Good information! This is really more “a vote against flashcards” than “a vote against spaced repetition”, though, at least given your concrete issues with flashcards. Spaced repetition is an algorithm for figuring out when to review material that you want to memorize; flashcards are one thing that spaced repetition is applied to, because it’s easy to stick flashcards in a computer. As far as I know, no matter what object-level mnemonic devices you’re using, spaced repetition is still strictly better than “when I feel like I’m forgetting” or “right before a test” or any of the other obvious review strategies, if you can deal with the cognitive load of scheduling things, or get a computer to do it for you.
Is there space for some sort of SRS that allows for input of the more helpful types of memorizations that you listed (pictures, venn diagrams, etc.)?
You are absolutely correct; this is a hair worth splitting. I meant “spaced repetition flashcards”, and I have only seen formal spaced repetition algorithms applied to flashcards. In my particular case, I end up with 30 or so “pages” of related information, as opposed to 500 flashcards. I agree that using spaced repetition algorithms to tell me when to study which page is likely better than alternative methods, though I haven’t found an algorithm optimized for that sort of thing, and at the moment my intuition of “when I’m forgetting” is sufficient for the low number of separate objects to study.
[For this comment, I will use the term “page” to mean any collection of related information, be it a list, table, memory palace, notes on a single topic, etc.]
To be explicit: I vote against using spaced repetition (of any sort) to identify specific facts within a “page” of information. When reviewing a page, of course you can go quickly over the parts you know well and dwell on the parts you don’t, but I would encourage the student to not completely ignore the other details “until it’s time.”
As an example: I have a collection of facts that can be represented as a large table or as individual facts. If I study it in a table, then I get the advantage of keeping the “big picture in mind”, plus I can activate spatial memory as well as rote. If I study it as separate facts: the Pro is I can use spaced repetition to greater effect, not reviewing the parts I know better, but the loss of the picture and the spatial memory makes it not worth the cost. (Note that the “big picture” isn’t a single sentence I can write down; it’s noticing trends in the data, how column A and B are similar except in in key areas, etc.)
As always, my experience is only in high volumes of information that can be organized vaguely hierarchically. (That said, I think if you look hard enough, you can find categories or hierarchies for any large volume of information, outside of truly random things like the sequence of a deck of cards.)
Caveat #2: “self testing” is really important! So if you are quitting flash cards, make sure you find some new way to quiz yourself, don’t just read passively.
Conclusion:
-Spaced repetition algorithms might be viable, though I don’t know any suitable to my needs
-I claim that spaced repetition flashcards are not useful for large volumes that have categories and/or relationships between facts.
-benkuhn rightfully points out that my vendetta is mostly against flashcards, or any method focusing on “terminal facts” in random order without also studying closely related facts.
Life application:
1) If you see a table, don’t vaporize it into your flashcards. Rather, study the table.
2) If you see a mass of new data: ask yourself if you can organize it in a way meaningful to you, then study it in the structure you built.
If I study it as separate facts: the Pro is I can use spaced repetition to greater effect, not reviewing the parts I know better, but the loss of the picture and the spatial memory makes it not worth the cost. (Note that the “big picture” isn’t a single sentence I can write down; it’s noticing trends in the data, how column A and B are similar except in in key areas, etc.)
Why not do both? The thing about SRS flashcards is that facts that you have strong recall of are nearly costless to add, because the review schedules get spaced into very large intervals (weeks, then months or even years), and it only takes a split second to see a card and realize it’s easy. So you could learn from tables if that really is more efficient for you, but having the data in SRS card form as well is a good insurance policy.
Frankly, because at the volume I was running, it was far too great an investment of time. When I stopped, I had about 75-100 scheduled (learned) flashcards per day if I added nothing the day before, though I usually added 60-some every day. The cards would take me 1-2 hours, and the amount kept building as I was adding to it faster than I was pushing them “out”.
Additionally, here our mileage may vary, but even with easy flashcards I occasionally find myself staring dumbly at it for ten or more seconds before I realize what it’s asking and smack my head. So I end up trimming out the stupid-easy ones, but that starts to defeat the purpose. Thus, for myself personally, I won’t duplicate in flashcards what I’m already memorizing elsewhere.
