The Forgetting Curve Explained: Ebbinghaus, Memory, and How to Beat It

In 1885, a German psychologist named Hermann Ebbinghaus published a small book that would quietly become one of the most influential texts in cognitive science. Über das Gedächtnis (On Memory) described a series of experiments he had run on himself, memorising lists of nonsense syllables and tracking how quickly he forgot them. The graph he produced from those experiments — a steep curve sloping downward from full recall to near-total loss — is what we now call the forgetting curve, and it still defines how we think about memory loss today.

This article explains what the forgetting curve actually says, why it works the way it does, what later research has confirmed and refined, and most importantly how to use the science to remember more of what you learn.

What the Forgetting Curve Shows

Ebbinghaus memorised lists of three-letter nonsense syllables — combinations like DAX, BUP, ZOL — chosen specifically because they had no meaning and therefore could not be remembered through association with prior knowledge. He then tested himself at various intervals to see how much remained.

The pattern he found was consistent and stark:

• After 20 minutes, he had forgotten roughly 42% of the material
• After 1 hour, around 56% was gone
• After 24 hours, about 66% was lost
• After 6 days, retention had dropped to roughly 25%
• After 31 days, only around 21% remained

Plotted on a graph, this produces a curve that drops steeply at first, then gradually flattens. Most of the forgetting happens in the first 24 hours after learning. After that initial collapse, the remaining material decays slowly.

The forgetting curve is not a law of nature in the strict sense — the exact percentages vary by individual, by content type, and by how the material was originally learned. But the shape of the curve has been replicated thousands of times since 1885, including in modern studies using fMRI and large-scale online learning data. It is one of the most robust findings in cognitive psychology.

Why the Brain Forgets

The forgetting curve looks like a failure of memory, but it is better understood as a feature. The brain is not a recording device; it is a prediction machine. Information that does not get reinforced is treated as low-value and gradually pruned.

There are three main mechanisms at work:

Decay

The simplest model of forgetting is that memory traces fade over time without active use. Modern neuroscience refines this — the physical synaptic connections that constitute a memory are not destroyed but become harder to access. With no retrieval pressure, the trace becomes weaker relative to competing memories.

Interference

New information competes with old information for retrieval pathways. If you learn a Spanish vocabulary list today and a different Spanish list tomorrow, the second list partially interferes with the first. This is why your memory of a single concept feels sharper than your memory of one of fifty similar concepts.

Retrieval failure

Sometimes the memory is still encoded but the cue you are using to retrieve it does not match the cue that was present at encoding. The "tip of the tongue" experience is a classic example. The information has not vanished — the access route has.

The forgetting curve captures the combined effect of all three. The steep initial drop reflects the brain's default behaviour: in the absence of retrieval signals, assume the information is not needed and let it decay.

What Ebbinghaus Also Discovered

The forgetting curve is the famous half of Ebbinghaus's work. The less-famous half is what he discovered next, and it is what makes the forgetting curve actionable rather than depressing.

When Ebbinghaus re-studied the same material after a delay, the second forgetting curve was shallower than the first. Each subsequent review produced a shallower curve still. With enough spaced reviews, the curve flattened almost completely — meaning the material was retained almost indefinitely.

This is the principle that became known as spaced repetition, and it underpins every modern flashcard system from Anki to Duolingo. The forgetting curve tells you what happens if you do nothing. Spaced repetition tells you what to do about it. We cover this in detail in our science of spaced repetition guide.

How Modern Research Refined the Curve

Ebbinghaus's experiments were limited in important ways. He was a sample size of one. His material was deliberately meaningless. His test environment was small and controlled. Later researchers have replicated the basic finding while also identifying conditions under which the curve looks different.

Cepeda et al. (2006) — a meta-analysis of 254 studies of distributed practice — confirmed the basic shape of the forgetting curve but showed that the optimal gap between reviews depends heavily on how long you need to retain the material. For a one-week retention target, reviews should be spaced about 24 hours apart. For a one-year target, reviews should be spaced weeks apart.

Murre & Dros (2015) — a direct replication of Ebbinghaus's original experiments — confirmed his forgetting curve to within a few percentage points more than 130 years later. The 1885 numbers were close enough to be remarkable.

Carpenter et al. (2018) — meta-analyses of educational settings — showed that spaced study produces an average of 20% better retention than massed study at the same total time.

Baumgartner et al. (2021) — a Psychological Bulletin meta-analysis — identified spaced repetition as one of the most reliable effects in experimental psychology, with effect sizes that hold up across age groups, content domains, and time scales.

The cumulative picture is that Ebbinghaus was essentially right. Modern research has filled in detail, but the core curve — and the spaced-repetition counter to it — is settled science.

How to Beat the Forgetting Curve in Practice

The actionable version of the forgetting curve research is simple. You do not need a complex schedule or specialist software. You need to do three things.

1. Review within 24 hours

The single biggest leverage point is the first review. Material reviewed once within 24 hours of initial learning retains roughly twice as much at the one-week mark as material that was learned and never reviewed. A 5-minute review the next day will outperform a 30-minute review a week later.

