The Forgetting Curve and Why Spaced Repetition Works

In 1879, a young German psychologist named Hermann Ebbinghaus decided to do something slightly insane. He invented a list of more than two thousand nonsense syllables (think ZIB, KAJ, RUT) so that no prior associations could help him remember any of them. Then he memorized batches of them. Then he tested himself at carefully measured intervals to see how much remained. He was both the experimenter and the entire subject pool. The work took years. He published the results in 1885 in a book called On Memory, and the curve he drew in those pages is one of the most quietly important findings in the history of psychology. It has been replicated, refined, and extended for one hundred and forty years, and the basic shape has held.

Here is the shape. After you learn something, you forget most of it within the first twenty-four hours. You forget a smaller portion of what is left over the next few days. What survives the first week tends to survive much longer. The curve is steep at the start and shallow at the end. The implication is that the moments just after you encounter something matter much more than the moments much later. If you can intercept the forgetting at the right point, you can keep the memory alive almost indefinitely with surprisingly little effort.

The spacing effect, refined

The most important follow-up to Ebbinghaus was the recognition that when you revisit material matters more than how often you revisit it. This is called the spacing effect. It says that two encounters spaced a day apart will produce more durable memory than two encounters spaced a minute apart, even though the total study time is the same. The result is one of the most robust findings in cognitive psychology, replicated across hundreds of studies in dozens of domains.

In 2008, Nicholas Cepeda and colleagues published a meta-analysis in Psychological Bulletin that synthesized decades of work and gave us a usable rule of thumb: if you want to remember something for a long time, the optimal gap between review sessions is roughly ten to twenty percent of the time you want the memory to last. To remember a word for a week, review it after a day. To remember it for a year, review it after a month or two. Beyond that, the curve gets shallow and the math becomes forgiving. What matters most is that the gaps grow.

Mental events that follow each other in time become so connected that one of them, when it reappears, calls up the others.

Why cramming feels effective and is not

There is a reason students cram and pay the price later. Cramming produces a strong feeling of fluency in the moment. The material feels available, accessible, easy. The problem is that the feeling of fluency is a terrible predictor of long-term retention. Robert and Elizabeth Bjork at UCLA spent decades studying this and named the principle desirable difficulty: techniques that feel harder in the moment, including spacing, often produce better learning over time. Techniques that feel easier in the moment, including massed practice and rereading, often produce worse learning over time. Your sense of how well something is going in is almost always biased toward the cramming condition.

This is a hard truth and it explains a great deal about why most attempts to learn things fail. We optimize for the wrong signal. Ten minutes a day, spread over a month, will produce a much more durable result than five hours in one weekend afternoon. Almost everyone does the weekend afternoon. The reason is that it feels like progress while you are doing it.

From the lab to your phone

For most of the twentieth century the spacing effect was an academic curiosity. Then in the 1970s and 80s, a small group of researchers and developers, including Sebastian Leitner with his flashcard box system and Piotr Wozniak with his early version of SuperMemo, began to translate the laboratory finding into practical tools. Their core insight was simple: if the optimal review interval grows with the strength of the memory, then a good study tool should track each item separately, present it more often when it is fresh and less often as it becomes durable, and schedule reviews to land just before the predicted moment of forgetting.

This is the architecture behind every modern spaced-repetition system, from Anki and Quizlet to language apps like Duolingo and the medical-school standby SuperMemo. Each item carries a state that updates each time you see it. The schedule is personalized. Items you find hard come back faster. Items you find easy drift further apart. Over months and years, the algorithm learns the rough shape of your own forgetting curve and adapts to it.

It is not just for vocabulary

The earliest spaced repetition systems were aimed at language learners, where the value is most obvious. The scope of the effect, however, is much broader. Cepeda and colleagues documented spacing effects for trivia, anatomy, statistics, music theory, motor skills, and even procedural reasoning. The effect appears in young children and in older adults. It appears across cultures. It appears for material that is meaningful and for material that is not.

The implication is that almost anything you want to remember well, you should encounter on a spaced schedule rather than all at once. A poem you love. A passage from a book that changed how you think. The names of your second cousin's children. The verses of scripture that mean something to you. A piece of advice you keep meaning to live by. None of those need a textbook. They need a rhythm.

The cost of effortful retrieval, and why it is good for you

There is one more piece of the picture worth knowing. The act of pulling a memory out of your head, also called retrieval, is itself a powerful strengthener of the memory. Henry Roediger and Jeffrey Karpicke published a famous 2006 paper in Psychological Science showing that students who tested themselves on material remembered it dramatically better one week later than students who simply restudied it, even when restudy time was longer. The finding is now called the testing effect, and it travels well with spacing. The combination, called retrieval practice on a spaced schedule, is the single most evidence-backed study technique known.

The trick to making this work outside a classroom is to find low-friction ways to be re-exposed to material. A flashcard works. A wall calendar with a daily verse works. A frame that quietly rotates through lines you have loved over the past year works. None of these require effort. The exposure is the active ingredient.

If you want to read more

Ebbinghaus, On Memory, 1885, is now in the public domain and is shorter than you would expect. Cepeda, Pashler, Vul, Wixted, and Rohrer, Spacing Effects in Learning, Psychological Science, 2008, is the cleanest modern synthesis. Roediger and Karpicke, Test-Enhanced Learning, Psychological Science, 2006, is the foundational testing-effect paper. Bjork and Bjork, Making Things Hard on Yourself, but in a Good Way, is the best one-page explanation of desirable difficulty. Make It Stick, by Brown, Roediger, and McDaniel, is the popular book-length treatment if you want the trade version.