Feynman Technique is a four-step learning method that uses plain-language explanation as a diagnostic of understanding. The learner studies a concept, explains it in simple terms as if teaching a novice, identifies any gaps exposed by that attempt, then returns to source material to repair those gaps and consolidate knowledge.
The name derives from physicist Richard Feynman's approach to understanding, though the four-step method was formalised by later writers rather than by Feynman himself.
How it works
The cognitive engine behind the technique is self-explanation. When a learner attempts to reformulate a concept in plain language, the act of production forces the construction of inferences that connect new information to existing schemas. Chi et al.'s foundational work demonstrated that learners who explained worked examples aloud solved more problems and made fewer errors than peers who re-read the same material without explaining 1. The explanation attempt does not merely express understanding; it creates it.
Reformulating knowledge in one's own words also engages retrieval practice: actively reconstructing information strengthens memory traces far more than equivalent restudy time alone 2. A second related mechanism operates upstream: even the expectation of having to teach activates more coherent encoding of material. A Kobayashi meta-analysis found that studying with the expectation of explaining produced a near-medium effect size (g = 0.48) on learning outcomes, whereas studying without that expectation yielded no significant benefit over passive study (g = -0.02) 4.
The technique also functions as a diagnostic instrument. Attempting to produce a plain explanation exposes illusions of knowing, the common state in which familiarity with a label or term masks the absence of genuine conceptual understanding 3. Where explanation fluency breaks down, the gap is visible rather than hidden. The learner returns to source material specifically to resolve those failures, making revision targeted rather than diffuse.
Learn by Teaching
The Feynman technique — study it, teach it in plain words, find the gaps, then refine and simplify.
g = 0.55
effect size for self-explanation across 64 studies
An analyst studying a complex financial model closes the reference material and attempts to explain the valuation methodology in plain terms, as if briefing a colleague with no financial background. When the explanation falters on the distinction between enterprise value and equity value, the gap is immediately apparent. The analyst returns to the source, re-reads only the relevant section, and reconstructs the explanation until it holds.
The breakdown in explanation reveals precisely what understanding is missing, converting a vague awareness of ignorance into a specific, actionable target.
Why it matters
Most conventional study methods fail to distinguish between recognition and recall. Highlighting, re-reading, and passive review expose the learner to information without requiring reconstruction, creating a sense of familiarity that correlates poorly with actual performance on transfer tasks. The Feynman Technique disrupts this pattern by making production mandatory. A Bisra et al. meta-analysis of 64 studies found a weighted mean effect size of g = 0.55 for self-explanation prompts on learning outcomes, placing it among the most effective low-cost study interventions identified across the education research literature 3.
The practical consequence extends beyond retention. Chi et al. showed that self-explainers transferred successfully to novel problems, whereas learners who relied on surface-level pattern matching without self-explanation failed on transfer tasks 1. Self-explanation produces comparable gains across subject areas, educational levels, and both declarative and procedural knowledge types, indicating the mechanism is not domain-specific 3.
Frequently asked
What are the four steps of the Feynman Technique?+
The four steps are: study the concept from primary material, write or speak a plain explanation as if teaching a novice, identify where the explanation breaks down or becomes vague, then return to source material to fill those specific gaps and refine the explanation until it holds without qualification.
Does the Feynman Technique actually improve learning and test performance?+
Yes. Across 64 studies and approximately 6,000 participants, self-explanation prompts produced a weighted mean effect size of g = 0.55 on learning outcomes. Separately, studying with the expectation of teaching a concept improved outcomes significantly (g = 0.48), whereas studying without that expectation showed no reliable benefit over passive study.
Is the Feynman Technique effective across subjects other than science?+
Yes. A meta-analysis of 64 studies found that self-explanation prompts improved learning outcomes across multiple subject areas, educational levels, and both conceptual and procedural knowledge types. The benefits are not confined to scientific or quantitative subjects; the effect is consistent wherever learners are required to build genuine understanding rather than rote recall.
How is the Feynman Technique different from summarising or highlighting?+
Summarising and highlighting are passive: they expose the learner to information without requiring reconstruction. The Feynman Technique demands active production, reformulating a concept in one's own words, which activates retrieval practice and forces the construction of inferences. This generation requirement is what converts surface familiarity into durable, transferable understanding.
