave little behind. Our post on the testing effect covers this half of the recipe in detail.

What happens in the brain, briefly

The mechanisms are still being studied, so it is worth being careful here.

One well-supported idea is consolidation: a new memory is fragile, and over the hours and days after learning, especially during sleep, the brain stabilizes and reorganizes it. Spaced sessions give that process several chances to work. Our post on sleep and learning to spell explains how that plays out.

Brain-imaging work adds another piece. A 2019 study by Feng and colleagues in the Journal of Neuroscience found that spaced learning led the brain to reproduce its original pattern of activity more faithfully when an item was studied again. That faithful replay predicted better memory later. The honest summary is that researchers know spacing works very reliably, and are still filling in exactly why.

A spaced-repetition week, with paper and pencil

You do not need software to start. You need a small set of words and three piles or columns: today, in a few days, and later.

1. Start small. Take eight to twelve new words, or five to eight for a young child.
2. Recall from the first session. Say each word, spell it from memory, check it immediately, and fix any mistake by trying again from memory.
3. Sort honestly. Words spelled right without hesitation move to "in a few days". Words you missed or hesitated over stay in "today" and come back tomorrow.
4. Stretch the gaps. Each time a word is right after a gap, move it further out: a week, then a few weeks, then a couple of months.
5. Mix old and new. Every session starts with the words that are due, then adds a few new ones.
6. Keep sessions short and frequent. Five to fifteen minutes on most days works far better than an hour on Sunday.

The most common mistake is letting words drop off the end. A school list ends on Friday, but the words should stay in the "later" pile for months, returning less and less often. That is what turns a passed test into a word you actually own.

If you are wondering how many new words to add each day without drowning in reviews, our post on the honest math of spaced repetition works it through.

Common myths about spaced repetition

"It is just flashcards." Flashcards are one format. The core is timed recall. Dictation, spelling aloud, or typing a word you hear all count, as long as the word is produced from memory at the right time.

"More often is always better." Reviewing a word every five minutes for an hour is massed practice wearing a disguise. It feels productive and fades quickly. The gaps are the point.

"Children will not put up with it." Most children put up with five minutes a day far more happily than a forty-five-minute cram. Seeing a word "graduate" to a longer gap is motivating in its own right.

"If I got it right once, I know it." One correct answer on the day you learned a word shows only that it is in short-term memory. A correct answer after a gap of a week is far better evidence.

How software schedules the reviews

Keeping track of dozens of words, each on its own timer, is tedious by hand. That bookkeeping is what spaced-repetition software is for.

The idea goes back to the Polish researcher Piotr Wozniak, whose SuperMemo program in the late 1980s used an algorithm called SM-2 to set the next review from how well you answered. The free flashcard app Anki adapted SM-2 and made it widely known. More recently, an open-source algorithm called FSRS, developed from 2022 onward, models each memory's stability and predicts when the chance of recalling it will drop to a set level; Anki added it as an option in 2023. Our post on how a scheduling algorithm decides when you see a word again explains how these models work.

Spelling.School uses FSRS for exactly this job: you hear a word, type it, see which letters went wrong, and the word is scheduled to return just before you would be likely to forget it. The method is what matters, though, and paper works too.

Putting it together

Strip away the vocabulary and the science comes down to three habits. Spell words from memory rather than looking at them. Check each attempt straight away. Come back to each word after gaps that grow longer every time you get it right.

Start this week with ten words you or your child genuinely need, three piles, and ten minutes a day. If you want the scheduling done for you, you can try five words on Spelling.School without signing up. Either way, the words that come back are the words that stay.

Questions people ask

What is spaced repetition in simple terms?

Spaced repetition means reviewing something you are learning several times, with gaps between reviews that grow longer each time you remember it correctly. For spelling, you might review a new word the next day, then after a few days, then a week, then a month. Each successful review makes the memory stronger and slower to fade. It works far better than practicing the same word many times in one sitting.

How often should you review spelling words?

A good starting schedule is the day after learning, then two or three days later, then a week, then a few weeks, then a couple of months. If you get a word wrong, bring it back the next day and start the gaps again. Short sessions of five to fifteen minutes on most days are better than one long session a week. What matters most is that words keep returning after the test is over.

Is spaced repetition better than cramming?

For anything you want to remember beyond the next day or two, yes. Cramming can produce a good score on a test the following morning, but the memory fades quickly afterwards. Reviews of hundreds of experiments have found that spreading the same amount of practice over several sessions produces much more durable learning. Cramming feels efficient because the material is fresh, which is exactly why it misleads.

Does spaced repetition work for spelling specifically?

Yes. Spelling depends on remembering an exact sequence of letters, which is the kind of detailed memory that fades quickly without review. Studies with children have found that practicing recall of spellings leads to better results days later than copying or tracing the words, especially when each attempt is checked right away. Combining that recall with spaced reviews is the most reliable way to make spellings last.

What is the best spaced repetition algorithm?

Several work well. SM-2, created by Piotr Wozniak for SuperMemo and later adapted by Anki, has been used for decades. FSRS, a newer open-source algorithm, models how stable each memory is and schedules reviews so that recall stays likely, and it usually needs fewer reviews to reach the same retention. For most learners the choice of algorithm matters less than reviewing consistently and recalling from memory rather than rereading.

Can you do spaced repetition without an app?

Yes. Use three piles of index cards or three columns on a page labeled today, in a few days, and later. Test each word from memory, move words you spell correctly to a longer gap, and keep missed words in the today pile. An app mainly saves you the bookkeeping of tracking when each word is due, which becomes tedious once you have more than a few dozen words.

Fix the words you actually get wrong

A few minutes a day, spaced repetition, built only for spelling. Try five words without signing up.