Spaced repetition can allow for infinite recall • Hacker News Board

Re: Spaced repetition can allow for infinite recall

#111

Post by Phiwise_ » Tue, Aug 09, 2022, 5:44 AM UTC

buscoquadnary wrote: post #65

ozim wrote: post #31

Earlier quoted context omitted.

That is false dichotomy that is running around. To really understand something first you blindly memorize - but that is not enough, once you do examples and exercises using what you blindly memorized you get to understand things quicker, a lot quicker. Not memorizing stuff and figuring things as you go is mostly recipe for disappointment. Like in chess - people think that chess players are somehow super intelligent -…

There's actually an interesting section of the book "Moonwalking with Einstein" where it talks about a study was done on chess masters where they showed them the board in positions that would be impossible under the rules and suddenly the master chess players didn't do all that much better than random people. The suggestion in the book was that really Grandmasters have spent so much time practicing, that they have me…

Off-topic, but I dislike chess and adore chess960/fischerrandom chess for essentially this reason. It's almost disheartening enough to want a different hobby when you review an online game and realize what you thought was a clever solution to an interesting "puzzle" of a board position was really just one you learned by experience a week ago. More variety in piece arrangement makes playing feel far more like doing chess problem solving than remembering the last time you messed up the same chess problem.

Re: Spaced repetition can allow for infinite recall

#112

Post by allenu » Tue, Aug 09, 2022, 5:45 AM UTC

zozbot234 wrote: post #109

allenu wrote: post #48

Earlier quoted context omitted.

That growing backlog is a big issue with this type of scheduling. I think it's one of those almost "unexpected" side-effects of spaced repetition. I mean, if you did the math up front, you should know generally how busy you're going to be over the next few weeks based on how many cards you're reviewing per day, how many new ones are coming, and how you're scoring with each card, but it's not really something people d…

> I think memorizing a smaller set of facts consistently is better In general, existing SRS does a very bad job of supporting priority and prerequisites between items. There's no way of saying "learn A, B, C, then learn X, Y, Z once you know the former well enough." This limits the domains where SRS can be applied most effectively to ones with very little structure, such as raw language vocabulary.

> In general, existing SRS does a very bad job of supporting priority and prerequisites between items.

This is a feature I've wanted to implement for a while, but I think it would be hard from a user's perspective to set up the dependencies. Specifically, I mean marking a card unlearnable until another card (or cards) is learned. It's something that I think requires a good UI so you can see at a glance which cards depend on which other cards. You'd go in and essentially make a skill tree of all the cards. It might be something more useful for more serious deck creators.

My dream is being able to create a deck where you have both vocab and then sentences that use the vocab. You'd make sentences that use the vocab dependent on the vocab being learned first, so that over time you'd gradually unlock more material that uses your base knowledge.

Re: Spaced repetition can allow for infinite recall

#113

Post by bayesian_horse » Tue, Aug 09, 2022, 5:47 AM UTC

tastysandwich wrote: post #82

Based on my own experience, a caveat is that you need to be applying what you're learning outside of flashcards. Otherwise, it's like your brain has different memory banks for different contexts, and you'll remember everything when using your flashcards, but a real-world scenario will come up and you'll forget. I went from zero Japanese to JLPT N2 in just a few years and I attribute a great deal of that to Anki. Howe…

In my opinion the memorization (with whatever method) gets you started. There is no good replacement, and in that phase any other learning activities really aren't as important.

Only when memorization has laid the basic foundation does it make sense to spend more time to read and write in the new language. In my opinion reading is the most efficient activity at higher levels.

Re: Spaced repetition can allow for infinite recall

#114

Post by bayesian_horse » Tue, Aug 09, 2022, 5:58 AM UTC

This is obviously a joke.

But people get spaced repetition wrong. Most people think you need a spaced-repetition schedule for long term memory of facts. You really don't. The only proposition of spaced repetition is that you spend less time studying when using this schedule than with a more dense repetition schedule, if both lead to the same long term memory effect.

Another schedule might be to review your complete vocabulary list every day. This will also lead you to long term memory, eventually. Earlier probably (in terms of calendar time). But you'd spend more hands-on-time studying. However, you'd much prefer the former method over the latter when preparing for an exam on a specific date.

Also: formal methods of managing the schedule aren't really necessary. Reading in a foreign language counts as repetition of the words. Exposure to the language and the vocabulary counts, for example what Duolingo does with its various activities is partly spaced repetition, but less formally so.

Re: Spaced repetition can allow for infinite recall

#115

Post by zozbot234 » Tue, Aug 09, 2022, 6:02 AM UTC

allenu wrote: post #112

zozbot234 wrote: post #109

Earlier quoted context omitted.

> I think memorizing a smaller set of facts consistently is better In general, existing SRS does a very bad job of supporting priority and prerequisites between items. There's no way of saying "learn A, B, C, then learn X, Y, Z once you know the former well enough." This limits the domains where SRS can be applied most effectively to ones with very little structure, such as raw language vocabulary.

> In general, existing SRS does a very bad job of supporting priority and prerequisites between items. This is a feature I've wanted to implement for a while, but I think it would be hard from a user's perspective to set up the dependencies. Specifically, I mean marking a card unlearnable until another card (or cards) is learned. It's something that I think requires a good UI so you can see at a glance which cards de…

Where this feature would be especially useful too is idioms and collocations ("words best used together"), not just full sentences. It's something that's not often addressed in formal language learning but a big part of native fluency. And yes, it's a lot of data. Though it could be made manageable by e.g. importing it from external knowledge graphs/mindmaps and the like.

