The Feynman Technique: How to Learn Anything by Teaching It

Who Was Richard Feynman (and Why His Learning Method Works)

Richard Feynman (1918-1988) was a theoretical physicist who won the Nobel Prize in 1965 for his work on quantum electrodynamics. But among students and self-learners, he's better known for something else: his obsession with clear explanation.

Feynman had a reputation at Caltech for demanding clarity. When colleagues presented ideas using dense jargon, he'd stop them and ask for a plain-English version. He believed that complexity in explanation almost always signaled confusion in thinking. His undergraduate lectures, published as The Feynman Lectures on Physics, remain among the most accessible physics texts ever written, 60 years after their original delivery.

The "Feynman Technique" as a named method wasn't something Feynman published in a paper. It was distilled from his approach to learning by later writers and educators who recognized a pattern in how he worked. The core insight is this: the act of explaining is not a way to demonstrate understanding. It's a way to create understanding.

This isn't just biographical trivia. Cognitive science has caught up with what Feynman practiced intuitively. Fiorella and Mayer (2016) published a comprehensive review of "generative learning" strategies and found that explaining material to others (or even to an imagined audience) produced significantly stronger learning outcomes than re-reading, highlighting alone, or summarizing without an audience in mind.

---

The Four Steps of the Feynman Technique

The method is simple enough to fit on an index card. That's part of its power.

Step 1: Choose a Concept

Pick one specific idea you want to understand. Not a broad topic like "economics" or "machine learning," but a focused concept: "comparative advantage," "gradient descent," "how mRNA vaccines work." Write the concept name at the top of a blank page.

The narrower the concept, the better the technique works. If you find yourself writing for 20 minutes and still haven't covered the basics, your topic is too broad. Split it.

Step 2: Explain It in Simple Language

Write an explanation as if you're teaching it to someone with no background in the subject. Use short sentences. Avoid jargon entirely. If you must use a technical term, define it immediately in everyday words.

The target audience matters. Feynman's original framing was "explain it to a child," but a more practical standard is: explain it to an intelligent adult who knows nothing about this specific field. A first-year college student in an unrelated major is a good mental model.

Step 3: Identify Gaps and Go Back to the Source

This is where the real learning happens. As you write your explanation, you'll hit points where you get stuck, wave your hands, or resort to vague language. These are your knowledge gaps.

Go back to the original material: the textbook, the paper, the article, your highlighted passages. Study specifically the parts you couldn't explain. Then return to your written explanation and fill in the holes.

Step 4: Simplify and Use Analogies

Once your explanation is complete and accurate, make it simpler. Cut unnecessary words. Replace abstract descriptions with concrete analogies. If you wrote "mitochondria perform oxidative phosphorylation to generate ATP," rewrite it as "mitochondria are like tiny power plants inside each cell; they burn fuel from your food to produce the energy currency your body runs on."

Good analogies aren't decorative. They're structural. They map the relationships in the new concept onto relationships the learner already understands.

---

Why Teaching Forces Understanding

The Feynman Technique works because of a well-documented cognitive phenomenon: the protege effect. When you prepare to teach something, your brain processes the information differently than when you study it for yourself.

Koh et al. (2018) ran a series of experiments where participants either studied material for a personal test or studied it expecting to teach it to another person. The "teaching expectancy" group consistently outperformed the "test expectancy" group, even when they never actually taught anyone. Just the intention to teach changed how deeply they encoded the material.

Why? Teaching requires you to do several things that passive studying doesn't:

• Organize information hierarchically: You need to decide what comes first, what depends on what, and what can be skipped.
• Identify causal relationships: You can't just list facts. You need to explain why things happen.
• Anticipate questions: You mentally model what a listener would find confusing, which forces you to address ambiguities you might otherwise ignore.
• Translate between registers: Moving from technical to plain language requires genuine comprehension of the underlying concept, not just familiarity with the terminology.

Fiorella and Mayer (2016) found that the "explaining" effect was strongest when learners generated their explanations without looking at the source material. This aligns perfectly with Step 3 of the Feynman Technique: the gaps you discover when explaining from memory are precisely the areas where your understanding is weakest.

Dunlosky et al. (2013) reviewed ten common study techniques in a landmark paper for Psychological Science in the Public Interest. They rated highlighting alone and re-reading as "low utility." Self-explanation and practice testing received "moderate to high utility" ratings. The Feynman Technique combines both of these high-utility strategies into a single workflow.

---

The Feynman Technique vs. Other Learning Methods

The Feynman Technique isn't the only evidence-based learning strategy. Here's how it compares to other popular methods:

Sources: Retention estimates synthesized from Dunlosky et al. (2013), Karpicke & Blunt (2011), Fiorella & Mayer (2016), and Roediger & Karpicke (2006).

