Two techniques account for most of the difference between people who retain what they study and people who don't. Both have decades of evidence behind them. Both feel worse to use than the methods they replace, which is the main reason they stay unpopular.
What are spaced repetition and active recall?
Active recall means retrieving information from memory without looking — testing yourself rather than reviewing. Spaced repetition means revisiting that material at increasing intervals over time. Used together, they produce substantially better long-term retention than re-reading or highlighting, for the same time invested.
The illusion that wastes study time
Re-reading produces fluency: the text becomes familiar, processing feels smooth, and your brain reads that ease as knowledge. It isn't. Recognising material when it's in front of you is a different skill from producing it when it isn't.
Roediger and Karpicke found exactly this pattern — repeated studying increased students' confidence in their ability to remember while producing worse retention on delayed tests than testing did. Confidence and competence moved in opposite directions.
So the discomfort of active recall isn't a sign it's going badly. Struggling to retrieve something is the mechanism working; smooth familiarity is the thing that fools you.
How to do active recall
- Read a section, then close it. Write down everything you can remember, from memory.
- Check what you missed and note the gaps specifically.
- Turn material into questions rather than summaries — questions force retrieval, summaries permit copying.
- Answer before revealing. Always attempt it, even when you're fairly sure you don't know.
- Explain it aloud as though teaching someone. Gaps in understanding surface immediately.
- Prefer producing over recognising. Short-answer beats multiple choice, because recognition is easier than recall.
Getting an answer wrong then correcting it is genuinely productive — the failed attempt appears to prepare the memory for the correction.
How to space it
| Review | When |
|---|---|
| First | Later the same day or the next day |
| Second | Two to three days later |
| Third | About a week later |
| Fourth | Two to three weeks later |
| Then | Monthly, until it's solid |
The intervals are approximate. The principle is to review as the memory begins to fade but before it's gone — a recall that takes effort strengthens memory more than one that's instant. If it's effortless, space it further out. If you've forgotten entirely, shorten the gap.
Note that the research is mixed on whether expanding intervals genuinely beat equally spaced ones. Spacing clearly beats cramming; the precise schedule matters much less than people assume, so don't over-engineer it.
Combining them
The two work as one system: each spaced review should be a retrieval attempt, not a re-read. That's what flashcard software automates — it asks you to produce the answer and schedules the next appearance based on how it went.
The tool is optional. A stack of index cards, or a list of questions with dates beside them, achieves the same thing. What matters is retrieving, and spacing it out.
Two mistakes worth avoiding
Making cards you don't understand. Memorising a sentence you can't explain produces a card you'll fail repeatedly and resent. Understand first, then commit to memory.
Letting the system become the work. Elaborate colour-coded decks are a comfortable form of procrastination — they feel productive while deferring the uncomfortable retrieval. Fewer cards, actually tested, beat a beautiful deck you avoid.
Protect your sleep around all of this. Consolidation happens overnight, so a shorter session followed by decent sleep beats a longer one that eats into it — more in how to improve memory.
Frequently asked questions
How long should sessions be? Short and frequent beats long and rare. Twenty to thirty focused minutes with real breaks is more productive than three unbroken hours.
Does this work for skills, not just facts? The principles transfer — practice testing and spacing help procedural learning too, alongside interleaving different problem types rather than blocking one kind.
Do I need an app? No. Software handles scheduling for large volumes of material; paper cards or a question list work fine for smaller amounts.
Learning well depends on the same foundations as everything else — sleep, attention, and a nervous system that isn't overloaded. Finding Steady Ground covers those in the mental wellbeing guides. Related: how to improve focus.
Source: Roediger & Karpicke, Test-Enhanced Learning, Psychological Science (2006).