Home → Blog → Memory Science

The Forgetting Curve Explained: What Ebbinghaus Found

·6 min readMemory Science

The forgetting curve describes how quickly newly learned information fades from memory over time if you never review it, dropping fastest in the first hours and days, then leveling off more gradually. It comes from research by psychologist Hermann Ebbinghaus, published in 1885, and it's the main scientific basis for why spaced review works better than studying something once and moving on.

Who Ebbinghaus Was and What He Did

Hermann Ebbinghaus was a German psychologist who wanted to study memory scientifically, at a time when memory was mostly discussed philosophically rather than measured experimentally. To avoid the problem of prior knowledge and meaning affecting results, he memorized lists of nonsense syllables, meaningless three-letter combinations like "zub" or "fik", and tested himself on his own retention at different time intervals after learning them.

He was both the researcher and the only subject in his experiments, which is a real limitation by modern standards; findings from one person's memory, using one specific memorization method, don't automatically generalize to everyone. Still, his basic finding, that forgetting happens fastest soon after learning and then slows down, has held up reasonably well and influenced memory research ever since.

What the Curve Actually Shows

The general shape, replicated conceptually in later research even if exact numbers vary by study and material, looks like this:

Time since learning Relative retention (general pattern, not exact figures)
Right after learning Highest
Within the first day Noticeable drop
After a few days Continued, slower decline
After a week or more Decline continues but flattens further

The key shape is a steep early drop followed by a long, shallow tail, not a straight-line decline. This is why cramming the night before a test can work for a next-day exam (you're testing near the top of the curve) but fails for material you need to remember weeks or months later (you've fallen far down the curve by then).

What the Curve Does Not Show

It's easy to find graphics online claiming the forgetting curve proves you forget "50% within an hour" or "70% within a day." Treat specific numbers like these with skepticism. Ebbinghaus's original data came from memorizing meaningless syllables through rote repetition, not from meaningful classroom material, real-world skills, or diverse learners. Modern retellings often round his findings into clean, dramatic statistics that go beyond what the original research supports.

What the curve reliably shows, and what's relevant for studying, is the shape of forgetting: fast early loss, slower loss later, and the loss is slowed by meaningful connections and sped up by meaningless, disconnected information. The precise numbers matter far less than that general pattern.

It's also worth noting the forgetting curve on its own doesn't tell you how to fight forgetting. For that, you need a separate body of research on review timing.

Why Spaced Review Works Against It

The practical insight from forgetting-curve research is that review timed to happen before you've fully forgotten something is far more efficient than relearning it from scratch after you've forgotten it. Each successful review appears to make the next forgetting curve for that item flatter, meaning you retain it longer before needing to review again.

This is the mechanism behind spaced repetition systems: instead of reviewing everything daily (wasteful, since you still remember most of it) or reviewing it once and never again (which lets it fall off the curve entirely), you review each item right around the point you'd be about to forget it. Over several cycles, the gaps between reviews get longer because the memory gets more durable. See how spaced repetition works for the scheduling logic behind this.

Cepeda et al.'s 2006 meta-analysis of distributed practice studies supports the broader principle: spacing study sessions apart, rather than massing them together, produces better long-term retention across a wide range of material, consistent with the idea that timed review counteracts the forgetting curve more efficiently than cramming does.

A Worked Example

Say you learn 30 new vocabulary words on Monday with no review.

Turn your own material into a study course

Upload a PDF, lecture recording or YouTube link. gigabrainz builds chapters, quizzes and spaced-repetition flashcards from it in a couple of minutes. Your first 3 courses are free, no card needed.

Try gigabrainz free →

This is the specific problem flashcard scheduling is built to solve: gigabrainz's flashcards use spaced repetition scheduling so that review timing adapts to how well you already know each card, aiming to catch material right as it's about to fall off your personal forgetting curve rather than on a fixed daily or weekly schedule.

Factors That Change How Fast You Forget

The steepness of your own forgetting curve isn't fixed; several factors change how quickly a given piece of information fades.

None of these factors change the basic shape of the curve, fast early decline followed by a slower tail, but they do change how steep that initial decline is for any given piece of information you're trying to hold onto.

How to Use This Practically

  1. Review new material within a day or two of first learning it, not weeks later. This is while you're still high enough on the curve that review is cheap and effective.
  2. Don't rely on one exposure for anything you need long-term. A single reading or lecture puts you on the steep part of the curve; you need at least one follow-up review to flatten it.
  3. Space reviews out, don't repeat them back-to-back. Reviewing the same material five times in one evening mostly tests short-term memory, not the kind of durable recall the forgetting curve is describing.
  4. Prioritize review for material you're about to forget, not material you already know well. This is the core logic behind spaced repetition flashcards, and it's more efficient than reviewing your whole deck equally every day.

The Takeaway

The forgetting curve's real lesson isn't a specific percentage to memorize; it's that forgetting is fastest right after learning and slows down over time, and that timed review interrupts that decline more efficiently than either cramming once or reviewing everything constantly. If you build your study schedule, like the one in how to make a study schedule, around spaced review rather than one-time exposure, you're working with the curve instead of against it.

For material that resists spacing, like ordered lists or speeches, the memory palace technique gives you a second tool.

Frequently asked questions

What is the forgetting curve?

The forgetting curve is a graph showing how quickly memory of newly learned information fades over time without review. It was first described by psychologist Hermann Ebbinghaus in 1885 based on his own memorization experiments.

Did Ebbinghaus really find we forget 50% in a day?

Specific percentages you may see online vary by source and are often overstated or misquoted. Ebbinghaus's original work showed a steep initial decline in retention followed by a slower decline, using nonsense syllables and self-experimentation, not a precise universal percentage that applies to all material and all people.

Does the forgetting curve apply to everything you learn?

The original research measured memory for meaningless syllables, which forget faster than meaningful, well-understood information. Real-world material, especially if connected to things you already know, generally resists forgetting better than Ebbinghaus's syllables.

How does spaced repetition fight the forgetting curve?

Each time you review and successfully recall information, the subsequent forgetting curve for that memory tends to flatten, meaning you forget more slowly next time. Spaced repetition times reviews to happen before you'd fully forget, repeatedly flattening the curve.