Feynman Technique

There is a moment every student knows: a friend asks you to explain something you have “studied,” and three sentences in, you hear yourself saying “well, it’s basically, you know…” and trailing off. You did not suddenly forget. You never actually had it. Reading about a concept and being able to rebuild it in your own words are different achievements, and exams pay for the second one.

The Feynman technique industrializes that uncomfortable moment. Named for the physicist Richard Feynman, who insisted that anything truly understood can be explained simply, it is a four-step loop: explain a concept in plain language from memory, find where the explanation breaks, repair the break at the source, and simplify again. It is essentially active recall with a plain-language constraint bolted on, and like all retrieval-based methods it gets even better when the explanations are repeated on the kind of spaced schedule described in spaced repetition, explained.

It is also routinely oversold as a do-everything study method, which it is not. Here is how to run it, a worked example, and an honest account of where it stops.

The four steps

Step 1: Choose a concept and write it at the top of a blank page. One concept. “Osmosis,” not “cell transport.” “Price elasticity,” not “microeconomics.” The technique works at the resolution of a single idea.

Step 2: Explain it in plain language, from memory, as if teaching a smart twelve-year-old. Write or speak the explanation without looking at your notes. The audience constraint is doing real work here: a twelve-year-old has no patience for jargon, so every technical term you’d normally hide behind has to be unpacked or replaced. Use examples. Use analogies. Keep going until you have covered what it is, why it happens, and what it predicts.

Step 3: Find the breaks. Somewhere in step 2 you stalled, got vague, or smuggled in a term you cannot actually define. Mark every one of those spots. Then go back to the textbook or lecture and re-learn precisely those pieces, not the whole chapter. This is the step people skip, and it is the entire payoff: the technique’s product is a map of your specific gaps.

Step 4: Simplify and retell. Rewrite the explanation incorporating what you fixed. If it still leans on a memorized phrase, translate the phrase into something you could have invented. When the explanation runs start to finish without a stall, you are done with this concept for today.

Why does this expose gaps that re-reading hides? Because explanation is generative. Re-reading lets you ride the author’s structure and mistake recognition for understanding. Explaining from a blank page forces you to supply the structure yourself, and missing pieces have nowhere to hide. Research on related habits backs this up: self-explanation earned a moderate-utility rating in Dunlosky et al.’s 2013 review, and the retrieval component connects it to the testing effect documented by Roediger and Karpicke (2006). When students explain why something works rather than re-confirming that it works, retention and transfer both improve.

A worked example: osmosis

Here is the technique running live on a standard biology concept.

First attempt (step 2): “Osmosis is when water moves across a semipermeable membrane from an area of high concentration to low concentration… wait. High concentration of what? Water or solute? It’s water moving toward the saltier side, so… toward high solute concentration. Why does it do that? Something about equilibrium. And it’s passive, no energy needed. Why doesn’t it need energy?”

Already two breaks: the direction got tangled, and “something about equilibrium” is a hand wave. That confusion is invisible during re-reading, because the textbook sentence always sounds right when the textbook is the one saying it.

Repair (step 3): Back to the notes. Direction: water moves toward higher solute concentration, because that side effectively has less free water. Energy: it is diffusion, driven by random molecular motion, so no pump required; molecules just statistically end up more spread out.

Retell (step 4): “Imagine a tank split by a screen with holes too small for salt but big enough for water. The salty side is crowded with salt particles, so it has fewer water molecules per scoop. Water molecules bounce around randomly and pass through the screen in both directions, but more cross into the salty side simply because more of them are free to travel from the fresh side. Net result: water flows toward the salt until both sides are equally crowded. No pump, no energy, just statistics. That’s why salted slugs shrivel and why your fingers prune in the bath.”

Notice what happened. The first attempt would have scored maybe half marks on an exam question. The final version can handle a question the textbook never asked, because it carries the mechanism, not the sentence. The same loop works on supply and demand, recursion, comparative advantage, action potentials, anything with a why inside it.

