@heynavtoor - Feynman Method Meets Claude
@heynavtoor on X A creator on X lays out a four-step learning routine borrowed from Richard Feynman and adapts it into four prompts you can run in Claude in about 20 minutes. The core idea is to stop passively rereading and instead explain material in your own words, find your gaps, and lock ideas with analogies.
- The post cites Ebbinghaus's forgetting curve, claiming roughly 80 percent of material is gone after 31 days without active reinforcement.
- It leans on research like Koh et al. (2018) on teaching as a learning tool, and Karpicke and Blunt (2011) on retrieval practice beating rereading, though the post's specific figures (10 to 20 percent gains) are worth checking against the originals.
- The workflow is four prompts: find the five load-bearing ideas, explain them simply, get a gap report, then attach analogies, all in a 20-minute session.
The method at the center of this piece is the Feynman Method, a four-step way of learning that the author says can be run inside Claude in about 20 minutes. The steps are: pick one concept and write it down, explain it as if teaching a 12-year-old, hunt down every gap or vague spot in that explanation, then go back to the source, fix the gaps, and tie the idea to an analogy. The author's claim is that most people forget most of what they read, citing Hermann Ebbinghaus's 1885 forgetting curve, which shows steep losses within hours and up to 80 percent within a month without reinforcement. The deeper problem, per the article, is the "fluency illusion": a clear paragraph feels like understanding when it's really just recognition. The author's line that reading proves attention while explaining proves understanding is the core idea the whole piece hangs on.
The article grounds the method in research it attributes to a few sources. It cites Koh, Lee, and Lim (2018), reporting that students who taught material scored 10 to 20 percent higher than those who only studied, and Karpicke and Blunt (2011, Science), finding that self-testing outperformed rereading. It also invokes Robert Bjork's concept of "desirable difficulty" from UCLA in 1994, the idea that effortful learning sticks better than easy learning. The article presents Feynman's own origin story as evidence of the approach: his "Notebook of Things I Don't Know" from his Princeton exam prep, where he worked only on what he couldn't yet explain. Feynman himself never published research on the method, so the scientific backing comes from later studies of retrieval practice and teaching, not from his own work.
The practical core is four prompts for Claude: one to extract the five load-bearing ideas from a topic, one that asks you to write your own plain-language explanations before comparing them to Claude's model answers, one that reads your explanations and reports where your understanding breaks, and one that builds two analogies per idea along with a note on where each analogy fails. The author suggests a timed workflow across roughly 20 minutes per topic, with the prompts spaced across minutes 1 to 20. Worth keeping in mind: the piece is promotional for a prompt set, and its strongest evidence is for retrieval and self-explanation in general, not for any specific AI tool.