@Daichi I think you're correct about me not spreading my scores. I usually don't take into account how easy or hard the card is, i'll just hit space if I remember it. I hadn't thought about it that until you brought it up. The other options you mentioned might be helpful as well.

Thank you very much!

[RawToast]

Daichi Wrote:The other option you might wanna consider is increasing the "New Interval" on lapses from 0% to something like 5%, so a fail won't completely decimate your interval just for a fail.

Thanks for that. My mature card rate is 96% at the moment, but I hated when I got a mature card wrong and I had to go through the 1d, 2d, etc steps and ended up pressing easy every time! I have set it to 10% so a 30 day card would fall to 3 days, or a 60 day one to 6 days. That way the card deserves to be reset if you get it wrong the second time around.

Saiga - I highly reccommend the sorted core. I find my highest failure rates came form vocab outside of the ordering (e.g. I did JLPT cards at the same time.)

Saiga Wrote:Not usually, I have lately been trying to dig through Tae Kim before doing any native material.
I have made it so that leeches are only tagged, I have 115 leeches at this point. Perhaps I should suspend them.

So you have a vocabulary up in the thousands and still don't use it? Encountering words in their native context is important. Even if it is something like NHK Easy News, reading something every day will help you remember stuff; there is more to this language than example sentences.

A thing that people should keep in mind is that once your level of vocabulary reaches a certain point, you no longer really need to study vocabulary in order to learn more of it. I believe the transition point is somewhere in the neighborhood of 15,000-20,000 words.

I've recently gotten to that point, and I can guess meanings from context quite reliably. I can also just spontaneously understand a lot of idiomatic stuff reflexively just from having gotten used to the language. Recently someone asked about 踊り食い, and I wondered for only a split second. Then it instantly popped into my head what that word must mean. The same thing happened a few days later with someone talking about 耳鳴り.

Your handling of new vocabulary in your foreign language, ideally, will eventually become similar to the way you handle new vocabulary in your native language. Maybe it won't be the same, but you'll start to have similar thought processes when you come upon a new word.

Also, a bit of advice to the people who aren't achieving the kind of 85%-90% retention that a lot of us are. There's a lot of us here that have been studying Japanese for a long time. I'm getting that kind of retention now, but I definitely wasn't getting that kind of retention at the beginning. It was kind of a slog for a while.

That changes over time as you learn new words, though. Learning and remembering vocabulary gets easier as you go. You get used to the internal logic of the language, and so you start having the feeling that a lot of things are 当たり前 despite the fact that you at point had to explicitly learn them.

You can get there at some point if you keep pushing forward. You won't realize it's happening while it's happening either. I've spent most of my time living in Japan recently feeling like my Japanese isn't getting better. However when I look back on old material or look through letters I used to send to my parents, it's clear that my abilities actually have improved.

I also agree with everyone saying it's a use it or lose it kind of proposition. If I learn a word through flashcards or a textbook that word kind of feels fuzzy in my head. Then if I have some more exposures of it in different contexts outside of my flashcards then it becomes a lot clearer. The flashcards I see as priming myself for things I'll eventually gain a deeper understanding of upon encountering in the real world. They also act as a reminder to make sure I don't forget those things. They aren't the be all and end all. They're maintenance for your language skill.

Quote:A thing that people should keep in mind is that once your level of vocabulary reaches a certain point, you no longer really need to study vocabulary in order to learn more of it. I believe the transition point is somewhere in the neighborhood of 15,000-20,000 words.

I'd say even that many words isn't a necessity. My English vocab is somewhere around 12,000 known words and probably half of it was learnt "in context". I was lazy in school but active in the interwebs.

More important is probably actually getting used to the language, not necessarily the amount of vocab one learns. There are so many words that I "know" but can't translate to Finnish or explicitly define them without looking at my English dictionary. Many times a word pops in my head and I am left wondering "how the hell do I even know this word and how to use it?"

Perhaps Japanese is different...but certainly the kanji should help one decipher the meaning.

Yeah, I think that's definitely true. If those 12,000 words were sufficiently varied enough to give you a good handle on the internal logic of the language then it would definitely be possible to reach that point earlier. I learned a lot of my vocabulary in the context of sentences, but not in the context of larger works. I also built up my vocabulary through vocabulary textbooks where you learn sets of words at a time.

