Spaced Repetition Explained: the Science of Remembering What You Study

Forgetting is not the enemy — it is the mechanism

In the 1880s Hermann Ebbinghaus measured how quickly he forgot lists of nonsense syllables and produced the first forgetting curve: a steep drop in the hours after learning, then a slow decline. Over a century of replication has confirmed the shape. Left alone, most of what you study today is gone within a week.

The useful discovery came later. Each time you successfully retrieve something as it is fading, the curve flattens — the memory decays more slowly than it did before. Forgetting is not a failure to be prevented; it is the signal that tells you when a review will actually do something.

Why the timing matters more than the volume

Reviewing material you still remember perfectly feels good and teaches you almost nothing, because retrieval was effortless. Reviewing material you have half-forgotten is uncomfortable, slow, and far more durable. This is the spacing effect: the same number of reviews produces substantially better retention when they are spread out than when they are massed together.

That is also why re-reading a chapter five times in one evening is such a poor use of an evening. Five reviews spread across three weeks would cost the same time and leave you with far more.

A schedule you will actually follow

Formal algorithms exist, but the intervals do not need to be precise — they need to expand. A perfectly serviceable pattern for exam preparation:

• Same day you first learn something
• Two to three days later
• About a week later
• About three weeks later
• Once more shortly before the exam

Anything that expands the gap after each success works. What does not work is reviewing everything on a fixed weekly cycle regardless of how well you know it — you will spend most of your time on the material you have already mastered.

Space your practice questions, not just your notes

Spacing is usually described in terms of flashcards, which undersells it. The same principle applies to practice questions, and questions add something flashcards cannot: they force you to recognise a concept from an unfamiliar description, which is exactly what an exam asks you to do.

The practical version: when you get a question wrong, it should come back to you in a few days, and again a couple of weeks later if you get it right. When you get it right twice in a row at increasing intervals, it can leave the rotation. This is what QuizLab's review scheduling does with your own answer history, so the questions you are shakiest on resurface without you having to track them.

Interleave while you are at it

One more finding worth pairing with spacing: mixing topics within a session beats finishing one topic before starting the next. Blocked practice — twenty questions on one domain, then twenty on another — produces better performance during the session and worse retention afterwards. Mixed practice feels messier and lasts longer, because you have to identify which concept applies before you can apply it.

What this looks like on a Tuesday evening

Thirty minutes: ten minutes on questions you got wrong last week, fifteen minutes on a mixed set across everything you have covered so far, five minutes writing one sentence about each mistake. No new material at all. Sessions like that feel unproductive precisely because they are hard — the difficulty is the part that is working.

Three things to take away

Review things as you are starting to forget them, not while you still know them. Expand the interval every time a review succeeds. And apply it to questions rather than only to definitions, because retrieving a concept from a novel description is the skill the exam actually measures.