4 ms·

1. Spaced repetition is for the end of the knowledge acquisition process. It crystallizes what you already understand pretty well. Sugar is for the very begin

1. Spaced repetition is for the end of the knowledge acquisition process. It crystallizes what you already understand pretty well. Sugar is for the very beginning: making kids aware that some subjects are interesting enough to learn. 2. Spaced repetition is boring, especially for kids.

Havoc 7y ago
Don't see why a tool can't actively manage both - introducing new concepts and re-inforcing them on the right schedule. Yes you don't want a rigid spaced rep schedule...but I'd like to see some of that integrated in a soft fashion anyway
em-bee 7y ago
if kids choose to play an activity over and over again (not continuously, but every day or so pick it up and play for a while), is that not also spaced repetition? how would that be boring?
mikekchar 7y ago
I'm going to disagree with you, but I'm actually very interested to hear your reasoning behind your comments on spaced repetition. It is a very interesting perspective that I haven't heard before so I'm quite keen to hear more. Having said that, here's my point of view. Spaced repetition is not really directly involved with knowledge acquisition. When I say "acquisition" I have a very specific meaning in mind. In recent language acquisition theory there is a distinction between "learning" and "acquisition". Learning means to be able to go from not knowing something to being able to remember something. Acquisition means to go from not being able to use something to being able to use it appropriately without undue effort. Learning does not always lead to acquisition (something that's still not penetrated the organisations that do education, unfortunately). Spaced repetition is wonderful for learning. You can go from not knowing something to being able to remember something for a very long time with minimal effort. Again, it does not necessarily lead to acquisition. From the perspective of language, to acquire language you need to be exposed to it in realistic contexts where you understand the meaning. Personally, I have found that learning can help with acquisition because it kind of gives you a super fast dictionary to look things up and allow you to comprehend things when you are exposed to them. You need a lot of exposure and comprehension to actually acquire the knowledge, though -- and spaced repetition alone won't get you there. I think the same is true of all skill acquisition, but there is a lot of research to be done to show that (as an example, imagine learning all the rules of calculus -- you still won't be able to write proofs easily, but it will help you to save massive amounts of time when you start studying proofs). Where I think you might be coming from is that there is a period when learning something where you just can't remember it for a significant amount of time. In most spaced repetition software (like Anki), you go from a 10 minute review spacing to a whole day -- and even going from 1 minute to 10 minutes can often be a jump. What's interesting about that is that there is a different concept called "spacing" (which is unfortunately very similarly named to "spaced repetition"). With "spacing" the idea is that you will reduce the slope of the forgetting curve faster if you forget something. This is a crazy idea, but it seems to work very well. If you want to read a tonne of papers on the subject, google "desirable difficulties". So my idea is that you initially want to space a repetition longer than you would for spaced repetition so that you have to struggle to remember it. This will help you learn it faster. At that point you should switch over to spaced repetition. I say "my idea", because there are very few (or possibly none ;-) ) papers that discuss mixing spacing and desirable difficulties with spaced repetition.... If only I were a psychology prof... Anyway, there is another concept called "Interleaving". With interleaving, what you do is intentionally change the subject frequently while you are learning something. This makes it difficult to remember and potentially reduces the amount of spacing you require to get the "spacing effect". Many papers suggest changing subjects every 10 minutes (which seems crazy -- imagine 10 minutes of math, 10 minutes of English, 10 minutes of Foreign language, 10 minutes of geography, then back to 10 minutes of math, etc, etc, etc -- but the studies I've read show really incredible rates of learning) But again, the key is that learning is distinct from acquisition -- so while learning is useful it doesn't necessarily lead to skill acquisition. As for #2 I don't think spaced repetition is boring at all -- especially for kids. Give a small child a cartoon on a DVD. Watch as they repeat it over and over and over and over again. Watch at they view the video today, the same video tomorrow, the same video the next day. It's really incredible. Kids love that kind of repetition. I get your point that Anki might be boring ;-)
- mnemonicsloth 7y ago
  > If you want to read a tonne of papers on the subject, google "desirable difficulties". I do. I memorize a lot of stuff (I'm a medical student) and my life will be way less stressful if almost remembering something means you're on the right track. You probably aren't reading this, but thanks for the pointer.