I'd learn like 50 new words in a sitting, but they were all related to a certain domain. I was increasing my vocabulary, but not necessarily increasing my understanding of that internal logic.

I also understand what you mean about knowing what a word means, but not being able to explain it in a different language. There's quite a lot of words like that for me in Japanese. They just don't map very well to western concepts or concepts in English. There's a lot of words with profoundly deep and complex meanings that you'd just be at a complete loss to adequately describe without having to type an entire paragraph describing situations or giving cultural context.

When you look things up in a dictionary it can give you a handful of similar words in your native language that put you in the right ballpark, but they pale in comparison to having an actual intuitive understanding of the word.

Also your brain starts to just compartmentalize itself. I think about Japanese in Japanese. Even explanations of simple words sometimes momentarily elude me until I can reboot my brain into English mode.

Granted, I am quite sure one can speed up the process by using tools like Anki and actually trying to learn new words, instead of letting them accumulate naturally. My knowledge had been built up over a span of maybe 10 years, so there was more than enough time to really internalize all the vocab I came in contact with. To the learners of Japanese here, 10 years would probably be unacceptable. But let's be clear, given enough time even Core6k's worth of vocab "studying" should suffice if one wants to communicate in and comprehend Japanese like I do in English.

nadiatims Wrote:

corry Wrote:

nadiatims Wrote:As for core6k, I tried stuff like that when i first started and found it extremely boring and demotivating.

What did you start with then before you could understand normal Japanese stuff.

Initially after doing RTK I figured the next step should be similarly organized I and searched around and found different word lists and tried various systems for learning things in a rather hap hazard manner. After doing RTK (and this is possibly one of its disadvantages) I somehow thought that learning the language would simply involve learning and mastering each component separately via some book or system that spoon feeds me what I need to learn. After abandoning that, I just ended up reading and listening to a lot of varied media. A lot of it was rather hit and miss. Iirc I think manga was really good, as were junior high school level books, but i also started trying to read stuff way out of my league which was painful to say the least, but i'm sure i got something out of it. I think my listening improved leaps and bounds after i started recording any japanese videos I watched (youtube/anime) and then looped them on my ipod. I found this type of listening practice much more useful than listening to non-watched content or music.

Thanks for your reply. I think I will have to try this looping technique.

But I am wondering how you were able to understand any spoken Japanese with only RTK. Did you manage to find videos with Japanese transcripts or something?

corry Wrote:

nadiatims Wrote:

corry Wrote:What did you start with then before you could understand normal Japanese stuff.

Initially after doing RTK I figured the next step should be similarly organized I and searched around and found different word lists and tried various systems for learning things in a rather hap hazard manner. After doing RTK (and this is possibly one of its disadvantages) I somehow thought that learning the language would simply involve learning and mastering each component separately via some book or system that spoon feeds me what I need to learn. After abandoning that, I just ended up reading and listening to a lot of varied media. A lot of it was rather hit and miss. Iirc I think manga was really good, as were junior high school level books, but i also started trying to read stuff way out of my league which was painful to say the least, but i'm sure i got something out of it. I think my listening improved leaps and bounds after i started recording any japanese videos I watched (youtube/anime) and then looped them on my ipod. I found this type of listening practice much more useful than listening to non-watched content or music.

Thanks for your reply. I think I will have to try this looping technique.

But I am wondering how you were able to understand any spoken Japanese with only RTK. Did you manage to find videos with Japanese transcripts or something?

Well at first I couldn't understand much at all. However, before discovering and starting RTK I had already been living in Japan and studying japanese for about 5 months. So I knew hiragana, katakana, some vocabulary (less than 500?) and basic grammar. I passed the old 3kyuu jlpt exam just before starting rtk so that gives you some idea of my level. Ie. very low but not a complete beginner.
After spending maybe a year after that dicking around with various random paper and online grammar/vocab resources and passing 2kyuu, What I eventually realised is that the only thing that was really having an appreciable effect was reading/listening/using real japanese and I started trying to do more of that. My listening ability lagged massively for a very long time and I think there a couple of reasons for this. One is that this is normal for beginners. Another is that Much of my study time was at work at my desk with books (I couldn't really watch anime or youtube at work). Another reason is that I was listening to stuff which provided little to no context (mostly music). Eventually I realised that i tended to understand speech when I knew the words (sounds obvious right) and also when I could understand from context roughly what was being spoken about. Someone (i forget who) once said something like "you can only learn what you already know." Once I realised this I started watching a lot of anime and youtube and I started focusing on vocabulary. You'll have a much better chance of recognising and understanding in context a word that you have already learned somewhere else. I hope this answers your question.