The key distinction: most methods focus on input (how you consume information). The Feynman Technique focuses on output (how you produce an explanation). That reversal is what makes it so effective. The Feynman Technique is a form of active recall. For more on active recall as a study method, see our deep dive.

These methods aren't mutually exclusive. Combining the Feynman Technique with spaced repetition creates a powerful loop: explain a concept today, review your explanation in three days, refine it in a week. Each review cycle deepens your understanding while the spacing prevents forgetting.

---

How to Apply the Feynman Technique with Digital Tools

Feynman used notebooks and chalkboards. You have better options. Digital tools can remove friction from every step of the technique and make the whole process faster and more thorough.

Step 1: Capture What You're Learning

As you read articles, papers, or books online, use Glasp's web highlighter to mark the key passages that define or explain the concept you're studying. Color-code your highlights: one color for definitions, another for examples, a third for claims you want to verify.

This creates a curated source library you can return to in Step 3 when you need to fill gaps. It's far more efficient than re-reading an entire chapter to find the one paragraph you half-remember. For more on effective highlighting strategies, see the science of highlighting.

Step 2: Write Your Explanation

Open a blank document and write your explanation without looking at your highlights. This is critical. The point is to discover what you can't explain from memory. If you write with your highlights visible, you'll copy rather than reconstruct, and the technique loses its power.

Some people prefer writing. Others prefer speaking out loud, recording a voice memo, or even making a short video. The format doesn't matter as long as you're producing an explanation for an audience, real or imagined.

Step 3: Use AI to Stress-Test Your Understanding

This is where modern tools add something Feynman didn't have. After writing your explanation, paste it into Glasp's AI chat feature along with your original highlights. Ask the AI to:

• Identify factual errors or oversimplifications in your explanation
• Point out concepts from your highlights that you missed in your explanation
• Generate follow-up questions that test whether your understanding goes deep enough
• Suggest better analogies for your weakest explanations

The AI acts as a simulated student who asks difficult questions. It's like having a study partner available at any hour.

Step 4: Share and Get Feedback

Post your refined explanation or highlights on the community feed. When other learners comment on or highlight the same concepts differently, you get external perspectives that challenge and refine your understanding. Teaching doesn't require a classroom. A shared highlight with a thoughtful note attached is a micro-lesson.

For video-based learning, YouTube Summary lets you grab transcripts and key points from lectures, then apply the Feynman Technique to the content without rewatching the full video.

---

Real-World Applications Across Fields

The Feynman Technique adapts to any domain where understanding matters more than memorization.

Software Engineering

A developer learning a new framework can write a plain-language explanation of how its state management works. If they can't explain the data flow without using framework-specific terms, they'll struggle to debug it when something breaks. Many senior engineers use a version of this technique when writing documentation or onboarding junior team members.

Medicine

Medical students face enormous volumes of factual content. The Feynman Technique helps them move beyond rote memorization of symptoms and treatments toward understanding mechanisms. Explaining how a drug works at the cellular level, in plain language, reveals whether you've actually grasped the pharmacology or just memorized a flowchart.

Finance and Investing

Warren Buffett famously avoids investing in businesses he can't explain simply. This is the Feynman Technique applied to financial analysis. If you can't explain a company's business model in two sentences without jargon, you probably don't understand it well enough to predict its future performance.

Language Learning

The technique works for grammar concepts that resist memorization. Instead of drilling conjugation tables, explain why a language uses the subjunctive mood. What situations trigger it? How does it change meaning? Teaching grammar rules in plain language to an imagined student forces you to internalize the logic, not just the patterns.

Personal Knowledge Management

If you're building a second brain, the Feynman Technique serves as a quality filter. Before adding a note to your knowledge system, try explaining the concept in one paragraph. If you can't, the note needs more processing before it becomes useful knowledge. This prevents the common trap of accumulating thousands of notes you never actually understand. To remember what you read, you need to do more than save passages; you need to reconstruct them in your own words.

---

Common Pitfalls and How to Fix Them

Pitfall 1: Choosing Topics That Are Too Broad

"Explain quantum mechanics" is not a Feynman Technique exercise. It's a semester. Pick one concept within a field: "wave-particle duality," "the uncertainty principle," "quantum entanglement." Each of these is a single session.

Fix: If your explanation takes more than one page (roughly 300-500 words), your topic is too broad. Split it into sub-concepts and tackle each one separately.

