Ebbinghaus Forgetting Curve: Why You Forget Words (And How to Stop)

You study 50 new vocabulary words. You feel confident. A week later, you can barely remember 15. This experience is almost universal among language learners, and it is not a character flaw or a poor memory. It is a predictable consequence of how human memory works — a phenomenon that Hermann Ebbinghaus first described in 1885 and that neuroscience has confirmed ever since.

Understanding the forgetting curve is not just interesting — it fundamentally changes how you should study.

Who Was Ebbinghaus?

Hermann Ebbinghaus was a German psychologist who became the first person to study human memory with scientific rigor. In the 1880s, he conducted extensive experiments on himself, using nonsense syllables (like "DAX" and "BUP") to memorize and then track how quickly he forgot them.

His methodology was meticulous: he would memorize a list, wait a set period, then test how many repetitions it took to relearn the list. The difference in effort between the first learning and the relearning told him how much memory remained.

The result of these experiments was one of psychology's most important discoveries.

The Forgetting Curve: What It Shows

Ebbinghaus plotted his forgetting rate over time and found a consistent pattern. After learning something new:

After 20 minutes: You forget roughly 42% of what you learned
After 1 hour: About 56% is gone
After 1 day: Approximately 67% has faded
After 1 week: Around 77% is lost
After 1 month: Up to 79% may be gone

The curve is steep at first and then flattens. Most forgetting happens within the first few hours and days after learning — not gradually over weeks.

What makes this finding so important is its universality. The curve applies to everyone. It does not matter how intelligent you are or how hard you try during the initial study session. Without intervention, forgetting is the default outcome.

Why Cramming Fails

The most common student response to an upcoming exam is to cram — to study intensively in a single session immediately before the test. Cramming can produce short-term gains because the material is still in active memory when the test happens.

But cramming exploits a loophole in the forgetting curve rather than changing it. The information learned through a single intensive session follows the same decay curve as any other learning. Within a week of the exam, most of that crammed material is gone.

For language learning specifically, cramming is particularly destructive because:

Vocabulary requires production, not just recognition. Knowing a word well enough to pick it out of a multiple-choice list is not the same as knowing it well enough to use it in conversation. Production requires deeper encoding that single-session study rarely achieves.
Language builds on itself. If your vocabulary from month one decays before you reach month two, every subsequent lesson becomes harder because it lacks the foundation it depends on.
The volume of language material is too large to cram. A working vocabulary requires 2,000–5,000 words. No single study session can handle that load.

How Spaced Repetition Beats the Curve

Ebbinghaus himself identified the solution in the same series of experiments that revealed the problem. He found that reviewing material at spaced intervals dramatically reduced the rate of forgetting.

Each successful review does two things:

It restores the memory to its full strength
It makes the memory more resistant to future decay — the next forgetting curve after a review is flatter and longer

After multiple spaced reviews, a memory that would have vanished within a week becomes resistant to forgetting for months or years. Ebbinghaus called this the spacing effect, and it is one of the most replicated findings in cognitive psychology.

The practical implication is straightforward: instead of one long study session, distribute your reviews across time. The same total study time produces dramatically better retention when spaced than when massed.

The SM-2 Algorithm: Spaced Repetition Made Practical

In the 1980s, Polish researcher Piotr Wozniak formalized the spacing effect into a computable algorithm called SM-2 (SuperMemo 2). SM-2 calculates the optimal interval for reviewing each piece of information individually, based on how well you recalled it during previous reviews.

The core logic:

If you recalled a word easily, the next review is scheduled further in the future (the interval grows)
If you struggled to recall a word, the interval resets to a shorter period
Over time, well-known words appear rarely; difficult words appear frequently

This makes study time maximally efficient. You never waste time reviewing words you already know well, and you never allow a struggling word to slip back into forgetting.

Apps like Voccle implement the SM-2 algorithm automatically. When you flip a flashcard and rate your confidence, the algorithm updates your review schedule in the background. You do not have to think about the math — you just study, rate, and let the system handle the timing.

A Practical Review Schedule Without an App

If you prefer a manual system, here is a simplified schedule based on spaced repetition principles:

Day 1: Study new vocabulary
Day 2: Review everything from Day 1
Day 4: Review Day 1 words again
Day 8: Another review
Day 16: Review again
Day 30: Monthly review
Day 60: Final consolidation review

Words you recall easily at each stage move to the next interval. Words you forget get reset to Day 1. This manual system requires discipline and tracking but demonstrates the principle clearly.

Memory Consolidation: The Science Behind Why This Works

Spaced repetition works because it aligns with how biological memory consolidation functions. When you first encode a memory, it exists in a fragile, temporary state in the hippocampus. With time and sleep, memories undergo consolidation — they are gradually transferred to long-term storage in the cortex.

Each retrieval of a memory (each time you successfully recall a word) strengthens the neural pathways associated with that memory and partially reconsolidates it. The act of retrieval is itself a form of learning — a phenomenon called the testing effect or retrieval practice effect.

This means that the active process of trying to recall a word — even if you struggle — produces stronger memories than passively re-reading a word list. Flashcard apps that require you to attempt recall before revealing the answer are leveraging this effect directly.

Applying This to Your Vocabulary Study

The most important changes to make based on the science:

Review within 24 hours of learning something new. The forgetting curve is steepest in the first day. A brief review before sleep dramatically reduces that initial loss.
Use active recall, not passive review. Cover the answer and try to produce the word before checking. Struggle is productive.
Use a spaced repetition tool. Manually calculating intervals for hundreds of words is impractical. Voccle and similar apps automate this completely.
Study daily rather than in occasional long sessions. Fifteen minutes every day beats three hours once a week for long-term retention.
Trust the algorithm when it says you know a word. Learners often over-review easy words because it feels productive. This is an inefficiency — trust the spacing, and spend that time on genuinely difficult material.

The forgetting curve is not your enemy. It is a map. Follow the map in reverse, and you will retain vocabulary that most learners lose forever.