I know that everyone is different, so this is just my experience and what I have observed in other people. If others continue to have success with SRS, then far be it from me to insist they fix what isn’t broken.
even with easy flashcards I occasionally find myself staring dumbly at it for ten or more seconds
To me, that suggests that the card is either too complex and should be split up further, or that you simply do not have solid recall of the relevant facts, so you should just flip to the answer and mark the card difficult. It’s quite normal to forget even some basic info over time; the point of SRS is to refresh these memories at the lowest viable cost.
I appreciate the input, truly, but I can confidently state that’s not the case in my situation. This happens even on the simplest questions that I know cold, and is a problem with mental fatigue, monotony, and reading. After the 100th card, I would expect similar results from “what color is the sky” occasionally. I highly doubt I am dyslexic, but I might be a little ADHD. Once again, I do not presume everyone has similar results, but when I did 150 cards per day (and lord help me if I missed a day), easy cards posed a significant drain on my time and mental energy.
Interesting. If you get that kind of mental fatigue from reading, maybe flashcards really are relatively inefficient for you. If it turns out that dyslexia is the problem, there is an open source font that can help with that issue. Some people have set their SRS up to read questions aloud using computer-generated speech. But yes, most of the time it’s a signal that you should take a break and perhaps switch to some other activty.
One thing that’s relevant to this discussion is that the latest SRS versions can actually cope quite well with missed reviews. Yes, you’ll still be presented with a backlog of suggestions, but the system gives you improved credit if you can recall a card easily even after the increased delay. Because that implies your memory of it was quite good in the first place, so it can be refreshed less often in the future.
Anki is very extensible. I think writing easy-to-use Anki plugins would be a great way to practice coding and get some useful stuff out there. In fact, I’m adding that to my list of things to look into.
This is an idea I had only toyed with but have yet to try in practice, but one can create meta-cards for non-data learning. Instead of creating cards that demand an answer, create cards that demand a drill, or a drill with a specific success outcome.
I find it a bit hard to find “the best example” for this, perhaps because the spectrum of learnable-skills is so broad, but just for the sake of illustration: if you’re learning to paint, you can have “draw a still object”, “draw a portrait”, “practice color”, “practice right composition”, “practice perspective” &c, cards. After you finish your card-prompted drill, you move to the next card.
Or if you’re practicing going pro at a game (with existing computer program AIs), you can have “Play AI X in a game situation S and achieve A”, “Practice game opening against AI until (able to reach a certain state)”, “practice a disadvantaged end-game situation against AI and bring the game to a draw”, and so on, cards.
Of course reviewing the cards would take longer, but they are only meant to be used as scaffolding to harness the Anki spacing algorithm. The numeric parameters of the algorithm might need an adjustment (which is easy to do in Anki) for that, but I think that qualitatively it should work, at least for specific skills. Of course, this set-up, especially if it needs a major parametric-overhauling[1], is an investment, but every human breakthrough required its avant-gardians.
[1] Which is not granted: perhaps the algorithm is only problematic at the beginning of the “learning”, being too frequent, in which case you can just “cheat” carefully and “pass” every other review for a while, which is not a major disturbance. Or, on the contrary, perhaps “well learned cards” (interval > 3 months, or even 1 month, for example) should be discarded for more challenging ones (i.e, “beat the expert AI” replacing “beat beginner AI”, or “juggle 5 balls while riding a unicycle on a mid-air rope” replacing “juggle 4 balls”), which is even less of a problem, as you should immediately recognize well-learned skills (i.e. “practice counting up to 20”).
Is there space for some sort of SRS that allows for input of the more helpful types of memorizations that you listed (pictures, venn diagrams, etc.)?
Unfortunately, Anki and other SRS software do not seem to support card dependencies (i.e. “only show card X if cards Y, Z… are firmly set in memory, as predicted by the spaced-repetition model”). If that was supported, using SRS to memorize heavily-structured data would simply be a matter of setting up appropriate dependencies. (A “memory palace” is really the same thing, except that the highest level of your hierarchy is a spatial model, i. e., your “palace” or “room”, containing the information you want to memorize.)
(One could generalize this by also supporting the option: “prioritize card X when cards Y, Z… are more easily recalled.” Then the deck could even include such things as mutual dependencies, loose associations etc. and the app could use them to “branch out” from what you know already, showing info that can most easily be committed to memory in a highly clustered way.)
