[0kl]

If anyone else is interested in research on note taking, this is a survey of the research with citations: https://www.cultofpedagogy.com/note-taking

Unfortunately it doesn’t look like there’s been as much research on specific methodologies (e.g. Cornell method), but it may also be the case that the methodology doesn’t matter at all, only that the strategies are employed.

I’m surprised it doesn’t mention the encoding and external storage paradigm, but https://www.cc-seas.columbia.edu/node/31875 has some additional context for anyone that wants some vocabulary and a very short “methods” and “strategies” document.

[btrettel]

> Unfortunately it doesn’t look like there’s been as much research on specific methodologies (e.g. Cornell method), but it may also be the case that the methodology doesn’t matter at all, only that the strategies are employed.

I recall that in the book Your Memory by Ken Higbee he said that various study/notetaking methods don't differ much and just combine strategies that are known to work. So he picked a popular system to recommend to readers: https://en.wikipedia.org/wiki/SQ3R

You can immediately see similarities between SQ3R and Cornell Notes.

[Bartweiss]

I've heard both teachers and education researchers argue that once you understand an idea and represent it competently, repeated exposure is the only thing that matters for learning.

If people who are using a specific technique get better results, it's probably just survivorship bias, because everything in education is survivorship bias. Even if people who adopt the technique all do well, it's probably just selection bias on the sort of people who seek out new study techniques. And if you teach it to students and they improve, it's probably still the Hawthorne effect, with novelty and optimism meaning it gets used more consistently at first.

There are a few pretty miraculous results, but they're mostly on memorizing lots of completely arbitrary info (e.g. memory palaces). Outside of that, it seems like picking anything easy to use and being consistent is the winning approach.

(I'd allow a special exception for things like Anki: if one tool supports data relationships another doesn't, it can obviously be better for learning that sort of content.)