Why You Forget What You Revised Last Week
Every GCSE student knows the experience: you spend an evening on circle theorems, feel completely on top of them, and two weeks later a mock question on the same material draws a blank. Nothing went wrong with the original revision. What you ran into is the normal behaviour of human memory, and it was described systematically by the psychologist Hermann Ebbinghaus in the nineteenth century.
Ebbinghaus studied how quickly newly learned material fades, and the shape he found is now called the forgetting curve. Memory for new material drops away steeply at first, then the decline flattens: most of what you will forget, you forget early, and what survives the early period fades much more slowly. Crucially, the curve is not fixed. Each time you successfully revisit material, the subsequent forgetting is slower, and the fade takes longer to set in.
That single fact is the entire case for spaced repetition. If forgetting is fastest soon after learning, then the highest-value moment to review a topic is before the fade has done its damage, and each successful review buys you a longer gap before the next one is needed. Revision that ignores this, covering each topic once in a long block and never returning, is working against the shape of your own memory. Revision that works with it gets dramatically more retention from the same number of hours.
- Forgetting is steepest soon after learning, then levels off
- Ebbinghaus described this shape: the forgetting curve
- Each successful review slows the subsequent forgetting
- Reviewing before the fade beats re-learning after it
The Testing Effect: Recall Beats Re-Reading
Spacing answers the question of when to review. A second, equally well-established finding answers the question of how. The testing effect is the observation, replicated across decades of memory research, that retrieving information from your own head strengthens memory far more than re-exposing yourself to it. Attempting a question, or recalling a method from nothing, does more for retention than re-reading the notes that contain it, even when the re-reading feels smoother and more productive.
That feeling is the trap. Re-reading produces fluency, the comfortable sense of recognising material, and recognition is precisely what a maths exam never asks for. The paper hands you a blank space and a question; the only thing that counts is whether you can produce the method unprompted. Retrieval practice rehearses exactly that act, which is why it transfers to exam performance in a way that highlighting and note-copying do not.
The two principles multiply each other. A review session that consists of attempting questions closed-book is worth far more than one spent re-reading, and a retrieval attempt made after a gap, when the memory has started to fade and recall takes genuine effort, strengthens the memory more than an easy recall made minutes after learning. Effortful recall at a well-timed interval is the single most efficient revision act available to you. Everything else in this guide is arranging your schedule so that it happens on purpose.
- Retrieving from memory strengthens it more than re-exposure
- Re-reading builds recognition; exams demand production from a blank page
- Effortful recall after a gap strengthens memory more than easy recall
- Spacing and retrieval practice compound each other
What Spaced Repetition Looks Like in Practice
Strip away the theory and spaced repetition is a simple scheduling rule: after learning something, review it at increasing intervals, with each successful review earning a longer gap before the next.
Concretely, suppose you cover ratio sharing properly on a Monday: worked examples, then a set of questions closed-book. Under a spaced schedule you would return to it briefly the next day or the day after, ten minutes of questions, no notes. Recall it well, and the next visit moves out to perhaps a week later. Succeed again and the gap stretches to a fortnight, then towards a month. Each visit is short, because you are maintaining a memory rather than rebuilding one, and each success pushes the topic further into the background of your plan.
The schedule also self-corrects. If a review goes badly and the method has genuinely faded, the topic drops back to short intervals: revisit tomorrow, then next week, and rebuild from there. Struggling on a review is not the system failing; it is the system finding a weak memory while there is still time to repair it, which is infinitely better than a mock finding it for you.
Compare the usual alternative, massed practice: three hours on ratio in one evening, then nothing for six weeks. The three-hour version feels more thorough on the night, and the fortnight-later result is usually indistinguishable from not having revised at all. The spaced version spends less total time and leaves a method you can still perform in the exam hall.
