What you cram the night before an exam vanishes almost as soon as the exam is over. Yet material you reviewed across several sessions stays with you for months. Where does that difference come from?
The answer lies not in how many times you review, but in the intervals between reviews. Learning science calls this the spacing effect, and it is one of the most thoroughly replicated findings in all of psychology.
Cramming versus spacing: what is the difference?
Even with the same total study time, how you distribute it changes the outcome.
- Massed practice (cramming): reviewing the same material three times in a row
- Spaced practice: reviewing the same material three times, with gaps in between
In 2006, a research team led by Cepeda analyzed hundreds of experiments spanning the previous 100 years and confirmed that the groups that spaced out their study consistently scored higher on later tests. The pattern held regardless of the type of material: vocabulary, prose, mathematics, or motor skills.
The counterintuitive part is that if you look only at the moment right after studying, cramming feels more effective. Repeat something three times in a row and by the third time you answer fluently. But that fluency is short-lived and breaks down within a few days. Spaced practice is harder, because each review demands an effort to recall, but that very difficulty is what builds long-term retention.
Why "just as you start to forget" works
The key is what happens in the brain during a review.
When you review after a gap, the memory is still there, but retrieving it takes some effort. The leading explanation today is that recalling something in this effortful state strongly strengthens the memory (this is known as a "desirable difficulty").
The harder you work to recall a memory, the more strongly it stays. The easier the review, the less of it stays.
Tracing over the answer while looking at it is easy, but for the brain it is not practice at retrieval. Each time you pull a half-forgotten memory back out on your own, the pathway to it gets stronger. That is why the best time to review is just before you would forget completely, a little before the forgetting curve bottoms out.
- Reviewing too early: you still recall it easily, so the strengthening effect is weak
- Reviewing too late: you have forgotten it entirely, so it is almost like learning it again
- Reviewing as you start to forget: maximum strengthening for minimum time
Recalling is itself a form of learning
The other pillar alongside the spacing effect is the testing effect (also called retrieval practice).
In a 2006 experiment, Roediger and Karpicke compared a group that repeatedly reread a passage with a group that took a recall test after reading it. On an immediate test the rereading group came out ahead, but one week later the result had completely reversed, and the group that practiced recall outperformed by a wide margin.
In other words, reviewing does not mean looking over the material again. It means trying to recall what it said without looking at it. Close your notes and see whether you can state the key points, cover the meaning of a word and answer, rework a problem: any format is fine, but whether a recall step is included makes the decisive difference.
Why the intervals keep getting longer
With each review, the forgetting curve becomes flatter (this is explained in detail in the forgetting curve article). In other words, the length of time a memory can hold up grows with every review.
If so, it makes sense for review intervals not to be fixed but to expand step by step.
- First review: the next day (to stop the initial steep drop)
- Second review: a few days to a week later
- Third review: a few weeks later
- Fourth review onward: a month or more
In early research, these expanding intervals were used as a rule of thumb, but a later large-scale study (Cepeda et al., 2008) found something interesting: the optimal review interval depends on how long you want to remember the material. If the exam is a month away, every few days; if you want to retain it for a year, every few weeks. The further away the target date, the wider the intervals that proved efficient.
This has a practical implication: it makes the most sense to build your review plan by working backward from your exam or target date.
A system you can sustain beats perfect intervals
It is worth adding that there is nothing magical about the specific numbers "1 day, 1 week, 1 month." This is not so delicate that being a day off ruins everything. What the research shows is how robust the principle is: recall material roughly as you start to forget it, at expanding intervals.
In real study, what actually breaks a plan is not imprecise intervals but a breakdown in managing them. Once you have 50 or 100 items, keeping track in your head of which to review when is close to impossible, and this is where many people give up on spacing and go back to cramming.
The principle is simple. What separates people who keep going from those who don't is whether they have a system that keeps it running every day without breaking down, one where the items due for review today automatically appear in front of them.
Summary
- For the same study time, spacing out your reviews beats doing them all at once (the spacing effect, confirmed by 100 years of research)
- Make reviews recall practice, not rereading (the testing effect). Don't be fooled by how well it seems to go right afterward
- The best time is just as you start to forget. Reviewing while recall is still easy is inefficient
- Review intervals are not fixed but expand step by step. The further away the target date, the wider the intervals
- What determines results is less the precision of the numbers than a system that keeps running without breaking down