The Feynman technique is a study method in which you explain a concept in plain language, as if teaching a beginner, and treat every place where the explanation stalls as a gap to fix. It takes about four steps and a blank page, it works best on concepts you need to understand rather than facts you need to memorize, and it can be done entirely on paper or on a device that never sends your notes anywhere. This guide covers the steps, two worked examples, how to structure your notes, what the evidence does and does not show, and where the method fits next to spaced review.
TL;DR
- The method has four steps: pick one narrow concept, explain it simply in writing or aloud, mark where you stall or reach for jargon, then return to the source and rewrite the explanation.
- Its value is as a gap-finder. The confusion you feel when an explanation breaks down is the signal that real learning is still needed.
- Budget roughly 30 to 60 minutes per concept, so reserve it for foundations such as mechanisms and cause-and-effect, and use flashcards for plain facts.
- A small 2026 classroom study found a significant gain for an AI-integrated version of the technique, but it is a single study of seventh graders, so treat it as encouraging rather than conclusive.
- Your explanations reveal what you do and do not know, so keep them in a notes app that processes data on the device.
Table of Contents
- What the Feynman technique is, and what it is not
- The four steps
- Two worked examples
- How to take notes with the Feynman technique
- Does the Feynman technique actually work?
- Where it fails, and how to avoid that
- Our perspective: your explanations are sensitive data
- Studying privately with Echo Chamber Pro
- FAQ
- Sources
What the Feynman technique is, and what it is not
The method is named after the physicist Richard Feynman, who shared the 1965 Nobel Prize in Physics for work in quantum electrodynamics. The study routine that carries his name is best understood as a popular reconstruction, and one overview describes it as Feynman's reported study method. That matters because there is no single official version. The guides you find online differ in whether they use three, four or six steps, but they share one core idea: you learn by producing an explanation, not by rereading one.
The Ohio State University Dennis Learning Center describes it as a study method where you try to explain concepts simply, and notes that your problem areas show up where you get stuck or fall back on complex terminology. That sentence is the whole method in miniature.
One clarification helps before you start.
- It is not a summary. A summary compresses what the textbook says. The Feynman technique asks you to rebuild the idea so that someone without the vocabulary could follow the reasoning.
The four steps
A 2026 research paper on the technique as a self-directed learning strategy lays the process out as four steps, with an estimated 30 to 60 minutes per concept. Here is how each step works in practice.
Step 1: Pick one narrow concept
Choose something you can name in a short phrase. "Why bread rises" is workable. "Baking" is not. "Why a bond's price falls when interest rates rise" is workable. "Finance" is not. If the concept is too wide, you will produce a vague tour instead of an explanation, and you will not stall anywhere, which defeats the purpose.
Write the concept at the top of a blank page or an empty note. Then study it from your usual source, whether that is a lecture, a chapter or a set of slides, and close the source before you move on.
Step 2: Explain it simply, from memory
Write or say the explanation as if you were speaking to a bright 12-year-old. The sources agree on this framing. Use short sentences. Include at least one concrete example. Do not look at your material.
Doing this from memory is the important constraint. Learning-science guidance for tutors makes the same point: the learner should produce the answer from memory before seeing the material again, because recognition ("yes, that looks familiar") is a weaker signal than recall.
Step 3: Mark the gaps
Reread your explanation and mark every spot where you:
- stopped and could not continue,
- wrote a vague phrase such as "it basically just sort of works",
- leaned on a technical term without defining it, or
- gave a definition but could not say why it is true.
Practitioners describe the useful moment as the specific, localized confusion you feel when an explanation turns circular or jargon-heavy. That is where your mental model breaks. Do not skip past it, and do not feel bad about it. A list of five gaps is a good outcome, because you now know exactly what to study.
Step 4: Go back, patch, and simplify again
Return to the source for only the marked gaps. Then rewrite the explanation, replacing jargon with plain words and, where you can, adding an analogy. Many guides add a fifth beat here: repeat the cycle until the explanation runs without a stall. The Dennis Learning Center version ends the same way, with review and repetition to reinforce retention.
