Richard Feynman, the physicist, had a reputation for being able to explain anything to anyone. His method was not magic — it was a discipline: if you cannot explain it simply, you do not understand it yet. The technique named after him turns that observation into a learning loop that is more effective than re-reading, highlighting, or passive review, and it applies to any subject from physics to personal finance to programming.
What changed in 2026
- AI made passive consumption easier and understanding less necessary — until it isn't. With GPT-class tools able to summarize anything, the premium on actual comprehension shifted. The Feynman Technique is the counter-tool: it tests whether you understand something or just recognize it.
- Knowledge work moved faster. Picking up a new domain in weeks, not months, is now expected in many roles. Techniques that accelerate deep learning have more value than ever.
- Short-form content creates an illusion of understanding. Reading a thread or watching a video feels like learning. The Feynman Technique reveals whether you actually absorbed it.
The four steps
Step 1: Choose a concept. One specific idea, not a whole field. "How compounding interest works" not "personal finance."
Step 2: Write an explanation as if teaching it to a 12-year-old. Paper, notebook, text file — medium does not matter. No jargon unless you define it. Simple language forces simple thinking.
Step 3: Identify the gaps and fuzzy spots. Where you stumble, reach for jargon, write "etc.," or go vague — that is where understanding ends and recognition begins. Mark them.
Step 4: Go back to the source only at the gaps. Re-read the relevant section, watch the specific video, ask the targeted question. Then redo that part of the explanation.
Repeat until the explanation is complete and jargon-free.
Why it works
| Passive study |
Feynman Technique |
| Recognition: "I've seen this before" |
Recall: "I can reproduce and explain this" |
| Highlights apparent understanding |
Exposes actual gaps precisely |
| Easy, comfortable |
Effortful — which is why it encodes better |
| You learn what you already know |
You learn where you are wrong or incomplete |
The effort of generation — trying to produce an explanation from memory — is the encoding event. The gaps are the signal, not the shame.
How to use it in practice
For a new topic: After your first read or watch, immediately write the explanation. Do not re-read first. The gaps from your first explanation tell you exactly what to re-read.
For exam prep: Explain each concept on your syllabus in sequence. Rank them by how many gaps appeared. Study the gappy ones hardest.
For understanding something you think you already know: You will be surprised. Write the explanation of something you have used for years. The edge cases and exceptions show up quickly.
For technical documentation at work: Explaining a system or process to a newcomer — even on paper — reveals inconsistencies in your own model of it.
How to pick what to use it on
Not every learning task benefits equally:
- High value: complex concepts, foundational principles, anything you need to apply under pressure.
- Medium value: procedural knowledge (sequences of steps) — you can use a simpler checklist approach here.
- Lower value: pure memorization tasks (vocabulary, formulas) — spaced repetition is more efficient.
Common mistakes
Starting with clean notes. The technique is supposed to be rough. A polished explanation suggests you already understand it or are faking clarity you don't have.
Using jargon as a shortcut. Every term you use that you cannot define in plain language is a gap. Name them all.
Re-reading the source before writing. That removes the productive struggle. Write first, then identify what to re-read.
Treating the exercise as writing practice. The quality of the prose doesn't matter. The completeness and honesty of the explanation does.
Doing it once and moving on. Two or three cycles on a genuinely new concept is normal. One pass is rarely sufficient.
What to skip
- Teaching an AI chatbot as a substitute — it accepts nonsense without pushback; a real or imagined skeptical 12-year-old is a better audience.
- Explaining to yourself silently — writing or speaking forces output generation; silent thinking allows shortcuts.
- Using it for everything simultaneously — start with the three highest-priority concepts you need for a project or exam.
FAQ
Does it work for math and formulas?
Yes, but the explanation needs to cover what the formula does and why, not just the steps. If you can only reproduce the notation, you don't yet understand the concept.
Is the "12-year-old" framing too simplistic for advanced topics?
No. The goal is not to actually teach a child; it is to force you away from jargon. Expert-level understanding often produces the simplest explanations of all.
How long should each explanation session take?
10–20 minutes for a focused concept. If it takes longer, either the concept is too broad or you are polishing prose rather than testing understanding.
Can I use it to prepare for presentations or interviews?
Yes — and it is one of the best interview prep methods for technical roles. If you can explain the concept simply and handle follow-up gaps, you can handle technical questions under pressure.
Where to go next
See Active recall explained in 2026, Spaced repetition explained in 2026, and The Pomodoro Technique explained in 2026.