I think the OP and numerous other posters are completely correct in pointing out that you need to know good 10-20k words before you'll start really understand most things in their entirety.

@nadiatims: You got me thinking just now. I think I'm going to go back to listening to "Dig" again and prioritize it over the NHK JP news. (I only have so much time in the day to listen to stuff.) I think it'll help my everyday Japanese more than the NHK news will. (Although the NHK news is still pretty handy for picking up a ton of politically-related Kango.) XD

I'll still get some news coverage from "Dig," because they spend a lot of time talking about the news, but it won't be in that NHK Announcer style that always puts my brain to sleep. Maybe I'll just get the 15 minute NHK news instead of the 45 minute version.

Either way, if I catch myself tuning out, then it means it's time to change.

nadiatims Wrote:

corry Wrote:

nadiatims Wrote:Initially after doing RTK I figured the next step should be similarly organized I and searched around and found different word lists and tried various systems for learning things in a rather hap hazard manner. After doing RTK (and this is possibly one of its disadvantages) I somehow thought that learning the language would simply involve learning and mastering each component separately via some book or system that spoon feeds me what I need to learn. After abandoning that, I just ended up reading and listening to a lot of varied media. A lot of it was rather hit and miss. Iirc I think manga was really good, as were junior high school level books, but i also started trying to read stuff way out of my league which was painful to say the least, but i'm sure i got something out of it. I think my listening improved leaps and bounds after i started recording any japanese videos I watched (youtube/anime) and then looped them on my ipod. I found this type of listening practice much more useful than listening to non-watched content or music.

Thanks for your reply. I think I will have to try this looping technique.

But I am wondering how you were able to understand any spoken Japanese with only RTK. Did you manage to find videos with Japanese transcripts or something?

Well at first I couldn't understand much at all. However, before discovering and starting RTK I had already been living in Japan and studying japanese for about 5 months. So I knew hiragana, katakana, some vocabulary (less than 500?) and basic grammar. I passed the old 3kyuu jlpt exam just before starting rtk so that gives you some idea of my level. Ie. very low but not a complete beginner.
After spending maybe a year after that dicking around with various random paper and online grammar/vocab resources and passing 2kyuu, What I eventually realised is that the only thing that was really having an appreciable effect was reading/listening/using real japanese and I started trying to do more of that. My listening ability lagged massively for a very long time and I think there a couple of reasons for this. One is that this is normal for beginners. Another is that Much of my study time was at work at my desk with books (I couldn't really watch anime or youtube at work). Another reason is that I was listening to stuff which provided little to no context (mostly music). Eventually I realised that i tended to understand speech when I knew the words (sounds obvious right) and also when I could understand from context roughly what was being spoken about. Someone (i forget who) once said something like "you can only learn what you already know." Once I realised this I started watching a lot of anime and youtube and I started focusing on vocabulary. You'll have a much better chance of recognising and understanding in context a word that you have already learned somewhere else. I hope this answers your question.

I think the OP and numerous other posters are completely correct in pointing out that you need to know good 10-20k words before you'll start really understand most things in their entirety.

I have been doing some thinking and believe you're onto something.

SRS's only get us to 90%, and they do not build 'fluency' or fast retrieval speeds.

If you read a transcript, then listen to the audio until you reach comprehension, it will take a LOT of listens. A side effect of all that repetition to reach comprehension is that the meaning of each word will be drilled quite deeply. For languages that are especially hard to comprehend (japanese, chinese) this will be even more true. Maybe for something easier to understand, like Italian, more explicit memorization is required.

I still think an SRS is useful, but I am realizing the fastest progress I am making is from making SRS flashcards for Assimil lessons then listening to those lessons until I had 90+% comprehension.

@rich_f: Thanks for introducing "DiG" to me btw I listen to it everyday and I really love it!!!!

