The Spacing Effect: Why Cramming Fails (and What to Do Instead)

It’s the night before an exam, and the textbook is open at page one of a chapter you haven’t touched since September. The coffee is made. You’re going to get through all of it, and by about 3 am, a lot of it is going to be in your head.

And in the morning, a lot of it is. You sit the exam, you do fine, and you walk out feeling like cramming is a legitimate strategy. Then a fortnight later someone asks you something from that chapter and you draw a blank. Where did it go?

It went where all hastily stored knowledge goes. And the reason is one of the best-established findings in the whole science of learning.

The curve everybody follows

In the 1880s, the German psychologist Hermann Ebbinghaus did something that sounds miserable: he spent years memorising lists of nonsense syllables like WID and ZOF, then testing himself at different delays to see how much had gone. He chose nonsense so that nothing he already knew could help.

What he found was a steep, regular curve. A large share of what he’d learned was gone within an hour. More was gone by the next day. After that the decline slowed to a long, thin tail. In 2015, Murre and Dros repeated the experiment with a single modern volunteer and got a very similar shape, which is rare in psychology for something this old.

He found something else, and this one is more useful. When he went back and relearned a list, it took less effort the second time, and the relearned list faded more slowly. Each return to the material flattened the curve.

The spacing effect

Put those two things together and you get the spacing effect: the same amount of study, spread across days, leads to much better memory than the same amount packed into one sitting.

It’s not a one-off result. In 2006, Nicholas Cepeda and colleagues pooled 317 experiments on spacing and found the benefit again and again. It shows up with words, facts, maths, motor skills and foreign vocabulary, in children and adults, in labs and in real classrooms.

One of the longest tests of it was a study by Harry Bahrick and his family. They learned foreign-language vocabulary with different gaps between sessions, then tested themselves years later. The words learned with longer gaps were remembered better, at the end of the experiment as well as at the start. Same total study time, a very different outcome.

Why a bit of forgetting helps

The most convincing explanation is also the most surprising: forgetting is part of the mechanism.

When you’ve just studied something, it’s fresh. Recalling it takes almost no effort, and recalling it effortlessly barely strengthens the memory. When you come back after a gap, you’ve forgotten part of it, so you have to work to dig it out. That working is what builds a deeper memory.

The psychologist Robert Bjork calls this a desirable difficulty: a condition that makes learning feel harder in the moment and pays off later. Spacing is the classic case. It slows you down, it makes you feel less sure, and it works.

There’s a second reason. When you return to something on a different day, you’re in a slightly different mood, place and state of mind. The memory gets linked to more contexts, and that gives you more ways back to it when you need it.

Why cramming feels better than it is

If spacing works so well, why does everyone still cram? Because cramming has three powerful illusions on its side.

It works soon after. If the test is tomorrow, cramming can get you through it. The cost doesn’t arrive until next week, next term, or in the later course that assumes you remember this one. Because the cost is delayed, you never connect it to the cause.

Familiarity feels like knowledge. After three hours on the same chapter, every page looks familiar. Familiar and recallable aren’t the same. You can recognise an answer and be unable to produce it, as anyone knows who has stared at a blank exam paper thinking I know this.

It feels like hard work. A long, intense session feels like effort, and effort feels like it must be paying off. Spaced study feels like constant half-forgetting, which feels like failing.

Nate Kornell tested exactly this. People studied flashcards either massed or spaced, then were tested. About nine in ten did better with the spaced cards. Yet most said they believed cramming had worked as well or better. We’re bad judges of our own learning, and the methods that work best tend to feel worst.

How to space your study

You don’t need a perfect schedule. Any spacing beats none. Here’s a simple one that works for most things:

1. Start earlier, study in shorter bursts. Four one-hour sessions across two weeks beat one four-hour session the night before.
2. Review the day after you first learn it. That’s when forgetting is steepest, so a short review then does the most good.
3. Come back after three or four days, then a week or two later. Stretch the gaps as the material sticks.
4. Test yourself, don’t reread. Try to recall before you look at your notes. Spacing works best together with retrieval practice.
5. Make a card for anything that matters. Flashcards, paper or app, are a simple way to space many small facts at once. Each time you get one right, it moves to a longer gap; each time you miss, it comes back sooner.
6. Sleep in between. Sleep is when much of the day’s learning gets consolidated. An all-nighter trades that away.

How big should the gaps be?

It depends on how long you need to remember something, and Cepeda’s team ran a large study to find out. They taught people facts and tested them after delays from a few days to a year, with different gaps in between. The best gap depended on the delay: for a test a week away, about a day; for a test a year away, a few weeks.

A workable rule of thumb is a gap of roughly 10 to 20% of how long you need the memory to last. For an exam in a month, gaps of three or four days. For something you want to know for a year, gaps of several weeks. Go too short and you’re cramming again. Go too long and you’ve forgotten so much that you’re relearning from scratch, which is a lot less efficient. But the penalty for going too long is smaller than the penalty for going too short, so if in doubt, leave a bit more space.

What if the test really is tomorrow?

Then cram, and be honest about why. If you’ll never need the material again, cramming is a fair trade of tomorrow’s score for next month’s memory. It’s when you’ll need the knowledge for longer that it costs you, such as a course that builds on itself, a professional exam, or any language, instrument or skill.

If you’re cramming, do it well: test yourself rather than rereading, prioritise the topics most likely to come up, and sleep. Sleep often does more for your score than a few extra hours of reading.

Where this shows up in Cognosc

Cognosc is built on the spacing effect. After you learn an idea, it checks back with a single question, timed for just before you’d be likely to forget it. Each time you get it right, the next check comes further out, and each time you miss, it comes back sooner. It’s the Leitner box with the scheduling done for you, and it’s why a short session now and then beats a marathon.

Check your study habits

The how learning works test asks about cramming, rereading and many other study habits, and says what the research found about each. If you want to see the curve for yourself, the forgetting curve article has a toy where you set your own review days.