Where the technique earns its keep

The Feynman technique is expensive per concept, twenty minutes or more, so spend it where explanation-power matters:

• Load-bearing concepts that the rest of the course leans on: equilibrium in chemistry, opportunity cost in economics, inheritance in programming.
• Anything you can recognize but not produce. If you nod along in lectures but freeze on essay questions, that gap is exactly what this method measures.
• Exam formats that reward explanation: orals, essays, “explain why” short answers, teaching assessments.
• Concepts you got wrong on a practice test for reasons you cannot articulate.

For breadth, a faster cousin does the daily work: the blurting method dumps everything you remember onto a page without the plain-language polishing, covering a whole lecture in the time Feynman covers one idea.

The format is flexible, too. Writing is the default because it leaves evidence, but a voice memo on a walk works, and explaining to an actual person works best of all, since a real listener interrupts at exactly the spots you were hoping to glide past. Study groups are an underused venue for this: rotate who explains each concept cold, and the audience’s job is to ask “but why?” until the explainer either lands it or finds a gap. Either outcome is a win.

The honest limits

Now the part most write-ups omit. The Feynman technique is an understanding tool, and a lot of studying is not an understanding problem.

Take a pharmacology course with two hundred drugs. You can beautifully explain how beta blockers work, and that explanation will do nothing to help you remember that metoprolol is one of them and what its half-life implies for dosing. Names, doses, dates, vocabulary, anatomical structures, irregular verbs: this material is arbitrary. There is no deep mechanism that generates the French word for “seagull.” For raw recall at volume, the evidence points to retrieval practice on a spaced schedule, which means well-built cards reviewed at expanding intervals; a flashcard generator that turns your notes into question-and-answer cards covers in minutes what Feynman cannot cover at all.

Two more failure modes worth flagging:

1. Performance without retrieval. Some students “do Feynman” with notes open, producing a lovely explanation that is really transcription. The blank page is non-negotiable. Explaining while reading is just re-reading with extra steps.
2. One-and-done. A brilliant explanation written tonight still sits on the forgetting curve like everything else. The concept needs to be re-explained, or at least re-retrieved, days and weeks later. A good pattern: after a Feynman session, save the questions your gaps generated and fold them into your spaced reviews. StudyDone can run that follow-up loop from your notes, resurfacing each concept just before you would lose it.

Putting it into practice

This week, pick the three concepts in your hardest course that everything else depends on. Give each one a blank page and fifteen minutes: explain, mark the stalls, repair, retell. Keep the pages. The marked-up first drafts are the most precise diagnostic of your understanding you will ever get, far better than any highlighted textbook, and the questions they raise are exactly what your future review sessions should be made of.

FAQ

What are the four steps of the Feynman technique?

Pick a concept and write it at the top of a page. Explain it in plain language as if teaching a beginner, working from memory. Notice where you stall or reach for jargon, and go back to the source to fix those gaps. Then simplify the explanation again, with analogies, until it flows.

Did Richard Feynman actually invent the Feynman technique?

Not as a packaged four-step method; that formulation came later from writers describing how he worked. But the underlying habit was genuinely his: Feynman was famous for insisting that if you cannot explain something simply, you do not really understand it, and he prepared by rebuilding ideas from scratch in plain terms.

Is the Feynman technique a form of active recall?

Yes. Explaining from memory forces you to retrieve and reorganize the material rather than recognize it, which is the same mechanism that makes practice testing effective. It adds a layer on top: the plain-language constraint exposes fuzzy understanding that a flashcard answer might let slide.

How long should one Feynman session take?

Around 15 to 30 minutes per concept, including the trips back to your notes. That cost is why you should reserve it for the handful of load-bearing concepts in a course rather than trying to Feynman every fact on the syllabus.

Does the Feynman technique work for memorization-heavy subjects?

Only partially. It is excellent for the mechanisms and frameworks behind the facts, but there is no deep explanation that generates a drug name or a vocabulary word. For raw recall material, spaced flashcards are the right tool, with Feynman reserved for the concepts connecting those facts.