@Tori-kun: I picked it up from the podcast thread. I used to listen to it a lot, but started listening to the NHK news instead. It's hard for me to have time to listen to both, because they can eat up a lot of time. But for now, I'm going to go back to Dig. It's just more entertaining. (And I'm tired of hearing about the same 4-5 stories in the news every day, to be honest.)

Betelgeuzah Wrote:Granted, I am quite sure one can speed up the process by using tools like Anki and actually trying to learn new words, instead of letting them accumulate naturally. My knowledge had been built up over a span of maybe 10 years, so there was more than enough time to really internalize all the vocab I came in contact with. To the learners of Japanese here, 10 years would probably be unacceptable. But let's be clear, given enough time even Core6k's worth of vocab "studying" should suffice if one wants to communicate in and comprehend Japanese like I do in English.

I am curious, what do you mean here.

Ok sorry everyone. Looks like I stirred up a bit of a sh*tstorm with my remarks - especially the unanimous thing. So let's take a step back here. Your methods are all fine. What I mean to say is that I THINK this method is good, and I want to show some support for it. I like this vocab-centric way of looking at language acquisition. I think its efficient. But only under 2 conditions:
1. you have the patience for it, and it doesnt feel like pulling teeth
2. in some way, you make an effort to keep some basis in reality with the words you're learning (keeping in mind these are things you need to know how to USE not just some flashcard prompt like its a video game or something)

That's all I want to say. But who am I to listen to? Probably no one. I don't even have any results yet. But meh, talk to me in a year ...

corry Wrote:

Betelgeuzah Wrote:Granted, I am quite sure one can speed up the process by using tools like Anki and actually trying to learn new words, instead of letting them accumulate naturally. My knowledge had been built up over a span of maybe 10 years, so there was more than enough time to really internalize all the vocab I came in contact with. To the learners of Japanese here, 10 years would probably be unacceptable. But let's be clear, given enough time even Core6k's worth of vocab "studying" should suffice if one wants to communicate in and comprehend Japanese like I do in English.

I am curious, what do you mean here.

Learn the words in core6k, then just immerse yourself in Japanese and let the blocks fall in place. There's no need for Anki anymore, or studying vocab. You will learn the patterns, the words that you come across, how they are used and so on. It just takes a longer time.

Why are people talking about % correct stats? You either know it or you don't. No sense getting stressed about it. Next card please!

Put another way, there's no technique you can use to increase your % correct, so talking about it is pointless.

jcdietz03 Wrote:Why are people talking about % correct stats? You either know it or you don't. No sense getting stressed about it. Next card please!

Put another way, there's no technique you can use to increase your % correct, so talking about it is pointless.

This isn't quite true; there are adjustments that can be made to Anki's timing algorithms, and there are different ways of presenting the same terms. Plus obvious things like, doing your Anki at times of day when your recall is at its best, breaking study into manageable blocks so you don't start glazing over cards, etc.

Of course, if your retention is above 80% it may not be worth worrying about. Everything that you can do to improve retention costs time or convenience in other ways. Personally, I put a context sentence in, and both a kana and a kanji card for every new term. That eats up tons of time customizing cards. But my retention rate is much higher now. (It's definitely not worth it for scale learning, btw. I learn far less cards than I could by going with simple vocab cards; I do it that way for depth of learning and because I find simple vocab cards painfully dull.)

Similarly, shortening the scheduling periods or using 'learning mode' or increasing the time for 'learning mode' is going to take extra time reviewing.

It's not that nothing can be done, it's that there's tradeoffs. If you've got a six month old deck and a 50% retention rate, then it's probably better to look at using a different style of card or changing your timing or both. However, it's not worth fussing just because you're off a few points from someone else (especially with young decks where recognition success rate can fluctuate sharply.)

I believe Wozniak said that your learning efficiency is maximized at ~80% retention.

Any lower, and the cost of failing outweighs the higher ease factors.
Any higher, and you are paying for retention with extra reviews.

oefirouz Wrote:I believe Wozniak said that your learning efficiency is maximized at ~80% retention.

Any lower, and the cost of failing outweighs the higher ease factors.
Any higher, and you are paying for retention with extra reviews.

More accurately:

By "retention" I assume you mean "forgetting index", which are different, as generally retention is considered to be your average ability to recall at any given time.

