Home / Guides / What is the forgetting curve?

What is the forgetting curve?

Last updated July 7, 2026 · 5 min read

Who discovered the forgetting curve?

Hermann Ebbinghaus was a German psychologist who, in the 1880s, ran one of the first systematic experiments on human memory — on himself. He memorized long lists of meaningless three-letter syllables (deliberately meaningless, so prior knowledge couldn't help him), then tested how much he could still recall after different lengths of time. Plotting the results gave the shape now known as the forgetting curve: a decline that starts steep and gradually flattens out. It's one of the oldest findings in experimental psychology, and the general shape has been reproduced in later memory research using real words and real learners, not just Ebbinghaus's syllables.

Why does memory decay so fast at first?

The forgetting curve isn't evidence that you're bad at learning — it's the default behavior of memory for anyone, for anything new. Right after you learn something, it exists as a fragile, recently-formed trace that hasn't been reinforced by use. Most of what enters memory this way is never touched again, so it makes a kind of sense that the brain doesn't commit to holding onto all of it equally — the sharpest drop happens early, before a piece of information has proven it's worth keeping around. That's why a word you learned yesterday can feel completely gone today, even though you were sure you had it in the moment.

How review resets the curve

The forgetting curve isn't a fixed sentence for a given word — it's what happens by default, absent review. Every time you successfully recall something before you've fully lost it, the curve resets: the next decline starts from a stronger position and drops off more slowly than the one before. Do that enough times, spaced at the right intervals, and the curve keeps flattening until a word barely fades at all between reviews. This is the mechanism spaced repetition is built on — timing each review to land right at the point where the curve is about to steepen, so you catch the word before it's gone rather than after.

What the forgetting curve means for language learners

If you've ever learned a word in a lesson, felt confident about it, and then blanked on it a week later, you've experienced the forgetting curve directly — it wasn't a failure of effort, it's what happens to any newly learned word left unreviewed. The practical takeaway is that when you review matters as much as how you learned something in the first place. CatchFrase's smart review is built around this: instead of asking you to review everything on a fixed daily list, it uses the FSRS algorithm to bring each word back right around the point the curve suggests you're about to forget it — see what is spaced repetition for how the timing works.

Common questions

Is the forgetting curve real and proven?

Ebbinghaus's original experiments were on himself alone, using nonsense syllables rather than real vocabulary, so the exact shape he measured shouldn't be read as a universal constant. But the general pattern — fast early decay that levels off, and review slowing that decay — has held up across a large body of later memory research and is the basis for spaced repetition.

How fast do you forget a new word?

Faster than feels intuitive, and the exact speed varies by person, by how the word was learned, and by how it connects to things you already know. What's consistent is the shape: the first hours and days after learning something new are when most of the forgetting happens, which is exactly why an early review matters more than a later one.

Can you stop the forgetting curve completely?

Not permanently in one step, but you can flatten it repeatedly. Each successful recall before you'd have forgotten resets the curve and makes the next decline slower than the last. Enough resets and a word starts to feel less like something you're holding onto and more like something you just know.