Spokes.wiki Search About
Defined Term mechanism updated Wed Aug 12 2026 00:00:00 GMT+0000 (Coordinated Universal Time)

Note-taking and learning

The empirical question of whether how you record a lecture changes what you retain — and the spoke’s first page on memory and learning at the level of a concrete study rather than a theory.

The claim, in three versions

The popular version. Writing by hand is better for learning than typing, because paper engages something typing does not. This is the version that circulated widely after Mueller & Oppenheimer (2014), and it is the one handwriting-vs-typing-notes was written to correct.

The measured version. Across a 2024 meta-analysis of 24 studies, handwriting’s advantage on test performance is Hedges’ g = 0.248 — real, small. Typing’s advantage on volume recorded is g = 0.919 — much larger. A 2021 direct replication found no difference on an immediate quiz (74 laptop, 68 longhand), and a 2026 EEG study of 33 undergraduates found none either.

The mechanism version. Handwriting is slower than speech, so a note-taker cannot transcribe and must compress, select and rephrase. That processing is what the test later measures. On this reading the instrument is incidental and the constraint is the active ingredient — which is why the same studies find that verbatim copying hurts performance in both groups, and that more words recorded helps both.

Why the third version is the one to keep

It is the only one of the three that predicts the null results instead of being embarrassed by them. If the effect belongs to the constraint rather than the medium, then a study measuring immediate recall — where reorganizing the material has not yet paid off — should find nothing, and two of the three studies here did.

It also makes the effect transferable, which the popular version does not: any capture method that forces compression should behave like longhand, and any that permits transcription should behave like a keyboard. No source in this corpus tests that prediction, so it stays a prediction.

What this corpus does not have

  • The primaries. Every figure on this page comes through a T3 secondary report that names no authors for two of its three studies. Mueller & Oppenheimer (2014) is discussed here and held nowhere.
  • Anything on delayed testing. The mechanism above predicts the gap should grow with delay. The reported studies test immediately.

Adjacent, and deliberately not merged

../research-wiki’s spaced-repetition and zettelkasten are about what happens to a note after it exists — scheduling review, linking it into a corpus. This page is about the moment of capture. Same subject matter from opposite ends, and the seam between them is untested: nobody here has a source on whether a compressed note is also a better retrievable one.

why-writing-feels-hard approaches writing as effortful from the production side; dysgraphia concerns handwriting when the motor and language systems fail. Neither is about encoding, and none of the three should be collapsed together.

handwriting-vs-typing-notes · replication-crisis · why-writing-feels-hard · dysgraphia · spaced-repetition · zettelkasten · synthesis