What It Actually Is
If you study something once and never look at it again, you forget most of it fast. If you study it, wait, come back to it, wait longer, come back again — you remember it much better, for much longer.
That's it. That's the whole phenomenon. Psychologists call it the spacing effect: spreading your reviews out over time beats cramming them all together, even if the total amount of study time is the same.
This isn't a new discovery or a trend. It's one of the most repeated, most confirmed findings in all of psychology — over 100 years of experiments keep finding the same thing.
Same total study time. Spread out instead of crammed. Dramatically better long-term memory.
Where This Idea Comes From
In 1885, a German psychologist named Hermann Ebbinghaus wanted to study memory scientifically. The problem: if he tested people on real words or facts, their existing knowledge would mess up the results.
So he made up meaningless three-letter combinations (like "zuk" or "vom") and tested himself — memorizing lists, then checking how much he'd forgotten over time. He plotted his results and got what's now called the Ebbinghaus Forgetting Curve — a graph showing memory dropping fast at first, then leveling off.
You'll constantly hear "you forget 70% of what you learn within 24 hours." That number is a rounded-off, oversimplified version of what Ebbinghaus actually found.
His real numbers were closer to 67% forgotten (33% remembered) after a day — and that's specific to meaningless syllables he memorized once, tested only on himself. It's not a universal law about how you'll forget your Accounts syllabus.
To be fair to Ebbinghaus: modern scientists have redone his experiment and gotten a similar-shaped curve. So the idea holds up. It's the exact numbers that get exaggerated when people repeat this on YouTube.
Why Spacing Actually Works
Here's something most explainer videos won't tell you: scientists don't fully agree on why spacing works. There isn't one clean answer. There are several competing explanations, and each one has evidence behind it. The honest thing to say is that it's probably a mix of all of them.
Idea 01 Dig for it
When you come back to something after a gap, you can't just glance at it — you have to actually try to remember it. That act of trying to remember is what makes the memory stronger.
If you review something while it's still fresh, there's nothing to dig for, so nothing gets strengthened.
Idea 02 More paths
When you study something at different times, in different moods, maybe in different places, your brain attaches that memory to more "tags."
Later, more tags mean more ways to accidentally stumble back onto that memory when you need it.
Idea 03 Novelty
If you see the same thing twice in a row, your brain treats the second time as old news and barely pays attention.
Wait a while, and your brain treats it like new information again — actually processing it instead of skimming.
Idea 04 Consolidation
There's a biological process where memories get consolidated (stored more permanently) over hours and days.
A second study session that happens after that process has had time to work may benefit from it.
None of these cancel each other out — they're probably all a little bit true at once.
What the Evidence Shows
Two studies matter more than any others here.
Other research (Roediger & Karpicke, 2006) found that actively trying to recall something — quizzing yourself — builds a stronger memory than rereading it, even when rereading feels more productive in the moment.
Spacing and self-testing aren't two separate tricks. They work best combined: space out your self-tests, not just your rereading.
Desirable Difficulties
A researcher named Robert Bjork coined a phrase for all of this: desirable difficulties.
The idea: things that make studying feel harder right now — waiting before you review, forcing yourself to recall instead of looking it up, mixing topics instead of doing one at a time — are usually the same things that make the memory stick later.
Meanwhile, things that make studying feel smooth and easy right now — rereading, cramming, testing yourself immediately after learning — often create a false sense of "I've got this," which falls apart days later.
If studying feels a little effortful, that's usually a sign it's actually working.
The Flashcard App Myth
This is worth its own section because it's genuinely surprising, and almost nobody covers it.
You've probably heard that spaced repetition should use increasing gaps — review after 1 day, then 3 days, then 7 days, then 21 days, each gap bigger than the last. This is how most flashcard apps are designed, and it's presented everywhere as scientifically proven.
A study by Karpicke and Roediger (2007) directly tested increasing gaps against equal, fixed gaps (same number of days between every review).
Increasing gaps won on a quick test given 10 minutes later. But on a test given two days later — the kind of "did I actually remember it" test that matters — the equally-spaced schedule won instead. Several other studies have found the same reversal.
Overall, the research on this specific question is genuinely mixed: some studies favor increasing gaps, some favor equal gaps, and a good number find no real difference at all.
What the research agrees on more consistently isn't the shape of the schedule — it's that your very first review shouldn't happen too soon. Waiting a bit before your first review matters more than whether your later reviews increase, stay equal, or something else.
This is a genuinely honest correction to make, because it pushes back on how most spaced-repetition tools market themselves.
How the Tools Evolved
A memory trick from 1972 that used to be a manual box of index cards has quietly turned into a machine-learning problem — computers are now better than humans at guessing when you specifically are about to forget something.
Leitner System — 1972
The original manual version. A box with several compartments; flashcards move to a "further apart" compartment when you get them right, and back to a "closer together" one when you get them wrong. All done by hand.
SM-2 Algorithm — late 1980s, by Piotr Wozniak
The first algorithm to calculate review intervals automatically, instead of you deciding manually. This became the backbone of Anki (the most popular flashcard app) for over a decade.
FSRS — 2023–present
A newer algorithm that replaced SM-2 as Anki's default. Instead of one fixed formula for everyone, it uses machine learning, trained on hundreds of millions of real study sessions, to predict — for you specifically — when you're likely to forget something, and times your next review right before that point.
It genuinely produces fewer, better-timed reviews for the same level of memory.
SM-20 — 2026
The newest version of the original SuperMemo algorithm, where every part of the scheduling is now learned from data rather than hand-designed by a person.
The scheduling problem used to be something you solved with paper and discipline. Now it's something an algorithm solves for you — which means your job is reduced to doing the reviews when they show up, not figuring out when they should be.
Where This Doesn't Work
To keep this grounded, it's worth being upfront about where the evidence is weaker.
Facts vs. complex skills Biggest gap
Spacing is most strongly proven for facts and definitions — vocabulary, formulas, dates, terminology.
For complex problem-solving or applying concepts in new situations, the evidence is thinner and less consistent.
More gap isn't automatically better
If the gap is too long before you've actually learned something the first time, you just forget it completely instead of strengthening it.
There's a sweet spot, not a straight line where "more gap = more benefit."
Spacing without testing is a weaker version
Just glancing at old notes on a schedule isn't the same as trying to recall them on a schedule. The recall is the whole point — the schedule just makes the recall happen at the right time.
The Bottom Line
If someone only remembers five things from all of this, it should be these.