Feynman Technique for Exam Revision: A 4-Step Method

The Feynman technique is a revision method in four steps: pick one topic, write an explanation of it in plain language as if you were teaching a beginner, mark every point where the explanation breaks down, then go back to the source, repair those points and simplify again. The gaps you find are the study list.

It earns its place in a revision plan because writing a full explanation from memory exposes something reading cannot. You can follow a chapter line by line and still be unable to say why the method works, and the Feynman pass is where that difference becomes visible.

What is the Feynman technique?

The Feynman technique is a self-explanation routine named after the physicist Richard Feynman, who was known for insisting that an idea he could not explain in ordinary language was an idea he did not properly understand. The method turns that standard into a repeatable revision task.

The mechanism it exploits has a name worth knowing: the illusion of explanatory depth, the common gap between how well people feel they understand a mechanism and how much of it they can actually produce when asked to explain it. Familiar notes feel like knowledge. An explanation written with the notes closed is where the feeling gets tested, which is also why the method overlaps with active recall rather than rereading.

What are the four steps in practice?

Work on one topic at a time, with a timer and a blank page.

1. Write the topic at the top of a blank page. Keep it narrow: "why inflation erodes real returns" rather than "monetary policy". A unit you cannot explain in roughly a page is two units.
2. Explain it from memory, in plain language, to an imagined beginner. Close every source. Write in full sentences rather than bullets, because bullets let you skip the connective reasoning that is exactly what you are testing.
3. Mark every breakdown as you go. Underline any term you used without being able to define it, any step where you wrote "and therefore" without knowing why, and any place you slowed down or changed the subject. Do not stop to look things up; marking and looking up are separate steps for a reason.
4. Return to the source only for the marked points. Read narrowly, fix the specific gap, then rewrite the whole explanation without looking. The second draft should be shorter and should contain an analogy or example you chose yourself.

The fourth step is the one most people drop, and it is the one that does the work. Reading the answer closes a gap in recognition only; rewriting the explanation is what confirms the repair held.

A worked example: one pass on a topic you think you know

Take opportunity cost. The first draft often reads: "Opportunity cost is the cost of an opportunity you did not take." Written out, that sentence is visibly circular, and the circularity is the finding.

Marking the breakdowns gives three specific questions: what exactly is the cost measured against, is it every alternative or only one, and does money already spent count? The repair pass fixes each. Opportunity cost is the value of the best alternative forgone, not the sum of all alternatives; it is measured in whatever the best alternative would have produced, not only in money; and money already spent is a sunk cost, which is a different idea that does not belong in the explanation.

The rewrite then reads: "Opportunity cost is the value of the single best thing you gave up in order to do this instead. If a free afternoon could go to a mock test or a shift at work, the opportunity cost of the mock is the shift's pay, not the pay plus everything else you might have done."

That pass took about twenty minutes and produced one correct definition, one worked distinction and one genuine gap for the study list. Rereading the chapter would have produced the feeling of knowing all three.

How long does one Feynman pass take?

For a single topic, budget 20 to 30 minutes: roughly ten minutes writing the explanation from memory, five marking the gaps, and the remainder checking the source and rewriting. A first pass on an unfamiliar topic runs longer, because you are half-learning while you write.

Those are planning figures to replace with your own after a week of records. The useful signal is the trend: if a pass regularly takes an hour, the unit you chose is too big and should be split. If it regularly takes five minutes and finds nothing, you are explaining topics you already hold securely, and the same half hour would do more for you in a timed practice set. For a way of costing topics honestly across a whole syllabus, see how long it takes to learn one topic.

Which topics deserve a Feynman pass and which do not?

Reserve the method for topics where a mechanism has to be understood and where you suspect you are repeating phrases you cannot unpack. It is an expensive tool, so spending it on material that only needs drilling is a real cost.

For the pure-recall row, the right tool is deliberate encoding rather than explanation, which is what mnemonics and memory palace methods are for.

How do I use it when there is nobody to teach?

Write or speak the explanation to an imagined beginner rather than a real one, and impose the constraint that makes the imagined audience behave like a real one: no term may appear in your explanation unless you have already defined it in your own words.

Three practical substitutes work. Record yourself explaining the topic aloud and play it back the next day, when you will hear the hedging you did not notice while speaking. Write on a blank page with a pen, which is slow enough to make skipped steps obvious. Or draw the topic as a concept map first, then narrate the links out loud, since an unexplainable arrow is the same finding as an unexplainable sentence.

How do I turn the gaps into next week's study blocks?