Unfortunately, Anki and other SRS software do not seem to support card dependencies (i.e. “only show card X if cards Y, Z… are firmly set in memory, as predicted by the spaced-repetition model”). If that was supported, using SRS to memorize heavily-structured data would simply be a matter of setting up appropriate dependencies
The general idea of Anki is that you learn the knowledge first and then put it into Anki to avoid forgetting it.
The general idea of Anki is that you learn the knowledge first and then put it into Anki to avoid forgetting it.
Not necessarily. In some cases, the flashcard format is quite suited for learning new content as well—especially such things as vocabulary. Allowing inter-card dependencies could easily expand on these use cases.
It would also be directly useful in language learning: for instance, you could memorize some vocabulary words and then be prompted to learn related idioms, or collocations (i.e. words that are “often used together”). Despite its usefulness, this content is quite hard to memorize effectively in the absence of such specialized support.
This is not quite a “tech-tree” dependency structure, but you can use tags to stratify your cards and always review them in sequence from basic to dependent (i.e., first clear out the “basic” cards, then “intermediate”, then “expert”). Even if the grouping is arbitrary, I think you can go a long way with it. If your data is expected to be very large and/or have a predictable structure, you can always go for a “multiple-pyramid” structure, i.e, have “fruits basic” < “fruits advanced” < “fruits expert”, “veggies basics” < “veggies pro” tags &c, and perhaps even have an “edibles advanced” > veggies & fruits tag for very dependent cards.
On the assumption that the Anki algorithm works, just “reviewing down” to an empty deck every tag and proceeding thus sequentially from tag to tag, I think this would work too. Even if it so happened that by one Sunday you forgot “What is an American president” (basic) fact, it might still be profitable to rehearse that day the “Washington was the first president” card, despite the “20 rules” mentioned somewhere above. Presumably, if you had forgotten what a president is, the appropriate card is probably going to appear for review in the next few days, and so with a consistent (or even a semi-consistent) use of Anki, it would probably turn alright.
This is more for the anecdotal sake, but this reminds me a time when I burst out laughing out loud while at the dictionary. I was reading at the time “Three Men in a Boat”, and there was one sentence in which I didn’t know 2-3 of the words; the punchline clicked as I read the definition of the last of them.
Either way, somewhere higher on this commenting thread, I have also thought about the possibility (or rather, lack of) of creating dependencies in Anki. I’m actually thinking of creating an add-on/plugin to enable that—I’m learning Python these days (on which Anki runs), and I’m just about to start grad school (if I get admitted), so it seems like just the right time to make this (possibly major) meta-learning investment.*
* Not to mention that, since I’m learning Python, it’s also a (non-meta) learning investment. Win-win.
Good information! This is really more “a vote against flashcards” than “a vote against spaced repetition”, though, at least given your concrete issues with flashcards. Spaced repetition is an algorithm for figuring out when to review material that you want to memorize; flashcards are one thing that spaced repetition is applied to, because it’s easy to stick flashcards in a computer. As far as I know, no matter what object-level mnemonic devices you’re using, spaced repetition is still strictly better than “when I feel like I’m forgetting” or “right before a test” or any of the other obvious review strategies, if you can deal with the cognitive load of scheduling things, or get a computer to do it for you.
Is there space for some sort of SRS that allows for input of the more helpful types of memorizations that you listed (pictures, venn diagrams, etc.)?
You are absolutely correct; this is a hair worth splitting. I meant “spaced repetition flashcards”, and I have only seen formal spaced repetition algorithms applied to flashcards. In my particular case, I end up with 30 or so “pages” of related information, as opposed to 500 flashcards. I agree that using spaced repetition algorithms to tell me when to study which page is likely better than alternative methods, though I haven’t found an algorithm optimized for that sort of thing, and at the moment my intuition of “when I’m forgetting” is sufficient for the low number of separate objects to study.
[For this comment, I will use the term “page” to mean any collection of related information, be it a list, table, memory palace, notes on a single topic, etc.]
To be explicit: I vote against using spaced repetition (of any sort) to identify specific facts within a “page” of information. When reviewing a page, of course you can go quickly over the parts you know well and dwell on the parts you don’t, but I would encourage the student to not completely ignore the other details “until it’s time.”
As an example: I have a collection of facts that can be represented as a large table or as individual facts. If I study it in a table, then I get the advantage of keeping the “big picture in mind”, plus I can activate spatial memory as well as rote. If I study it as separate facts: the Pro is I can use spaced repetition to greater effect, not reviewing the parts I know better, but the loss of the picture and the spatial memory makes it not worth the cost. (Note that the “big picture” isn’t a single sentence I can write down; it’s noticing trends in the data, how column A and B are similar except in in key areas, etc.)
