The 3-Step Active Recall Method
The Levo method
The 3-step active recall method
Every Levo guide is a list of exam-style questions with the answers printed after them. You answer them from memory, check what you wrote, and answer them again. This page sets out the three steps, what happens in memory while you do them, and the research the method is built on.
The three steps
Work on one subtopic at a time. A subtopic is eight to twenty-five questions, which is the amount a person can hold in one pass.
-
Step 1 · Closed book
Cover the answers. Work through every question in the subtopic and write down what you can. Where you have nothing, leave the space blank and move on. Guessing is fine, stopping to look things up is not.
-
Step 2 · Open book
Go back to the first question. Read the printed answer and write it out in full, including the answers you already had right. Writing out the ones you got right is the part most people skip, and it is the part that fixes the exact wording.
-
Step 3 · Closed book again
Same questions, same order, answers covered, from memory. Score it. The distance between what you produced in step 1 and what you produce in step 3 is the only measurement the session needs.
Step 1 will feel bad the first time you work a subtopic. That feeling has no bearing on how much is being learned.
Then come back to it
One session moves a subtopic into memory. Keeping it there needs four short returns, and a return is step 3 on its own.
| Return | What you do | Roughly how long |
|---|---|---|
| Day 1 | Closed book, same subtopic | 5 minutes |
| Day 3 | Closed book, same subtopic | 4 minutes |
| Day 7 | Closed book, same subtopic | 4 minutes |
| Day 14 | Closed book, same subtopic | 3 minutes |
| Day 30 | Closed book, same subtopic | 3 minutes |
Anything you fail on a return goes back to step 2 for that question alone, then rejoins the schedule.
Why answering beats reading
Reading a page again raises how familiar the page feels. Familiarity is a different capacity from production, and the exam asks for production. The research on this is unusually consistent.
Roediger and Karpicke ran the comparison directly in 2006. Students studied prose passages. One group read the passage four times. Another read it once, then spent the remaining three sessions writing down everything they could remember with the text face down. Both groups spent the same total time. Tested five minutes later, the rereaders did better. Tested a week later, the rereaders produced 40 per cent of the passage and the recall group produced 61 per cent.
Karpicke and Blunt repeated the comparison in 2011 against a harder rival, elaborative concept mapping, using science texts of the kind used in teaching. On a short-answer test one week later the retrieval group scored 0.67 and the concept mapping group 0.45, a difference of one and a half standard deviations. The students in that study had been asked to predict their own results beforehand. They predicted that repeated study would serve them best and that practising retrieval would serve them worst.
In 2013 Dunlosky and colleagues reviewed ten study techniques against the whole evidence base and rated each one for general usefulness. Two were rated high utility: practice testing and distributed practice. Rereading, highlighting and summarising were rated low utility. The three-step method is the two high-utility techniques put into one procedure.
Cue-forming
Tulving and Thomson set out the encoding specificity principle in 1973: what is stored alongside a memory at the moment of learning becomes part of what retrieves it later. Godden and Baddeley tested it outside the laboratory in 1975 by having deep-sea divers learn word lists either on the beach or underwater, then recall them in the same place or the other one. Words learned underwater came back best underwater and words learned on the beach came back best on the beach. Matching the two environments raised recall by roughly 40 per cent.
Nobody can sit their revision in the exam hall. Four conditions can be matched, and each one is a cue that will be present on the day.
The question format
The exam puts a question to you and waits for an answer. The method puts a question to you and waits for an answer. Nothing in a Levo guide is laid out as notes or as a chapter, because a chapter cues nothing that the paper will supply.
Writing by hand
You will write the paper by hand, so write the answers by hand. There is also evidence that the act itself changes what is stored. Van der Weel and Van der Meer recorded high-density EEG from university students in 2024 while they wrote words by hand and typed the same words. Handwriting produced widespread connectivity across brain regions in the theta and alpha bands that typing did not produce.
The desk, and silence
Work at a desk, sitting upright, in silence, with the phone in another room. A bed and a playlist are two conditions the hall will not reproduce, so neither can cue anything there.