Two worked examples
Examples are where most guides stay thin, so here are two in different subjects. Each shows a weak first pass and the repaired version.
Example 1: Why bread rises (cooking, general science)
First pass: "Yeast produces carbon dioxide through fermentation, which makes the dough expand."
Gap found: What does "expand" mean physically? Why does the gas stay in the dough instead of escaping?
Repaired version: "Yeast are tiny living things that eat the sugars in flour and give off gas. The dough is stretchy because kneading builds a web of gluten, and that web works like the skin of a balloon. The gas pushes outward on the web, the web stretches and holds it in, and the dough puffs up. Baking sets the shape by firming the web."
The balloon analogy is the kind of picture used in a set of worked Feynman walkthroughs, and it passes the practical test: it lets you predict what happens in a case the textbook did not cover, such as why under-kneaded dough gives a flat loaf.
Example 2: Cash flow (personal finance)
First pass: "Cash flow is the net amount of money moving in and out of an account over a period."
Gap found: This is a definition, not an understanding. Why can a profitable business still run out of cash?
Repaired version: "Think of a bathtub. Money coming in is the tap, money going out is the drain, and the water level is your cash. A business can sell a lot, which looks like a full tap, but if customers pay in 60 days and bills are due in 30, the drain opens before the water arrives. The tub runs dry even though sales are strong."
The bathtub comparison appears in the same set of walkthroughs, and the test of a good analogy is again predictive: you can now explain why fast-growing companies often face cash crunches.
How to take notes with the Feynman technique
Readers often ask about notes specifically, so here is a layout that keeps the method honest. It uses one page per concept and four labeled zones.
| Zone | What goes in it | Rule |
|---|---|---|
| Concept | A narrow title, one line | If you cannot say it in a short phrase, split it |
| Explanation | Your plain-language draft, written from memory | No source open, no copying |
| Gaps | A list of stalls, vague phrases and undefined terms | Mark them as you reread, not as you write |
| Patch | The corrected explanation plus one analogy | Jargon only if you define it in the same sentence |
If you want to turn these pages into review material afterward, convert each gap into a single question and review those questions on a schedule. Our own pieces on private active recall on iOS and on flashcards versus notes go further on that handoff.
Does the Feynman technique actually work?
The honest answer is that its mechanism is well supported and its direct evidence is thin.
The mechanism. The technique combines retrieval (you must produce the explanation from memory) with elaboration (you connect ideas using examples and analogies). Study-skills overviews such as Athenify's guide to study techniques place it in that category and describe the process as explaining a concept in plain language to a complete beginner. Retrieval is the better-established ingredient: producing information is generally a stronger learning event than passively reviewing it.
The direct evidence. A 2026 within-subjects study of Grade 7 students compared an AI-integrated Feynman intervention with a spaced-repetition (Leitner system) intervention on cellular biology concepts. Mean scores rose from 24.67 to 26.30 after the Feynman intervention (p=.014), a statistically significant change, while the Leitner condition showed no significant difference from baseline (p=.187).
Read that result with its limits in mind:
- It is one study, in one school, with seventh graders and one biology topic.
- The intervention was AI-integrated, so it is not a test of the plain pen-and-paper method.
- The gain is about 1.6 points on the test scale, which is statistically detectable but modest.
- The spaced-repetition arm should not be read as proof that spaced repetition fails. Short interventions can favor methods built around understanding.
A separate study of the technique as a heutagogical strategy, meaning self-determined learning, argues it suits independent and remote study because the learner checks their own understanding and does not depend on outside validation. That is a reasonable design argument, though you should weigh it as a researcher's interpretation.
Put together: the evidence says the method is a sound way to practice retrieval and elaboration, and that it is promising, not proven, as a stand-alone intervention.
Where it fails, and how to avoid that
Pseudo-understanding. The most common failure is repeating textbook definitions in slightly simpler words and feeling fluent. The countermeasure is the no-jargon rule: if you cannot explain the concept without using the technical term itself, you have not yet understood the mechanism. Add a prediction test too. Ask what would happen in a scenario the source never mentioned.
