You studied 50 new words last week. Today, you can barely remember 12. Before you blame yourself, know this: your brain is working exactly as designed. The problem isn't you — it's that nobody taught you how memory actually works.

In this article, we'll look at the science of forgetting, why it happens, and the single most effective technique to beat it: spaced repetition.

The Experiment That Changed Everything

In the 1880s, German psychologist Hermann Ebbinghaus did something no one had done before: he turned memory into an experiment. Using himself as the only subject, he memorized thousands of nonsense syllables — meaningless combinations like "DAX", "BUP", "ZOL" — and meticulously tracked how quickly he forgot them.

Why nonsense syllables? Because Ebbinghaus wanted to isolate pure memory from meaning, context, and emotion. He wanted to see how the raw machinery of retention and forgetting worked, stripped of everything else.

His results, published in Über das Gedächtnis (1885), produced one of the most replicated findings in all of psychology: the forgetting curve.

The Forgetting Curve in Numbers

Ebbinghaus found that memory decay follows a roughly exponential pattern. Without any review:

Read those numbers again. After just one day, you've lost two-thirds of what you studied. After a week, three-quarters. This isn't a personal failing — it's the default behavior of the human memory system.

The shape of this curve has been replicated in hundreds of studies since Ebbinghaus, including a major 2015 replication study published in PLOS ONE (Murre & Dros, 2015). The curve is real, it's universal, and it applies to vocabulary learning with particular force because isolated words lack the rich contextual anchors that slow forgetting.

What's Actually Happening in Your Brain

When you learn a new word, your brain forms a memory trace — a pattern of neural connections, primarily in the hippocampus, that encodes the new information. Think of it as a faint path through a forest.

The problem is that your brain is constantly evaluating which paths are worth maintaining. Connections that aren't reactivated get pruned — a process called synaptic decay. This isn't a bug; it's a feature. Your brain has limited metabolic resources and can't afford to maintain every piece of information it encounters. It keeps what you use and discards what you don't.

For language learning, this means that a word you encountered once in a vocabulary list is, from your brain's perspective, barely worth remembering. It was used once, in a low-stakes context, with no emotional weight. The neural trace is fragile and fades fast.

Why Cramming Doesn't Work for Languages

Here's the cruel irony: cramming feels effective. You sit down for three hours before a test, drill 200 flashcards, and walk in confident. You might even pass the exam.

But what happens two weeks later? The words are gone.

This is what researchers call the spacing effect — one of the most robust findings in cognitive psychology. Massed practice (cramming) produces strong short-term performance but weak long-term retention. Distributed practice (spacing) produces the opposite: weaker short-term performance but dramatically stronger long-term retention.

A landmark study by Cepeda et al. (2006), published in Psychological Bulletin, analyzed 254 studies involving over 14,000 participants and confirmed that distributed practice is superior to massed practice across virtually all conditions tested. The effect is not subtle — it's one of the largest and most reliable effects in all of learning science.

For language learning specifically, this has a devastating implication: the weekend study marathon is one of the least efficient ways to learn vocabulary. Two hours on Saturday is dramatically less effective than 20 minutes each day for six days — even though the total time is the same.

The Solution: Spaced Repetition

Ebbinghaus himself discovered the fix. In his original research, he found that repetition spaced over time dramatically improved retention compared to massed practice. Each review at the moment of near-forgetting does two critical things:

  1. Restores the memory trace — the neural pathway is reactivated and strengthened.
  2. Slows the rate of future forgetting — each successful recall makes the memory more durable, so the next review can be delayed longer.

This is the core principle behind Spaced Repetition Systems (SRS): instead of reviewing everything every day, you review each item at the optimal moment — right before you would have forgotten it. Early reviews are frequent; as the memory strengthens, intervals stretch from days to weeks to months.

The Optimal Review Schedule

Research suggests the following approximate schedule for new material:

  1. First review: within 1–2 hours of initial study
  2. Second review: 24 hours later
  3. Third review: 3–7 days later
  4. Fourth review: 2–4 weeks later
  5. Fifth review: 2–6 months later
  6. Subsequent reviews: expanding intervals based on your personal forgetting rate

The magic is in the expanding intervals. Each successful retrieval doesn't just prove you remember — it changes the memory itself, making it more resistant to decay. After enough well-timed reviews, a word moves from fragile short-term storage into robust long-term memory that can persist for years.

Research shows that reviewing material within 24 hours of initial study can increase retention to roughly 80%, compared to the 33% you'd retain without review. Over 30 days, distributed practice is dramatically more effective than an equivalent amount of massed practice in a single session.

How to Implement This Tomorrow: Anki in 5 Minutes

The theory is elegant, but you need a practical tool. You can't manually track optimal review times for thousands of vocabulary items. This is where Anki comes in — a free, open-source flashcard application that automates spaced repetition scheduling.

Here's how to get started in 5 minutes:

1. Download Anki

Get it from apps.ankiweb.net. It's free on desktop (Windows, Mac, Linux) and Android. The iOS version costs $25 but is a one-time purchase worth every cent if you're serious about language learning.

2. Create Your First Deck

Create a deck named after your target language (e.g., "Japanese Vocabulary"). Keep it simple — one deck per language is enough to start.

3. Enable FSRS

Go to your deck options and enable FSRS (Free Spaced Repetition Scheduler). This is a modern algorithm based on the "Three-Component Model of Memory" that uses machine learning to personalize scheduling. It achieves the same retention as the older SM-2 algorithm with 20–30% fewer reviews. Set your desired retention to 90% — this is the sweet spot where you retain most of what you learn without drowning in reviews.

4. Add Cards the Right Way

Don't just put single words. Create sentence cards: put a complete sentence on the front (in your target language) with the translation and key vocabulary on the back. This gives your brain context, collocation patterns, and grammar — all in one card.

5. Set a Daily Limit

Start with 10 new cards per day. This sounds low, but remember: every new card generates future reviews. At 10 new cards per day, within a month you'll be reviewing 80–100 cards daily. Start lower than you think you should.

6. Do Your Reviews Every Day

This is the non-negotiable rule. Anki's algorithm only works if you show up consistently. Miss a day and your reviews pile up; miss a week and you're buried. Fifteen minutes daily beats two hours on the weekend — that's the whole point of spaced repetition.

The Bottom Line

Your brain forgets most of what it learns — and it does so rapidly. This isn't a flaw; it's an optimization. But it means that how you schedule your reviews matters more than how much time you spend studying.

The forgetting curve is the enemy. Spaced repetition is the weapon. And Anki is the tool that makes it practical.

You don't need to study more. You need to study at the right time.

This article is part of the series "The Science of Language Learning" — where we break down what research actually says about how adults acquire languages, and how to use that science to learn faster.

Next in the series: Input vs. Output: What Actually Drives Language Acquisition?

References: