Comment by wenc | Hacker News Reader

Comment by wenc | Hacker News Reader

Parent Full thread

Good point, but I would say "pretty bad" is relative in my opinion. I think of it as something like this:

Mental notes < written notes < SRS < repeated exposure + emotional connection.

Specifically around retention (and only retention), taking notes doesn't get you everything, but it gets you something.

(I do feel some of us put undue focus on taking notes at the expense of understanding the world through empirical means, but that's another discussion. I'm focusing solely on retention here.).

Active recall testing and distributed practice are great ideas. No disagreement there. (the Ebbinghaus curve is well known)

(also, thanks for the link to the publication -- those are interesting results that describe aggregate outcomes for the population studied, but there's usually variation in individual outcomes, and in self-selected groups like this one, a different subset of techniques may dominate)

To your point about Anki specifically, I'm a language learner and in my experience, flashcard-style SRS can help retention for atomic types of knowledge, but I find as I get to intermediate levels of a language, the utility of flashcards declines precipitously -- because one is called upon to synthesize structures rather than just words.

So far I haven't found an easy way to encode structures into flashcards (clozes etc are too simplistic)... furthermore, language learning is complicated by the need to acquire a "feeling" for the context of how certain words are used. It's a kind of tacit knowledge that sometimes defies simple flashcard-type encodings. You can encode context through narrative, but then your flashcards become too unwieldy.

I've found writing and speaking (with feedback) in small chunks over a long period of time to be much better at creating the "neurocircuitry" necessary for retention. Again, I wouldn't say flashcard SRS are bad at improving retention (I use them), but like taking notes, even if they don't get you all the way there, they get you somewhere.

I feel that at some point you need to be chunking flashcards in a way that reflects your evolving level of competency. This is an extra amount of effort and this is something I'm not sure I've seen any research or even much discussion about, so I don't know how effective it might be or whether it's worth the extra time. That said, after surveying all the blog posts, research and articles I could find, I don't think it's something that's really been explored for flashcards. A certain set of static rules have solidified to which most people adhere and they seem diametrically opposed to the idea of building a lattice of deeper, conceptual knowledge. Do flashcards really need to be simple and atomic? Do they need to be answerable in a second or two? Do they need to be easily parsed so they can be instantly answered? I think there's a lot of dogma there that needs to be reconsidered.

I've seen some interesting approaches around flashcards for mathematical concepts applied to sets of mathematical problems. I've experimented with the idea where a flashcard might ask me about algorithms in a way that challenges me to apply a specific algorithm to solve a problem or think about its properties and how they affect where/when they should be used, and I feel it's more useful then your average "vocab"-like flashcard.

If you google around, you'll find many descriptions of knowledge domains where flashcards either don't work or aren't efficient.

My layman's take is that flashcards are based on the idea of forcing retrieval through a trigger. It seems to me that atomicity is an inherent assumption, as well as independence, and that each flashcard maps to a finite set of deterministic answers. Feelings or a sense of things aren't as easily encoded. Tabular/graph relationships aren't as easily encoded either, and likewise sequences of decision trees aren't easily encoded. One can presumably design a chained sequence of cards complete with decision trees, but at this point flashcards end up being an unwieldy paradigm.

Fortunately one doesn't need limit oneself to flashcards. There are other methods like the method of loci etc. that work great for say, sequenced knowledge, like speeches or lyrics to a song, areas in which flashcards aren't the most natural fit.

To me however, memory is but one tool out of many for understanding something. I personally tend not to focus on memory itself too much. For the kind of stuff I'm interested in, memory is an outcome actually doing stuff, getting thing wrong, and then internalizing. In language learning, the memory that comes from embarrassment from making faux pas is both more quickly assimilated and retained longer than via SRS. I still remember mistakes made years ago even though they only happened once or twice.

Reply on news.ycombinator.com