Pitfall 2: Skipping the "Without Looking" Part

The entire technique collapses if you write your explanation while staring at the source material. You'll produce something that reads well but teaches you nothing, because you're transcribing rather than reconstructing.

Fix: Close the book. Close the tab. Put your highlights away. Write from memory first. The discomfort you feel is the technique working.

Pitfall 3: Mistaking Fluency for Understanding

You read an explanation, it makes perfect sense, and you feel like you understand it. This is the "fluency illusion," and it's one of the most well-documented traps in learning research (Bjork & Bjork, 2011). Understanding someone else's explanation is not the same as being able to produce your own.

Fix: After reading an explanation that seems clear, close it and try to reproduce the argument from scratch. You'll often discover that what felt obvious while reading becomes surprisingly hard to articulate.

Pitfall 4: Using Jargon Without Realizing It

Technical terms become invisible when you've spent enough time in a field. You stop noticing that "leverage," "iterate," "latency," or "regression to the mean" are jargon. Your plain-language explanation isn't plain if it contains words your audience wouldn't use in conversation.

Fix: Read your explanation aloud to someone outside your field (or imagine doing so). Every time they'd need to ask "what does that mean?", you've found a term to replace.

Pitfall 5: Stopping After One Pass

One round of explain-identify-gaps-simplify is good. Two or three rounds are significantly better. Karpicke & Roediger (2008) showed that repeated retrieval practice with feedback produced substantially better long-term retention than a single retrieval attempt.

Fix: Revisit your explanation after a few days. Try to improve it without consulting the source. Then check against your highlights. Each iteration deepens your understanding further.

---

Frequently Asked Questions

How long should a Feynman Technique session take?

Plan for 20-30 minutes per concept. About 5 minutes to review your source material, 10-15 minutes to write your explanation, and 5-10 minutes to identify gaps and revise. If a concept requires more than 30 minutes, it's likely too broad. Split it into smaller pieces.

Can I use the Feynman Technique for memorization, or is it only for understanding?

It's primarily a comprehension tool, not a memorization tool. For raw memorization (dates, vocabulary, formulas), spaced repetition is more efficient. But the two techniques complement each other well: use the Feynman Technique to understand a concept, then use spaced repetition to make sure you retain it over time.

Does it work for learning physical skills, not just intellectual ones?

Partially. The Feynman Technique is strongest for conceptual knowledge. You can use it to understand the theory behind a physical skill (biomechanics of a golf swing, music theory behind improvisation), but the physical execution still requires practice and repetition. Think of it as building the mental model that guides your practice.

What if I don't have anyone to teach?

You don't need an actual student. Write your explanation in a notebook, type it in a document, or record yourself talking. Fiorella and Mayer (2016) found that explaining to an imagined audience produced learning gains similar to explaining to a real person. The key variable is the act of generating an explanation, not whether someone receives it.

How is this different from just taking notes?

Standard note-taking is an input activity: you're transcribing what someone else said or wrote. The Feynman Technique is an output activity: you're generating an original explanation from your own understanding. Dunlosky et al. (2013) found that conventional note-taking, without further processing, had limited impact on retention. The Feynman Technique forces the processing that notes alone don't provide.

Can AI replace the "teaching" part of the technique?

AI can supplement it but not replace it. Using Glasp's AI chat to ask questions about your explanation is valuable for identifying errors and gaps. But the core cognitive benefit comes from you producing the explanation, not from the AI producing one for you. Reading an AI-generated summary feels like understanding, but it's the fluency illusion again. Write your own explanation first; then use AI to pressure-test it.

---

Conclusion: Think Clearly by Explaining Simply

Feynman once said, "The first principle is that you must not fool yourself, and you are the easiest person to fool." The Feynman Technique is, at its core, a method for not fooling yourself about what you know.

Most learning feels productive but isn't. You read, you highlight, you nod along, and three weeks later the concept is gone. The Feynman Technique breaks this cycle by demanding proof of understanding: a clear, jargon-free explanation that you produce from memory.

The research backs this up. Generative learning strategies like self-explanation consistently outperform passive review (Fiorella & Mayer, 2016). The protege effect shows that teaching intention alone improves encoding (Koh et al., 2018). And when you combine explanation with spaced review and active recall, retention rates can exceed 80% (Dunlosky et al., 2013).

Glasp fits naturally into this workflow. Highlight key passages as you read with Glasp's web highlighter. Write your Feynman-style explanation from memory. Use the AI chat feature to stress-test your understanding against your saved highlights. Share your insights on the community feed to get social feedback and discover how others interpret the same concepts.

You don't need to be a Nobel laureate to think like one. You just need a blank page and the honesty to admit what you can't yet explain.