Re: Spaced repetition can allow for infinite recall

#116

Post by bayesian_horse » Tue, Aug 09, 2022, 6:03 AM UTC

cehrlich wrote: post #50

I've been using Anki for language learning and it's a superpower. After just over two years I passed an exam that requires a passive vocabulary of about 10,000 words. Other people who grinded harder than me have managed it even faster, some in under a year. However one big mistake people make is to think an SRS helps you learn. That's not true. It helps you not forget things you've already learned. You still need som…

To use a language you need "over-learning". It's not enough to be able to recall a word.

You need to be able to recall it very fast, and that only comes through repetition beyond the "I won't ever forget this" stage. SRS doesn't make a statement about this, and you can still use it to achieve that over-learning, in my opinion. It may be better to achieve it through reading texts in the foreign language because the repetitions are faster.

Re: Spaced repetition can allow for infinite recall

#117

Post by bayesian_horse » Tue, Aug 09, 2022, 6:04 AM UTC

Timpy wrote: post #83

borroka wrote: post #56

Earlier quoted context omitted.

>> It helps you not forget things you've already learned. The argument is quite weak when we discuss learning words in a foreign language. You learn that lunch is "almuerzo" in Spanish, but is it learning or just memorization/association? To me the lines are blurred, even at first sight of the word. I used Anki myself to learn foreign languages and it is tremendously effective for words and short sentences too. I use…

I did Anki religiously every day for 2 years, I managed to internalize a bunch of facts about Japanese and no skills. I knew tons of vocab words and grammar points but I couldn't make a sentence in the heat of the moment to save my life. I know memorization is a part of the game, but it's not the whole game. It's like learning a bunch of music theory without actually touching an instrument. I am learning Mandarin now…

Try Duolingo. They do the rote memorization part, but they also provide a lot of sentences. Their Japanese course reportedly is also quite advanced.

Re: Spaced repetition can allow for infinite recall

#118

Post by zozbot234 » Tue, Aug 09, 2022, 6:08 AM UTC

kelseyfrog wrote: post #7

This is false if memory's are stored physically in the brain. Unless there is an increase in brain volume, this process will eventually hit the Bekenstein bound. I don't care if the upper bound is "effectively infinite," that's not what the proof claimed.

That explains it then. I was just trying to memorize something and kept hitting those weird "no free space left on device" errors...

Re: Spaced repetition can allow for infinite recall

#119

Post by barking_biscuit » Tue, Aug 09, 2022, 6:28 AM UTC

allenu wrote: post #112

zozbot234 wrote: post #109

Earlier quoted context omitted.

> I think memorizing a smaller set of facts consistently is better In general, existing SRS does a very bad job of supporting priority and prerequisites between items. There's no way of saying "learn A, B, C, then learn X, Y, Z once you know the former well enough." This limits the domains where SRS can be applied most effectively to ones with very little structure, such as raw language vocabulary.

> In general, existing SRS does a very bad job of supporting priority and prerequisites between items. This is a feature I've wanted to implement for a while, but I think it would be hard from a user's perspective to set up the dependencies. Specifically, I mean marking a card unlearnable until another card (or cards) is learned. It's something that I think requires a good UI so you can see at a glance which cards de…

>My dream is being able to create a deck where you have both vocab and then sentences that use the vocab. You'd make sentences that use the vocab dependent on the vocab being learned first, so that over time you'd gradually unlock more material that uses your base knowledge.

Pretty unnecessary in my experience. I only did around 2k sentences and that was for the express purpose of internalizing grammar, and found that 99 times out of 100 it was pretty natural to just learn the new vocab that came with the sentence. There would be the occasional one that would take a few times before I got it, but it was pretty insignificant problem.

After I smashed out all the grammar I just did purely vocab after that and it worked like an absolute charm.

Re: Spaced repetition can allow for infinite recall

#120

Post by barking_biscuit » Tue, Aug 09, 2022, 6:34 AM UTC

martindbp wrote: post #38

barking_biscuit wrote: post #26

I mean sure. The trouble is in order to keep the review time per day fixed you need to vary the rate at which you introduce new facts, and as time goes on the interval between when you can afford to introduce new facts will grow to be infinite. The model also doesn't account for how you feel about doing this activity. This includes the pain of not doing reviews for several days and then having to catch up. The probab…

Yeah, I've got 9000 reviews waiting for me in my old Chinese Anki deck, at some point you always fall off the wagon. I've changed my mind several times on SRS going back and forth, and my conclusion is that, for language, I think it's better to split your time with 90% on consuming meaningful input and 10% SRS, because learning in context with real content that you're interested in allows you to encode your memories…

I did the AJATT method way back in the day when it was relatively new. So heavily leaned on Anki, whilst also consuming loads of natural Japanese through music, shows, games, conversation etc.

In my experience doing both in tandem was the magic combination. Anki without native media is mildly pointless. Native media without Anki is super inefficient. Used in equal measure you find your access and understanding of native media continually ratchets up day by day every 3000 cards I would notice a very significant jump in my abilities when consuming native media and partaking in conversation.

Nothing else has really come close for me out of all the other things I've tried.