Wozniak Wrote:If you set your forgetting index to 10%, you will remember 90% of the material at repetitions. This does not imply that your knowledge retention will be 90% only. Your average retention will be nearly 95%! This comes from the fact that 90% refers to the retention at repetitions, while the initial retention right after the repetition is theoretically 100%. During the inter-repetition interval, retention is decreasing from 100% to 90%. On average you roughly remember 95% of the material.

For optimal forgetting index: "Simulation experiments have consistently pointed to the value of 25-30%." But he recommends no higher than 20% because generally most facts have some interdependancies and there's also a negative effect wrt esteem/motivation/stress from forgetting too much.

Source

overture2112 Wrote:

Wozniak Wrote:If you set your forgetting index to 10%, you will remember 90% of the material at repetitions. This does not imply that your knowledge retention will be 90% only. Your average retention will be nearly 95%! This comes from the fact that 90% refers to the retention at repetitions, while the initial retention right after the repetition is theoretically 100%. During the inter-repetition interval, retention is decreasing from 100% to 90%. On average you roughly remember 95% of the material.

I don't understand how this works. Lets say I've been studying using Anki for one year. I have 700 cards studied (so 700 cards will at some point come up for review at a future time). I have cards with intervals from 1 day to 2 years. My intervals are more or less evenly spaced from 1 day to 2 years (so, with 700 cards, about 350 cards will have an interval from 1 day to 1 year, and about 350 cards will have an interval from 1 year to 2 years). I have about one card due per day. I have about 20 "young" cards and the rest "mature" cards.

Now lets assume that my recall rate is 90%. If you look at the card due tomorrow, I will have about a 90% chance of getting it right. But if you look at cards farther into the future, I will have a much higher chance of getting them right. For instance, if I (jump ahead and) look at a card that should be due in a week, I will have more than a 90% chance of getting it right, because I'm reviewing it really early. If I look at a card that will be due in a month, I will have a REALLY high chance of getting it right. If I look at a card that should be due in a year, I will have an almost 100% chance of getting it right.

So while my recall rate is only 90%, the amount of cards in my deck that I will be able to recall at any one time should be close to 100%. Lets call this my total recall rate.


Now, my example deck above is very contrived. In real decks, if you have a lot of young cards and cards due soon, then your total recall rate would be much lower than if you had a lot of very mature cards and cards due far in the future. However, it's hard for me to believe that Wozniak could say "On average you roughly remember 95% of the material." Doesn't it really depend on the amount of cards due soon vs. the amount of cards due far into the future?


Edit: Maybe "forgetting index" doesn't mean (100% - recall_rate)?

^ Isn't that exactly what he's saying?

If your average correct rate is 90% at review time, you might actually remember more like 95% at any given time, because some cards are more recently reviewed.

mourei Wrote:^ Isn't that exactly what he's saying?

If your average correct rate is 90% at review time, you might actually remember more like 95% at any given time, because some cards are more recently reviewed.

My point was that the 95% figure seemed really arbitrary. I didn't understand where it came from.

If the majority of your cards are due far in the future, then it would be very likely you will have close to a 100% recall rate over all cards. However, if the majority of your cards are due in the next week, you will probably have close to a 90% recall rate over all cards.

I don't understand where he got that 95% number from.

partner55083777 Wrote:I don't understand how this works....it's hard for me to believe that Wozniak could say "On average you roughly remember 95% of the material."

1) The 95% is from a 10% forgetting index.
For other forgetting index values,

Code:

retention = -(forgetting index)/ln(1-(forgetting index))

2) I'm pretty sure you agree with him but are misunderstanding the wording. The idea is 10% forgetting index implies 90% recall rate at the time cards are due. But right after you review, the time they are farthest from their due time, they should be at 100% (since you literally just answered them less than a moment ago). Your ability to recall will slowly fade from 100% to 100-forgetting index (ie 90% in this case) at the time the card is due. Thus if you average your recall rate over all those periods of time in between, you can express it with the formula above (which is 95% for a 10% FI)

In the day and age of pop-up and electronic dictionaries it seems rather pointless to me to SRS vocabulary. Just keep reading and if you've forgotten a word, just look it up again. It's no biggie.

You gain way more benefit from reading real sentences within the context of real articles or stories than from just flash carding vocabulary, because it also teaches you grammar and gets you thinking in the target language for longer chunks of time. Basically it's more of a linguistic workout.