Using it on everything. The technique is time-intensive. Practitioners recommend treating it as a precision tool for conceptual foundations, and using spaced repetition for simple factual memorization, where it is more efficient. A reasonable split:
| Material | Better tool |
|---|---|
| Mechanisms, causes, "why" questions | Feynman technique |
| Proofs and derivations | Feynman technique, then practice problems |
| Vocabulary, dates, formulas, definitions | Spaced-repetition flashcards |
| Large volumes of facts before an exam | Flashcards plus self-quizzing |
If you are weighing flashcard tools for the second half of this workflow, our comparisons of other spaced-review apps and privacy-first flashcard options for Apple users lay out the tradeoffs.
Our perspective: your explanations are sensitive data
We build private AI applications for Apple devices, and we process data on the device rather than on remote servers. Studying with the Feynman technique is a good case for why that matters, and it is a point most study guides skip.
Consider what a set of Feynman pages contains. It is a record of what you do not understand: your stalls, your wrong first drafts, the topics where you needed three attempts. For a student, that is a map of weaknesses. For a professional preparing for a certification, a licensing exam or a career change, it can say more about them than a transcript does. Notes like these are personal in a way a published textbook is not.
That is why we separate two ideas that are often blurred. Data that is stored locally is not necessarily processed locally, a distinction we explain in Stored Locally Is Not the Same as Processed On-Device. A notes app can keep a file on your phone and still send its text to a server for analysis. When you evaluate a study tool, ask where the data path runs: what leaves the device, when, and whether you opted in. If you want a practical checklist, our guides on a five-minute iPhone privacy pass for student notes and on verifying that study notes are private on iOS walk through the settings.
There is also a tradeoff to name. A tool that analyzes your explanation can speed up step 3, and the 2026 classroom study above used an AI-integrated version. But the gap-finding is the learning. If software points out every gap for you, you skip the productive confusion that the method depends on. Our view is that AI works best as an optional second reader after you have found your own gaps, and that it should do that work on your device, with any network connection clearly explained and strictly opt-in.
— Alex
Studying privately with Echo Chamber Pro
If you want an on-device study app for your Apple devices, Echo Chamber Pro is one of the apps we publish. It is sold through the Apple App Store at 2.99 USD per month, 29.99 USD per year, or 79.99 USD as a one-off purchase. You can see the details on the Echo Chamber Pro page. Our wider approach is described on the Obsidian Ridge Labs home page.
FAQ
Does the Feynman technique actually work?
It works as a way to practice retrieval and find gaps, and one small 2026 study of Grade 7 students found a significant score gain for an AI-integrated version. The evidence base is still limited, so use it as a targeted method for hard concepts rather than a guaranteed shortcut. Pair it with spaced review to keep what you patch.
How long does one Feynman session take?
Plan for roughly 30 to 60 minutes per concept, based on the estimate in the heutagogy study. A narrow concept can take less, while a broad one will take more and should be split. If a session runs long, the topic is usually too wide.
What was Feynman's actual IQ?
A commonly cited figure is 125, from a school test in his youth. Forbes notes that the test was probably scored by the old quotient method, which is not how IQ tests are scored today, so the number is a poor guide to his abilities. It has no bearing on whether the study method works for you.
Sources
- The Feynman Technique, Dennis Learning Center, The Ohio State University
- Feynman Technique as a Heutagogical Learning Strategy
- Plus Ultra Paradigm: AI-Integrated Feynman and Leitner Active Recall Methods (Zenodo)
- Feynman Technique Examples: 3 Worked Walkthroughs
- Study Techniques: Active Recall, Spaced Repetition and More, Athenify
- Learning-science skill guidance on retrieval practice
- Learn Anything Faster With These Nine Techniques
- Richard P. Feynman, Princetoniana Museum
- Richard Feynman's IQ Score Was Only 125, Forbes