
Study and Learning
The Spacing Effect (Cepeda 2006) Explained
Spreading your reviews out beats cramming, and the further away you need something, the wider the gaps can be. What Cepeda and colleagues found in 2006 and 2008, and how to fit your reviews to the date that matters.
One memory palace in my course videos (a memory palace is a place you picture and fill with images of what you want to remember) is a room I generated with AI, styled like the Stranger Things living room (LOCI can now build one for you with AI). On its TV, a stream of words is being sucked into the screen: spaced retrieval practice. That palace I never clear. Others I load for a meeting and let fade. This one I come back to until I simply know it.
That difference is the spacing effect at work. The spacing effect is the finding that spreading your study or your reviews out over time beats cramming the same amount into one sitting, and that the longer you need to remember something, the wider the gaps between reviews can be. The two papers that pin this down best come from Cepeda and colleagues, in 2006 and 2008.
Here is what each one did and found, followed by how to use it.
Why it matters
Cramming feels efficient. One long session, job done, and the material feels fresh when you stand up. You have probably also heard that you should "space it out". What you are rarely told is how far apart. Every few hours? Every few days? Once a month?
That is the question that matters in real life, because you are never learning something for no reason. You have a pitch on Friday, an exam in May, a part to play in six weeks, or a subject you want for the rest of your career. The research says the answer depends on that date. That one idea is the most useful thing to take from the research.
The research: Cepeda and colleagues, 2006
- In short: Pooling hundreds of experiments, spreading study out beat bunching it together, and the ideal gap grew with how long the material had to last.
- What they did: A meta-analysis, which is a study of studies. They gathered experiments that compared massed practice, all in one block, with spaced practice, the same amount split across sessions. Those experiments covered a wide range of gaps between sessions and of delays before the final test.
- Who was involved: 839 assessments drawn from 317 experiments, reported in 184 articles, all on verbal learning.
- What they found: Spaced practice beat massed practice. The best gap between sessions was longer when the final test was further off.
- What they concluded: There is strong support for spreading practice out, and there is no single magic interval that suits everything.
The research: Cepeda and colleagues, 2008
- In short: The best gap before you review something depends on how far away the test is.
- What they did: People learned new material, came back for one review after a gap of anything up to three and a half months, and then sat a final test as much as a year later. The researchers varied both the gap and the delay to the test.
- Who was involved: Over 1,350 people.
- What they found: Every test delay had its own best gap. As a share of the time until the test, that gap was about 20 to 40% when the test was a week away, and it fell to about 5 to 10% when the test was a year away.
- What they concluded: Spacing helps at every scale, and the ideal gap grows as the time you need to remember grows, though not in proportion.
To make those percentages concrete: 20 to 40% of a week is roughly a day and a half to three days. 5 to 10% of a year is about two and a half to five weeks. So a longer horizon means a longer gap, but a smaller share of the total.
The first paper is a large meta-analysis of an effect that has been replicated for a very long time, so the core message, space it out, is about as solid as learning science gets. The second is one big study. Its exact figures are the best guide we have to gap size, but they come from that single study, so treat them as a guide rather than a law. A version you will see online, that the optimal gap is "10 to 20%" of the delay, is a popular simplification. It is not what the paper says.
What I take from it
A widening review pattern that has been in circulation for decades goes one hour, one day, one week, one month, three months, six months, and then as needed. For lines, a shorter version is common: an hour, a day, three days, a week. Both are good places to start. But be clear what they are. They are rules of thumb built on the spacing principle. No study tested those particular intervals. Researchers still debate whether gaps should get longer each time or stay even; growing gaps are used here because they fit the way people actually live.
What the 2008 study gives you is a better first question. Not "what is the schedule?" but "when do I need this?" It is the same question to ask when you design a palace: what is the goal? If it is a reusable palace for a meeting this afternoon, let it fade afterwards and plan no reviews at all. If it is a permanent palace, like the Stranger Things room, the reviews keep going, further and further apart, until you know it.
Two more things from practice, not from these papers. When something comes back easily, leave a longer gap before the next review. When it does not, bring it back sooner. And how strong the first picture was matters: a vivid image placed on a clear location needs fewer reviews than a vague one.
One caution. The 2008 study looked at a single review. Real life usually has several. Use its figures to set the size of the first big gap, then stretch from there.
What it does not show
- It does not prove any particular schedule. Not one hour, one day, one week, one month, and not the one in your favourite app. The principle is proven; the exact intervals are judgement.
- It does not prove that growing gaps beat even ones. That is still argued over.
- It is mostly about words. The 2006 review pooled verbal learning experiments. That is a lot of what people study, but it is what was tested.
- Spacing is not rereading on a timer. Each review should be a recall with the book closed, then a check. The case for that comes from Roediger and Karpicke, broken down in The Testing Effect (Roediger and Karpicke 2006) Explained.
- Spacing inside one lesson is a different finding. Kelley and Whatson's short-bursts-and-breaks hour has its own article: Spaced Learning: The One-Hour Lesson That Matched Four Months.