As always, my experience is only in high volumes of information that can be organized vaguely hierarchically. (That said, I think if you look hard enough, you can find categories or hierarchies for any large volume of information, outside of truly random things like the sequence of a deck of cards.)
Caveat #2: “self testing” is really important! So if you are quitting flash cards, make sure you find some new way to quiz yourself, don’t just read passively.
Conclusion: -Spaced repetition algorithms might be viable, though I don’t know any suitable to my needs -I claim that spaced repetition flashcards are not useful for large volumes that have categories and/or relationships between facts. -benkuhn rightfully points out that my vendetta is mostly against flashcards, or any method focusing on “terminal facts” in random order without also studying closely related facts.
Life application: 1) If you see a table, don’t vaporize it into your flashcards. Rather, study the table. 2) If you see a mass of new data: ask yourself if you can organize it in a way meaningful to you, then study it in the structure you built.
Why not do both? The thing about SRS flashcards is that facts that you have strong recall of are nearly costless to add, because the review schedules get spaced into very large intervals (weeks, then months or even years), and it only takes a split second to see a card and realize it’s easy. So you could learn from tables if that really is more efficient for you, but having the data in SRS card form as well is a good insurance policy.
Frankly, because at the volume I was running, it was far too great an investment of time. When I stopped, I had about 75-100 scheduled (learned) flashcards per day if I added nothing the day before, though I usually added 60-some every day. The cards would take me 1-2 hours, and the amount kept building as I was adding to it faster than I was pushing them “out”.
Additionally, here our mileage may vary, but even with easy flashcards I occasionally find myself staring dumbly at it for ten or more seconds before I realize what it’s asking and smack my head. So I end up trimming out the stupid-easy ones, but that starts to defeat the purpose. Thus, for myself personally, I won’t duplicate in flashcards what I’m already memorizing elsewhere.
I know that everyone is different, so this is just my experience and what I have observed in other people. If others continue to have success with SRS, then far be it from me to insist they fix what isn’t broken.
To me, that suggests that the card is either too complex and should be split up further, or that you simply do not have solid recall of the relevant facts, so you should just flip to the answer and mark the card difficult. It’s quite normal to forget even some basic info over time; the point of SRS is to refresh these memories at the lowest viable cost.
I appreciate the input, truly, but I can confidently state that’s not the case in my situation. This happens even on the simplest questions that I know cold, and is a problem with mental fatigue, monotony, and reading. After the 100th card, I would expect similar results from “what color is the sky” occasionally. I highly doubt I am dyslexic, but I might be a little ADHD. Once again, I do not presume everyone has similar results, but when I did 150 cards per day (and lord help me if I missed a day), easy cards posed a significant drain on my time and mental energy.
Interesting. If you get that kind of mental fatigue from reading, maybe flashcards really are relatively inefficient for you. If it turns out that dyslexia is the problem, there is an open source font that can help with that issue. Some people have set their SRS up to read questions aloud using computer-generated speech. But yes, most of the time it’s a signal that you should take a break and perhaps switch to some other activty.
One thing that’s relevant to this discussion is that the latest SRS versions can actually cope quite well with missed reviews. Yes, you’ll still be presented with a backlog of suggestions, but the system gives you improved credit if you can recall a card easily even after the increased delay. Because that implies your memory of it was quite good in the first place, so it can be refreshed less often in the future.
Anki is very extensible. I think writing easy-to-use Anki plugins would be a great way to practice coding and get some useful stuff out there. In fact, I’m adding that to my list of things to look into.
This is an idea I had only toyed with but have yet to try in practice, but one can create meta-cards for non-data learning. Instead of creating cards that demand an answer, create cards that demand a drill, or a drill with a specific success outcome. I find it a bit hard to find “the best example” for this, perhaps because the spectrum of learnable-skills is so broad, but just for the sake of illustration: if you’re learning to paint, you can have “draw a still object”, “draw a portrait”, “practice color”, “practice right composition”, “practice perspective” &c, cards. After you finish your card-prompted drill, you move to the next card. Or if you’re practicing going pro at a game (with existing computer program AIs), you can have “Play AI X in a game situation S and achieve A”, “Practice game opening against AI until (able to reach a certain state)”, “practice a disadvantaged end-game situation against AI and bring the game to a draw”, and so on, cards. Of course reviewing the cards would take longer, but they are only meant to be used as scaffolding to harness the Anki spacing algorithm. The numeric parameters of the algorithm might need an adjustment (which is easy to do in Anki) for that, but I think that qualitatively it should work, at least for specific skills. Of course, this set-up, especially if it needs a major parametric-overhauling[1], is an investment, but every human breakthrough required its avant-gardians.