2. Space subsequent reviews further apart

After the first review, the optimal gap roughly doubles each time. A common schedule looks like: review at 1 day, then 3 days, then 7 days, then 14 days, then 30 days. By the fifth review the material has effectively transferred into long-term memory.

The exact numbers do not matter as much as the principle. Expanding intervals work because they force retrieval just before the material would have been forgotten — which produces the strongest memory consolidation.

3. Use retrieval, not re-reading

The forgetting curve flattens fastest when reviews involve active retrieval — testing yourself rather than re-reading notes. Passive re-exposure produces only modest gains. Closing the book and trying to recall the material before checking is dramatically more effective. This is the testing effect, and it interacts powerfully with spacing. Our retrieval practice explained guide covers the mechanism in depth.

Why Microlearning Suits the Forgetting Curve

The forgetting curve is the cognitive-science case for short, frequent learning sessions over long infrequent ones. Five 10-minute sessions spaced across a week leave more in long-term memory than a single 50-minute session, even though total study time is identical.

This is the principle that gives microlearning its retention advantage. Apps that deliver short daily sessions are not just convenient — they are structured to match how human memory actually works. A platform like Chunks takes 5-to-10-minute story chapters and surfaces them in a daily cadence that mirrors the spaced-review intervals the research recommends.

This is also why cramming feels effective but rarely is. Cramming produces strong performance on a test taken immediately afterwards, because the material is still in working memory. The forgetting curve then collapses the gains over days or weeks. If the goal is to remember information for longer than the next 24 hours, cramming is one of the least efficient strategies available.

Common Misconceptions About the Forgetting Curve

"I have a bad memory, so the curve is worse for me."

The forgetting curve is roughly universal across healthy adults. Individual variation exists but is smaller than people assume. What looks like a "bad memory" is almost always a review habit problem — material is encoded but never reinforced. The same person who claims to forget everything they read will remember a song lyric heard 100 times without effort.

"If I just understand it well, I won't forget it."

Understanding helps encoding, but it does not prevent decay. The forgetting curve applies to material you understood deeply at the time of learning. Without retrieval, the memory still fades. Comprehension is a foundation for retention, not a substitute for it.

"Modern brains forget faster because of phones."

The 2015 replication of Ebbinghaus's experiments found forgetting rates within a few percentage points of his 1885 results, more than a century after smartphones were invented. The forgetting curve is a property of the brain, not of the era.

Frequently Asked Questions

What is the forgetting curve?

The forgetting curve is a graph showing how rapidly newly learned information is lost from memory over time. It was first described by Hermann Ebbinghaus in 1885 and shows that roughly 50% of material is forgotten within an hour, 70% within a day, and 80% within a month — unless the material is reviewed.

Who discovered the forgetting curve?

The forgetting curve was discovered by Hermann Ebbinghaus, a German psychologist, in experiments on himself published in 1885 as Über das Gedächtnis (On Memory). His findings have been replicated many times since, most notably in a 2015 study by Murre and Dros that matched his original numbers closely.

How fast do you forget what you learn?

Without review, you forget roughly 50% of new information within one hour, 70% within 24 hours, and 80% within a month. The forgetting curve is steepest in the first 24 hours, then flattens. The exact percentages vary by individual and content type but the shape of the curve is consistent.

How do you beat the forgetting curve?

You beat the forgetting curve by reviewing material at spaced intervals — first within 24 hours, then at gradually expanding gaps (3 days, 7 days, 14 days, 30 days). Active retrieval (testing yourself) is more effective than re-reading. This combination of spacing and retrieval is called spaced repetition and is the most effective study strategy known.

Is the forgetting curve still accurate today?

Yes. A 2015 direct replication of Ebbinghaus's experiments by Murre and Dros found forgetting rates within a few percentage points of his original 1885 numbers. The curve has been confirmed thousands of times across different populations, content types, and decades of research. It is one of the most robust findings in cognitive psychology.

Does the forgetting curve apply to skills as well as facts?

Partially. Skill memory (procedural memory) decays more slowly than fact memory (declarative memory) once the skill is well-learned, which is why you can ride a bicycle years later. But during the initial skill-acquisition phase, the forgetting curve still applies — practice sessions spaced over weeks produce more durable skills than the same amount of practice crammed into a single session.

Summary

The forgetting curve, first described by Hermann Ebbinghaus in 1885 and confirmed by direct replication in 2015, shows that human memory loses around 50% of new information within an hour, 70% within a day, and 80% within a month — unless the material is reviewed. The shape of the curve is a feature of how the brain prioritises information, not a flaw to be fixed. The way to beat it is spaced repetition: a first review within 24 hours, then expanding-interval reviews at roughly 3, 7, 14, and 30 days, using active retrieval rather than re-reading. This is why short, frequent learning sessions outperform long infrequent ones, and why microlearning platforms that surface daily lessons are structured to match how human memory actually works. The forgetting curve is not depressing once you know what to do about it — it is the strongest argument in cognitive science for studying in small, regular doses rather than cramming.