Regarding repetition and SRS, I feel that SRS is better for ensuring you don’t forget. As for repetition, I actually remembered Kana (as well as stroke directions), by massive repetition (and it took some time).
For initial remembering, I don’t really know the best way, but shorter-than-WaniKani-SRS may work as well. Some exposure to audio (subbed animes, songs) also made me remember without any real drilling.
Hearing words in practical use is a form of Spaced Repetition, without the System. SRS works, sure, but that doesn’t mean that spaced repetition in practical use doesn’t work. (Can I invent the term SRPU? ) And as always, what works best varies from person to person.
I tried going fast and had little long term memory for most of those words. So I slowed down.
Now I use the self test addon to review all my apprentice words. All the time. So many times. Some people say that ruins the 4hr / 8hr etc memory… but as other say, you should see these words out in other sources; real life, books, movie subs… Well, I see them in my 5 times a day self review! It seems to help me get words to stick longer.
These people are reading too much into the “quizzing you just before you forget” aspect of an SRS. I’m told the research is pretty clear that dredging something up from memory just prior to forgetting does indeed improve the connections and ability to recall, but my experience has been that this is far more important with long intervals (days/weeks) than short intervals (hours).
I’ve argued (often) that the most important aspect of an SRS is to give you more frequent reviews of the stuff you find hard, and less frequent reviews of the stuff you find easy (via longer intervals). In the case of Wanikani, the only way it knows what you find hard is if you answer incorrectly!
If 40% of the items are being answered incorrectly, too much of the workload is deemed “hard” and it will quickly start to feel like you aren’t getting anywhere.
I’m 100% confident that anyone can memorize anything with enough repetition. An SRS makes that repetition more efficient, but nothing is perfect and nobody can state with certainty what the ideal timing between intervals might be.
@Vouru says he or she has one session in the morning and one at night. That means at most two repetitions for anything, even stuff in the first two stages.
My strong, strong recommendation is to:
Stop doing lessons until your accuracy improves.
Aim to keep no more than ~100 items in the Apprentice bucket for a while. Don’t start doing lessons again until there are fewer Apprentice items. (I’m surprised, but it appears nobody has mentioned this.)
Before each review session, either:
a. use the new “Extra Study” feature to get in several extra reviews for “new lessons” and optionally, “recent mistakes”.
b. Use the self-study quiz userscript to review just items in stages 1 and 2. [This is what I do. I launch self-study from my Ganbarometer userscript, FWIW. I prefer the self-study quiz, because it allows me to re-quiz just the ones I got wrong.]
Either way, repeat the “extra” reviews as often as necessary, two or three times in a row, until you feel extremely confident with this limited subset of review items. Only then start your “real” reviews. Even if this takes so much time that you can’t finish all your outstanding reviews every day, I’m certain you’ll find the process much less stressful and that your accuracy will improve dramatically — decreasing your workload, improving your mood, and generally helping all around.
As long as you stop doing lessons until your accuracy improves, I’m confident this process will eventually get your current workload under control.
Thank you everyone for taking the time to read my post and give such useful and heartfelt replies!
I am defiantly going to take to heart and try the strategies mentioned here with my first step being taking a break from WK specifically for a bit and then coming back at a later time to doing offline revies via self study or equivalent till my accuracy improves.
While on break I’m going to focus on grammar and light learning (using apps like kawaiinihongo) as I’m finding grammar studies to be pretty fun and something different then trying to remember the meaning and spelling behind each kanji alone.
Have you tried making your own Japanese mnemonics? Might be a stupid idea but would get rid of the English pronunciation → Japanese issue
Like for 土 you might think “The moon (月→つ) shines down on the blood (血→ち) on the ground”) or for 角 (random cause I saw that one eariler) you might think “persimmons (柿→か) fall down at an angle and suffer (苦→く) when they fall (at which they also land at an angle)”
Can’t comment on anything else since I’m not generally any better at keeping up with like Anki reviews
This is how I study too. I write the kanji down and occasionally draw a picture. Sometimes it takes me a couple time of getting the word/kanji wrong and me having to “re-do” the mnemonic in my head for it to finally stick. But when I finally do find a mnemonic that works, it sticks like glue.
Note: What follows is advice from a stranger on the Internet. If any of it speaks to you, use it to empower yourself. Disregard whatever sounds like bovine excrement, as it’s no-more-useful-than the same.
I don’t think you are “doing something wrong” but rather your current frame (i.e.: what you believe about the Japanese language, what you believe about your progress, what you believe about your capabilities, etc.) are causing you serious emotional distress and anxiety.
Upper logical brain, you’re going through the physical motions of reviews. Lower emotional brain, is “readily looking for some distraction” to do something (anything) else because of the emotional distress.
I’d like to offer some study technique that would instantly turn things around, but the truth is probably that you need to examine your motivations.
Ask these questions of yourself:
Why is learning Japanese important to me?
What do I want to do with the Japanese language when I’ve mastered it?
How will I feel when I am fluent in Japanese and using it for what I’ve intended?