A clock
Give each pass a time and keep to it. The paper is a timed production task, and timed production is what should be practised.
Small personal markers work the same way and cost nothing: the same pen, the same drink, the same chewing gum at revision and in the hall. Anything permitted in the exam room that you use only while revising becomes another cue.
What the method does not do
The method puts the specification into memory. It does not teach content you have never met, and it does not train exam technique. Past papers do that second job: reading unseen data, structuring a six-mark answer, working to the clock across a whole paper. The order matters, because technique practice on content you cannot recall mostly measures the recall problem.
A workable division is one subtopic by the method, then one exam question on that subtopic from a past paper, marked against the mark scheme.
What a session looks like
Twenty minutes, one subtopic, paper and pen. Nothing else is decided while you work, which is the reason the method survives a bad week: there is no planning step to abandon.
Step 1, closed book, written from memory
Why can a transmission electron microscope resolve detail that a light microscope cannot?
electrons have a shorter wavelength than light
Step 2, open book, written out in full
A beam of electrons has a much shorter wavelength than visible light, so a TEM has a far higher resolution and can distinguish two points that a light microscope shows as one.
Step 3, closed book, twelve minutes later
Electrons have a much shorter wavelength than light, so the TEM resolves two close points separately where a light microscope shows one.
That question is one of 1,833 in the OCR A-Level Biology A guide.
Where the method is used
Every Levo guide is a set of questions with their answers, grouped by topic and subtopic, built to be worked in these three passes.
Questions about the method
What is active recall?
Active recall is producing information from memory rather than reviewing it. Answering a question with your notes shut is active recall. Reading the same page again is not.
Why is rereading not enough?
Rereading raises how familiar a page feels without raising the ability to produce its content cold. Dunlosky and colleagues rated rereading, highlighting and summarising as low utility in their 2013 review of ten techniques, and rated practice testing and distributed practice as high utility.
How many questions should I do at a time?
One subtopic, which is eight to twenty-five questions depending on the guide. All three steps on that subtopic take about twenty minutes.
Why write the answers out in step 2 when I already got them right?
Step 2 is where the wording is corrected. A half-right answer in step 1 usually reappears in step 3 in its half-right form unless the full version has been written in between.
Do I have to write, or can I do it in my head?
Write. You write the paper by hand, so handwriting is one of the four conditions the method matches deliberately. Van der Weel and Van der Meer found in 2024 that handwriting produced widespread brain connectivity that typing did not.
Why does step 1 feel so bad?
Because producing something from memory is harder than recognising it, and the difficulty is the part that works. In the Karpicke and Blunt study the students predicted that repeated study would serve them best and retrieval practice worst, and the result was the other way round.
How often should I come back to a subtopic?
Day 1, day 3, day 7, day 14 and day 30. A return is step 3 on its own and takes three to five minutes. Anything you fail goes back to step 2 for that question alone.
Do I still need past papers?
Yes. The method puts the specification into memory. Past papers train what you do with it: unseen data, extended answers and the clock. Exam boards publish them free.
References
- Roediger, H. L., and Karpicke, J. D. (2006). Test-enhanced learning: taking memory tests improves long-term retention. Psychological Science, 17(3), 249 to 255.
- Karpicke, J. D., and Blunt, J. R. (2011). Retrieval practice produces more learning than elaborative studying with concept mapping. Science, 331(6018), 772 to 775.
- Dunlosky, J., Rawson, K. A., Marsh, E. J., Nathan, M. J., and Willingham, D. T. (2013). Improving students' learning with effective learning techniques. Psychological Science in the Public Interest, 14(1), 4 to 58.
- 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.
- Tulving, E., and Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80(5), 352 to 373.
- Godden, D. R., and Baddeley, A. D. (1975). Context-dependent memory in two natural environments: on land and underwater. British Journal of Psychology, 66(3), 325 to 331.
- Van der Weel, F. R., and Van der Meer, A. L. H. (2024). Handwriting but not typewriting leads to widespread brain connectivity. Frontiers in Psychology, 14.
- Ebbinghaus, H. (1885). Uber das Gedachtnis. The original forgetting curve.