- Review at increasing intervals: next day, a week, a fortnight, a month
- Successful recall earns a longer gap; failed recall resets to short gaps
- Reviews are short because maintaining beats rebuilding
- Massed practice feels thorough and fades fastest
Intervals on a Real GCSE Timeline
Abstract intervals become much more useful when you pin them to the actual school calendar. GCSE Maths is a two-year course with exams in May and June of Year 11, and the right spacing looks different at each stage.
During Year 10 and the autumn of Year 11, while new content is still arriving, the priority is making sure each newly taught topic gets its early reviews: a short retrieval session within a couple of days of first meeting it, another about a week later, then a return visit two to four weeks on. This is also the period to start cycling back through old topics, because material from the start of Year 10 has had the longest to fade and is exactly what mocks expose.
From January of Year 11, the mix shifts. Little content is new, so revision becomes almost entirely review, and your intervals should now be driven by weakness rather than recency: shaky topics on short cycles of a few days, secure topics on long cycles of several weeks, checked briefly to confirm they are still there. Full past papers, sat weekly or so, act as honest audits that tell you which topics belong on which cycle.
In the final month, the gaps compress across the board. Weak topics appear every few days, strong ones every week or two, and daily mixed practice keeps everything warm. Done this way, exam week arrives with every topic recently retrieved, and nothing relying on a memory from months ago. How long each daily session should be is its own question, and our guide to how long to revise each day covers it properly.
- While content is new: review within days, then a week, then a month
- From January of Year 11, cycle length follows weakness, not recency
- Weak topics on short cycles, secure topics on long confirming cycles
- In the final month, compress all intervals and mix topics daily
Applying It to Topics, Flashcards and Past Papers
Spaced repetition is usually explained with vocabulary flashcards, and maths needs a translation, because a maths topic is a skill rather than a fact. The unit you schedule is the topic, and a review means attempting questions on it closed-book, not looking at it.
For topics, keep the mechanics light: a list of your topics with, next to each, the date you last practised it and how it went. Each revision day, pick two or three topics whose return date has arrived, and spend fifteen to twenty minutes per topic on questions without notes. Mark them, fix the errors, record the result, and set the next return date longer or shorter depending on how it went. Mixing two or three topics per session is deliberate: interleaving is slightly harder in the moment and produces more durable learning, as well as mirroring a real paper.
Flashcards still earn a place for the factual layer of maths: exact trigonometric values, index laws, the quadratic formula and the other formulae not on your board's sheet, circle theorem statements, key definitions. Review the deck on the same expanding schedule, and always answer before you flip.
Past papers are the top of the pyramid. A full paper is a spaced retrieval test over every topic at once, and its marking tells you which topics need their intervals shortened. In the final months, papers become the engine of the whole system: sit one, and let its dropped marks dictate the next week's short-cycle topics. The full revision guide shows how this loop fits into a complete plan.
- Schedule topics, and make every review a closed-book question attempt
- Two or three topics per session: interleaving mirrors the real paper
- Flashcards cover the factual layer: formulae, exact values, definitions
- Past papers audit every topic at once and reset your intervals honestly
How Exam Ladder's Smart Session Automates the Intervals
The honest weakness of do-it-yourself spaced repetition is the bookkeeping. The method asks you to track, for 70-odd topics, when each was last practised, how confidently, and when it is due back, and to recompute that every time a review goes well or badly. Plenty of students start a tracking sheet in September; very few are still updating it in March. The schedule quietly collapses back into revising whatever feels most urgent, which is usually whatever was revised most recently.
This is the specific problem Exam Ladder's Smart Session exists to solve. The platform records every question you attempt across every GCSE Maths topic, and holds a confidence score per topic that rises with correct, consistent answers and falls with errors and with time since you last saw the material. When you start a Smart Session, it selects questions for you: topics whose review is due, weighted towards where your confidence is lowest, with secure topics resurfacing at longer intervals to confirm they are holding. Answer well and a topic's next appearance moves further out; struggle and it comes back sooner. That is the expanding-and-contracting interval schedule from this guide, run automatically on your actual results instead of your guesses.