A pass that ends with a tidy explanation and no follow-up task has wasted most of its value. The marked breakdowns are the output; the explanation is a by-product.

Convert each marked gap into one line with a verb and a finish condition: "derive the formula for compound growth twice without notes", not "revise growth". Put those lines into the coming week as scheduled blocks rather than a floating list, and keep them visible next to the rest of your workload so they compete honestly for time. In MyStudyPlanner, the gap tasks go straight into the planner and calendar as dated to-dos, which stops them accumulating silently.

Then schedule a second look at the same topic on a later day. The Feynman pass tells you what you did not know today; whether the repair survived is a separate question, and the answer depends on returning to it, as set out in the spaced repetition revision timetable.

Is the Feynman technique better than active recall?

They answer different questions, so the comparison is about sequencing rather than superiority. Active recall asks whether an answer is available at all. The Feynman technique asks whether you can explain, justify and connect it.

Recall prompts scale: you can run forty of them across a syllabus in an hour and produce a coverage map. Feynman passes do not scale, because each one costs half an hour and returns depth on a single topic. The efficient combination is to use recall prompts broadly, then spend Feynman passes on the minority of topics where recall keeps producing a half-answer, a hesitation or a memorised phrase. If the answer simply is not there at all, the problem may not be explanation depth; work out why you forget what you study before spending 30 minutes per topic.

Does research support the Feynman technique?

Not under that name, and it is worth being straight about this. The Feynman technique is a practical routine assembled from older ideas rather than a studied intervention with its own evidence base. What it borrows from, self-explanation and retrieval practice, has been examined far more closely than the routine itself has.

Dunlosky and colleagues' review rated practice testing and distributed practice as highly useful across a wide range of learning situations. That supports the ingredients: explaining from memory is a retrieval attempt, and returning to a topic later is distributed practice. It does not establish that this particular four-step packaging outperforms other ways of doing the same two things, and no honest guide can tell you that it does.

What are the most common mistakes?

• Explaining with the book open. The whole diagnostic value comes from the source being closed during step two.
• Choosing a whole chapter as the unit. You will produce a summary rather than an explanation, and summaries hide exactly the joints you are trying to test.
• Skipping the rewrite. Looking up a gap feels like fixing it. Only the second explanation shows whether it is fixed.
• Simplifying into vagueness. "Plain language" means no unearned jargon, not fewer details. If the simplified version is no longer correct, it is not simpler, it is wrong.
• Running it on everything. Thirty minutes per topic across two hundred topics is not a revision plan. Triage first.
• Not recording the gaps. An unwritten gap is forgotten by the evening.

Frequently Asked Questions

What is the Feynman technique?

The Feynman technique is a revision method in four steps: pick one topic, write an explanation of it in plain language as if teaching a beginner, mark every point where the explanation breaks down, then return to the source, repair those points and simplify again. The gaps you find become the study list.

How long does one Feynman pass take?

For a single topic, budget 20 to 30 minutes: roughly ten minutes writing the explanation from memory, five marking the gaps, and the rest checking the source and rewriting. A first pass on an unfamiliar topic runs longer. If it regularly takes an hour, the unit you chose is too big and should be split.

Which topics deserve the Feynman technique?

Topics where understanding a mechanism matters and where you suspect you are repeating words you cannot unpack. It is poor value for pure recall items such as vocabulary, dates and constants, and for procedural fluency, where solving fresh problems under time is the better use of the same 30 minutes.

How do I use the Feynman technique with nobody to teach?

Write or speak the explanation to an imagined beginner rather than a real one. Set a rule that no term may appear unless you have already defined it in your own words. Recording yourself and playing it back, or writing on a blank page with the notes closed, both supply the missing audience adequately.

Is the Feynman technique better than active recall?

They answer different questions. Active recall tests whether an answer is available at all; the Feynman technique tests whether you can explain and connect it. Use recall prompts for coverage across a large syllabus, and reserve Feynman passes for the smaller set of topics where explanation keeps collapsing.

Does research support the Feynman technique?

Not under that name. It is a practical routine rather than a studied intervention, and it borrows from self-explanation and retrieval practice. Dunlosky and colleagues rated practice testing and distributed practice as highly useful across learning situations; treat the Feynman technique as a sensible way to apply those ideas, not as a proven method in itself.

Your next step

Pick the one topic you have reread most often and understood least, set a timer for 25 minutes, and run a single pass on it today. Write the marked gaps as dated tasks in your MyStudyPlanner planner before you close the page, or keep them in the free study templates if a spreadsheet suits you better.