The forgetting curve

In the 1880s, a German psychologist named Hermann Ebbinghaus ran one of the strangest experiments in psychology, on himself. He made up more than two thousand meaningless syllables, like WID and ZOF, so nothing he learned would connect to anything he already knew. He memorized lists of them until he could recite each one perfectly. Then he waited, anywhere from twenty minutes to a month, and timed how long it took to learn the same list again.

If relearning went faster than learning, some memory had survived, even when he couldn't recall a single syllable. That measure let him chart forgetting over time.

The shape he found has held up for over a century. A researcher repeated the whole project on himself in 2015 and got a very similar curve. Forgetting is fast at first, then slows down. Most of what you lose, you lose in the first hours and days. Whatever survives that early drop fades much more slowly.

The exact numbers depend on what you're learning. Nonsense fades faster than a story, and a story fades faster than an idea you understand and use. Be wary of anyone who tells you you'll forget a precise share of a lecture by tomorrow. The shape is reliable. The numbers aren't.

Bending the curve

Ebbinghaus noticed something else. When he spread his repetitions across several days, he needed fewer of them in total than when he packed them into one sitting. This is the spacing effect, and it's among the most consistently replicated findings in psychology. It shows up in children and adults, with everything from vocabulary to motor skills.

Try it below. Tap Cram, then Spaced. Both use four reviews. Then tap the chart to add or remove reviews of your own.

The numbers in that chart are invented to show the shape. Real curves vary with the person and the material. The pattern is what holds. The same reviews leave you with more when they're spread out than when they're bunched together.

Why spacing works

Each review sends the memory back up and makes the next fall slower. How much it helps depends on timing. Review something you saw ten minutes ago and the answer is right there. It's easy, so it does little. Review it after a couple of days, when you have to work to bring it back, and the memory gets a much bigger boost. The psychologist Robert Bjork calls this a desirable difficulty, meaning effort that feels worse in the moment and pays off later.

That's why cramming is so tempting. Five hours the night before a test feels productive, because everything is fresh and familiar. And it does work for tomorrow's test. A month later, little of it is left. Spaced study feels slower and leaves far more behind.

How far apart?

There's no single best gap. One large study of more than a thousand people found that the best spacing grows with how long you need to remember. For a test in a week, a review a day after first learning worked well. For something you need to keep for a year, a gap of about three weeks did best.

A simple plan works for most people. Review a new idea the next day, again a few days later, and again a week or two after that. Spaced repetition apps handle the schedule for you.