How to use it this week
The practical heart of this is to fit the gaps to the date. Here is how.
- Name the date. Write down when you need this, and whether it is reusable (needed once, then it can fade) or permanent (needed for good).
- Encode it well first. Understand it, then turn the key ideas into strong pictures. Spacing works on what you put in, so put in something worth keeping.
- Make the first review early. An hour later, then the next day. That part of the schedule rarely changes.
- Size the next gap to the date. Use the 2008 figures as a rough guide. These examples are one reading of them, not figures from the paper:
- A pitch on Friday, and it is Monday: recall it Monday evening, again Wednesday, and once on Friday morning.
- An exam five weeks away: after the first day, gaps of about a week, then a final recall a couple of days before.
- A professional qualification next year: let the gaps grow to several weeks once the material is solid.
- A subject for your career: keep going to months, and keep the palace permanent.
- Make every review a recall. Book closed, say or write what you can, then check.
- Adjust as you go. Easy recall earns a longer gap. A blank earns a shorter one and a stronger picture.
- Put it in the diary. Spacing only works if the reviews actually happen, so book them like meetings.
If you want the scheduling done for you, LOCI has spaced retrieval practice built in. LOCI turns your material into key ideas and images, places them on a palace it builds for you from Google Street View or with AI, and gives each image a flashcard with the real information on the back.
To go further, The Testing Effect (Roediger and Karpicke 2006) Explained covers what each review should look like. Spaced Learning: The One-Hour Lesson That Matched Four Months shows spacing inside a single lesson. The Forgetting Curve (and How to Beat It) shows what the gaps are fighting, and Sleep and Memory: Why the Night Does Half Your Learning covers what happens overnight between them.
Memory map
- The spacing effect: spread reviews out rather than cramming them. Image: a pile of books toppling off one desk beside the same books spread neatly along a sunny shelf.
- A permanent palace gets reviewed until it is locked in. Image: a cosy eighties living room where a stream of tiny objects is being sucked into an old television set.
- A reusable palace can be loaded for a meeting and left to fade. Image: a chalk drawing on a pavement softening in light rain.
- Across 317 experiments, spaced practice beat massed practice. Image: a long banquet table covered in hundreds of identical cakes, every slice cut the same way.
- There is no single magic interval. Image: a magician lifting his hat to find it empty.
- Fit the gap to the date: about 20 to 40% for a week away, 5 to 10% for a year away. Image: a short footbridge and a long suspension bridge, each lit by evenly spaced lamps.
- A longer horizon means a longer gap, but a smaller share of the total. Image: a mountain path with a few wide steps cut into a very long slope.
- The one hour, one day, one week schedule is a long-standing rule of thumb, not a tested schedule. Image: a carpenter measuring a plank with his thumb.
- Easy recall earns a longer gap; a blank earns a sooner one. Image: a gardener watering the wilting plant and walking past the healthy one.
- Every review is a recall, book closed. Image: a student looking up at the ceiling, eyes narrowed, a shut book under one hand.
- Book the reviews like meetings. Image: a wall calendar with small coloured stickers placed further and further apart.
Sources
- Mark Channon, How to Remember Anything (Hodder Education)
- Mark Channon, Learning Your Lines (Nick Hern Books)
- Cepeda, N. J., Pashler, H., Vul, E., Wixted, J. T., and Rohrer, D. (2006). Distributed practice in verbal recall tasks: a review and quantitative synthesis. Psychological Bulletin, 132(3), 354 to 380. https://doi.org/10.1037/0033-2909.132.3.354
- Cepeda, N. J., Vul, E., Rohrer, D., Wixted, J. T., and Pashler, H. (2008). Spacing effects in learning: a temporal ridgeline of optimal retention. Psychological Science, 19(11), 1095 to 1102. https://doi.org/10.1111/j.1467-9280.2008.02209.x
- Roediger, H. L., III, and Karpicke, J. D. (2006). Test-enhanced learning: taking memory tests improves long-term retention. Psychological Science, 17(3), 249 to 255. https://doi.org/10.1111/j.1467-9280.2006.01693.x (named in passing; broken down in The Testing Effect (Roediger and Karpicke 2006) Explained)
- Kelley, P., and Whatson, T. (2013). Making long-term memories in minutes: a spaced learning pattern from memory research in education. Frontiers in Human Neuroscience, 7, 589. https://doi.org/10.3389/fnhum.2013.00589 (named in passing; broken down in Spaced Learning: The One-Hour Lesson That Matched Four Months)
Loci
Build the memory palace in this article inside Loci — store your locations, link images, and rehearse with spaced repetition.
Try it in LociMemory School community
Learn directly with Mark inside the Memory School community on Skool — monthly lessons, practice, and Q&A.
Join Memory SchoolAbout the author
Mark Channon
Grand Master of Memory (1995). Author of three books on memory published by Hodder and Nick Hern. Creator of BBC1's Monkhouse's Memory Masters. Memory coach on Channel 4's Can I Improve My Memory?
Read full bio →