Around it sit the same layers described above: generated practice questions with worked solutions for the fix step, over 280,000 of them across the four boards (AQA, Edexcel, OCR and Eduqas), plus 276 real past papers with mark schemes and mock exams graded on real exam-board grade boundaries for the auditing step. You can run the whole method with a notebook and a pile of papers, and it works. The automation's contribution is that the intervals keep being computed in March, when the notebook version has usually been abandoned.
- Manual spaced repetition fails at the bookkeeping, not the theory
- Smart Session tracks per-topic confidence across every topic
- Due and weak topics surface sooner; secure topics return at long intervals
- Papers and mocks graded on real boundaries audit the schedule honestly
The Mistakes That Break a Spacing System
A few predictable errors undo most attempts at spaced revision, and knowing them in advance is the cheapest protection.
The first is reviewing by re-reading. Opening your notes on a due topic, nodding along, and marking it reviewed satisfies the schedule while skipping the retrieval that gives the schedule its value. A review must start with a closed book and a question. The second is only revisiting favourite topics. Left to feelings, everyone reviews what they are good at, because successful recall is pleasant and struggle is not; but the marks live in the topics you avoid, and a spacing system exists precisely to force those back in front of you. If your system never makes you do something uncomfortable, it has stopped working.
The third is abandoning the schedule after missing a few days. Spacing is forgiving: a topic reviewed two days late still benefits enormously compared with a topic never reviewed. Pick the schedule back up from today rather than declaring it failed. The fourth is starting so late that spacing has no room to work. Expanding intervals need weeks to expand into, which is why beginning in the autumn of Year 11 beats any quantity of Easter cramming: the same hours, spread earlier, simply produce more durable memory.
None of this requires perfection. A rough spacing system followed for months will comfortably beat an elegant one followed for a fortnight, and both beat massed cramming. Start earlier than feels necessary, review with a closed book, let struggle steer the schedule, and let the intervals do the compounding.
- A review means closed-book questions, never note re-reading
- Comfortable topics volunteer themselves; weak ones must be forced in
- A late review is still valuable: resume the schedule, never restart it
- Spacing needs weeks to compound, so starting early is the biggest lever
Frequently asked questions
What is spaced repetition in simple terms?
It is reviewing material at increasing intervals instead of all at once: shortly after learning it, then about a week later, then a fortnight, then a month. Memory fades fastest soon after learning, so each well-timed review interrupts the fade, and each successful recall means the next review can wait longer. The same total revision time, spread out this way, produces far more durable memory than one long session.
Does spaced repetition actually work for maths, or just for memorising facts?
It works for maths, with one adjustment: the thing you space is the skill, not a fact card. A review means attempting questions on a topic closed-book, so you are retrieving the method itself. Maths methods fade like any other memory, which is why a topic that felt secure in October can fail in a January mock. Flashcards still help for the factual layer, such as exact trig values and formulae not on the formulae sheet.
What intervals should I use for GCSE revision?
A practical default: review a topic within a day or two of first learning it, again about a week later, then a fortnight, then around a month, making each review a short closed-book question session. If a review goes badly, drop back to short gaps and rebuild. In the last month before the exams, compress everything, with weak topics every few days and secure ones weekly. The exact numbers matter less than the expanding pattern and the honesty of the reviews.
Is it too late to start spaced repetition a few months before my GCSEs?
No. The gains are biggest when you start early, but even a few weeks allows two or three spaced, closed-book visits to each weak topic, which comfortably beats a single long cram on each. With limited time, prioritise: put your weakest high-mark topics on short cycles, confirm strong topics briefly, and use past papers to decide what deserves the remaining slots. Spread hours always beat massed hours, whatever the total.
How does Exam Ladder's Smart Session use spaced repetition?
Smart Session keeps a confidence score for every GCSE Maths topic, based on your answers and on how long it has been since a topic last appeared. When you start a session it serves questions from topics that are due for review, weighted towards your weakest, and pushes confidently answered topics out to longer intervals. It is the expanding-interval schedule from this guide, computed automatically from your real results.