Havoc 7y ago

Don't see why a tool can't actively manage both - introducing new concepts and re-inforcing them on the right schedule. Yes you don't want a rigid spaced rep schedule...but I'd like to see some of that integrated in a soft fashion anyway

em-bee 7y ago

if kids choose to play an activity over and over again (not continuously, but every day or so pick it up and play for a while), is that not also spaced repetition? how would that be boring?

mikekchar 7y ago

I'm going to disagree with you, but I'm actually very interested to hear your reasoning behind your comments on spaced repetition. It is a very interesting perspective that I haven't heard before so I'm quite keen to hear more. Having said that, here's my point of view. Spaced repetition is not really directly involved with knowledge acquisition. When I say "acquisition" I have a very specific meaning in mind. In recent language acquisition theory there is a distinction between "learning" and "acquisition". Learning means to be able to go from not knowing something to being able to remember something. Acquisition means to go from not being able to use something to being able to use it appropriately without undue effort. Learning does not always lead to acquisition (something that's still not penetrated the organisations that do education, unfortunately). Spaced repetition is wonderful for learning. You can go from not knowing something to being able to remember something for a very long time with minimal effort. Again, it does not necessarily lead to acquisition. From the perspective of language, to acquire language you need to be exposed to it in realistic contexts where you understand the meaning. Personally, I have found that learning can help with acquisition because it kind of gives you a super fast dictionary to look things up and allow you to comprehend things when you are exposed to them. You need a lot of exposure and comprehension to actually acquire the knowledge, though -- and spaced repetition alone won't get you there. I think the same is true of all skill acquisition, but there is a lot of research to be done to show that (as an example, imagine learning all the rules of calculus -- you still won't be able to write proofs easily, but it will help you to save massive amounts of time when you start studying proofs). Where I think you might be coming from is that there is a period when learning something where you just can't remember it for a significant amount of time. In most spaced repetition software (like Anki), you go from a 10 minute review spacing to a whole day -- and even going from 1 minute to 10 minutes can often be a jump. What's interesting about that is that there is a different concept called "spacing" (which is unfortunately very similarly named to "spaced repetition"). With "spacing" the idea is that you will reduce the slope of the forgetting curve faster if you forget something. This is a crazy idea, but it seems to work very well. If you want to read a tonne of papers on the subject, google "desirable difficulties". So my idea is that you initially want to space a repetition longer than you would for spaced repetition so that you have to struggle to remember it. This will help you learn it faster. At that point you should switch over to spaced repetition. I say "my idea", because there are very few (or possibly none ;-) ) papers that discuss mixing spacing and desirable difficulties with spaced repetition.... If only I were a psychology prof... Anyway, there is another concept called "Interleaving". With interleaving, what you do is intentionally change the subject frequently while you are learning something. This makes it difficult to remember and potentially reduces the amount of spacing you require to get the "spacing effect". Many papers suggest changing subjects every 10 minutes (which seems crazy -- imagine 10 minutes of math, 10 minutes of English, 10 minutes of Foreign language, 10 minutes of geography, then back to 10 minutes of math, etc, etc, etc -- but the studies I've read show really incredible rates of learning) But again, the key is that learning is distinct from acquisition -- so while learning is useful it doesn't necessarily lead to skill acquisition. As for #2 I don't think spaced repetition is boring at all -- especially for kids. Give a small child a cartoon on a DVD. Watch as they repeat it over and over and over and over again. Watch at they view the video today, the same video tomorrow, the same video the next day. It's really incredible. Kids love that kind of repetition. I get your point that Anki might be boring ;-)

mnemonicsloth 7y ago
> If you want to read a tonne of papers on the subject, google "desirable difficulties". I do. I memorize a lot of stuff (I'm a medical student) and my life will be way less stressful if almost remembering something means you're on the right track. You probably aren't reading this, but thanks for the pointer.