Write these down, and review them each time before you start your WaniKani reviews.
DON’T do 100 reviews a day. Rather, guess a number that you can comfortably do, maybe that’s 50? or 30? or 20? then, do 80% of that. If you think “Sure, I can do 50 reviews.” then proceed to do 40 reviews (50 * 0.8 = 40) and STOP.
The point is to STOP before the negative emotions set in. At the end of your session, you are aiming to feel like, “Well, I could do ten more…” you should feel a little hungry to do more. But, you’re going to stop.
You’re going to think about how you’ll be able to do more tomorrow, and you can hardly wait for tomorrow to hurry up and arrive so you can actually do them!
One thing we’ve learned from Artificial Intelligence / Machine Learning is that error rates that are too low (i.e.: scores that are too high) are an indicator of overfitting or other problems in the model.
As this translates to the real world, if you are scoring 100% on your WaniKani reviews, chances are you aren’t being challenged enough.
Conversely, if you’re worried about a low score it means you’re trying to cram too much into your brain. In your case, this has nothing to do with your personal intellectual capability, but rather is because you are dragging yourself over broken glass emotionally to force yourself to do reviews.
Upshot: Go Okinawa 80% review style, STOP before it hurts. Let yourself look forward to doing reviews the next day.
I think once you do this, you’ll see shocking improvement in both how you feel about doing your reviews, and how well you are doing on them.
I’m late to the party, but good luck with your studies. I personally think that @Rrwrex 's advice is the best for you, just take it slowly and focus on strengthening the things you’ve learned before studying new stuff. I also believe that with enough repetition, people can remember almost everything so just keep at it! Also, if you have priorities outside of studying JP that might occupy your time and mind, I’d recommend you get that settled first. All in all, just go on slowly getting your accuracy up to like 80 percent and go and add lessons slowly. You have a lifetime sub so you have all the time in the world to improve.
I know this topic is specifically for WaniKani, but as much of the advice given so far can be applied to Anki as well, as someone who is currently struggling with obscene review loads, I recommend going through the thread to find a more balanced review diet
It’s honestly a bit of a shame that WaniKani doesn’t allow setting lesson and review caps like Anki does, because it’s way too easy for people to shoot themselves in the foot and be forced to reset levels or enable vacation mode to put a stop to the flood of reviews .
This is the part that worries me the most: do you only have time for Wanikani and are not doing any grammar at all? If it’s Wanikani you’re having trouble with, I’m sure we can figure out what might help you, as everyone else in this thread is already trying. But if you don’t have the time to study the language, then Wanikani isn’t going to do anything for you. You’re going to be reading Japanese like a Chinese person would attempt to. And often, your grammar studies would actually help make Wanikani a lot easier because you get more exposure to the language, which improves your memory and intuition.
Track 2 of Dragon Ball Z 18 1/2 Special Super Remix. It’s 4 remixed songs arranged onto 1 track, including the iconic Gohan vs Cell music that was originally going to be the series swan song theme.
When this track enters through your ear canal into the brain, through a complex and magical process, your brain briefly goes Super Saiyajin and becomes capable of tackling any task it is given. This is true of anyone in the world and totally not just of me anecdotally. This song is like if Betty White descended from heaven and jammed a 5G radio tower with healing rose quartz crystals duct taped to it directly into your brain stem.
Your first post expressed frustration about feeling overwhelmed.
The temptation is to learn grammar, build vocabulary, improve listening skills, practice production/output (speaking and writing), learn about pitch accent, and learn how to read all at the same time.
Speaking as someone who spent literally decades trying to learn the language this way, I can state definitively that there us a much more efficient and less stressful way: focus exclusively on that last item first, then start on the rest once you’ve got a foundation built.
The Japanese language is tightly bound to the written language. Much begins to make more sense once you can read a bit.
If you’ve already got at least the very basics of Japanese grammar down (enough to form simple “this is a red pencil” and “I went to the store” kinds of sentences) then my advice remains to focus exclusively on reading with WK for a while.
I‘d seriously consider leaving bunpro, etc. alone for a while, at least until WK starts to feel under control.
When is best to start branching out depends on the individual, but doing it when you’re feeling overwhelmed seems unwise.
By all means, set WK aside for a while if you’re starting to dread it, but there are real advantages to making reading the first step of your journey.
Honestly, I don’t see the appeal of Bunpro. I did the month free trial and quit it after a couple of days. What you mention, practicing with simple sentence patterns, is probably the most efficient way to learn IMO. “Monkey see, monkey do.” Using words results in acquiring grammar. Studying grammar doesn’t result in using words.
This x1000! My poor parents have been studiously and religiously using DuoLingo for Spanish for 3 years and can’t string together more than simple sentences about finding a toilet. Duo brings you up to a beginner toddler’s level and then tries to guilt trip you into practicing everyday and leaving you wondering why you’re not getting more proficient.
I felt bad and got them a bilingual reader and some verb flashcards to try to help them along. Of course they almost immediately shelved them and went back to their gamified illusion of learning.