[1] Which is not granted: perhaps the algorithm is only problematic at the beginning of the “learning”, being too frequent, in which case you can just “cheat” carefully and “pass” every other review for a while, which is not a major disturbance. Or, on the contrary, perhaps “well learned cards” (interval > 3 months, or even 1 month, for example) should be discarded for more challenging ones (i.e, “beat the expert AI” replacing “beat beginner AI”, or “juggle 5 balls while riding a unicycle on a mid-air rope” replacing “juggle 4 balls”), which is even less of a problem, as you should immediately recognize well-learned skills (i.e. “practice counting up to 20”).
Unfortunately, Anki and other SRS software do not seem to support card dependencies (i.e. “only show card X if cards Y, Z… are firmly set in memory, as predicted by the spaced-repetition model”). If that was supported, using SRS to memorize heavily-structured data would simply be a matter of setting up appropriate dependencies. (A “memory palace” is really the same thing, except that the highest level of your hierarchy is a spatial model, i. e., your “palace” or “room”, containing the information you want to memorize.)
(One could generalize this by also supporting the option: “prioritize card X when cards Y, Z… are more easily recalled.” Then the deck could even include such things as mutual dependencies, loose associations etc. and the app could use them to “branch out” from what you know already, showing info that can most easily be committed to memory in a highly clustered way.)
The general idea of Anki is that you learn the knowledge first and then put it into Anki to avoid forgetting it.
Not necessarily. In some cases, the flashcard format is quite suited for learning new content as well—especially such things as vocabulary. Allowing inter-card dependencies could easily expand on these use cases.
It would also be directly useful in language learning: for instance, you could memorize some vocabulary words and then be prompted to learn related idioms, or collocations (i.e. words that are “often used together”). Despite its usefulness, this content is quite hard to memorize effectively in the absence of such specialized support.
This is not quite a “tech-tree” dependency structure, but you can use tags to stratify your cards and always review them in sequence from basic to dependent (i.e., first clear out the “basic” cards, then “intermediate”, then “expert”). Even if the grouping is arbitrary, I think you can go a long way with it. If your data is expected to be very large and/or have a predictable structure, you can always go for a “multiple-pyramid” structure, i.e, have “fruits basic” < “fruits advanced” < “fruits expert”, “veggies basics” < “veggies pro” tags &c, and perhaps even have an “edibles advanced” > veggies & fruits tag for very dependent cards.
On the assumption that the Anki algorithm works, just “reviewing down” to an empty deck every tag and proceeding thus sequentially from tag to tag, I think this would work too. Even if it so happened that by one Sunday you forgot “What is an American president” (basic) fact, it might still be profitable to rehearse that day the “Washington was the first president” card, despite the “20 rules” mentioned somewhere above. Presumably, if you had forgotten what a president is, the appropriate card is probably going to appear for review in the next few days, and so with a consistent (or even a semi-consistent) use of Anki, it would probably turn alright. This is more for the anecdotal sake, but this reminds me a time when I burst out laughing out loud while at the dictionary. I was reading at the time “Three Men in a Boat”, and there was one sentence in which I didn’t know 2-3 of the words; the punchline clicked as I read the definition of the last of them.
Either way, somewhere higher on this commenting thread, I have also thought about the possibility (or rather, lack of) of creating dependencies in Anki. I’m actually thinking of creating an add-on/plugin to enable that—I’m learning Python these days (on which Anki runs), and I’m just about to start grad school (if I get admitted), so it seems like just the right time to make this (possibly major) meta-learning investment.*
* Not to mention that, since I’m learning Python, it’s also a (non-meta) learning investment. Win-win.
It was to be expected-- Someone had already created a “hierarchy Tags” addon: https://ankiweb.net/shared/info/1089921461
I haven’t used it myself, but a comment there said “Simple, nice, and easy.”