I should made an app for just typos like missing letters or switched azerty to qwerty or stuff like this. For the rest, I feel like I have to swallow up when a Guru 1 get back to Apprentice 3, it’s part of the process of learning.
I don’t get your reasoning for doing Wanikani. What’s the point of always ignoring your errors? It would be the same if you were reading a novel, came across a word you don’t know and looked it up. Why are you paying Wanikani for the service of reading words? You’re wasting your time right now, just go and read NHK Easy if you constantly ignore your mistakes you’ll get more out of it and it’s free.
19 Likes
As long as we’re admitting abuse of the Override script, I use it when I enter a kun reading instead of an on or vice versa. I even did it just yesterday: “Wait, what? 人 most definitely can be pronounced じん, what the hell? Oh, right, it wanted ひと, and that automatically makes me wrong. Cool. Overridden.”
1 Like
Well, in that case you were wrong since it was asking for the vocabulary word 人, not the kanji. じん is incorrect.
18 Likes
I get the same benefit that you and others get. I use it to learn kanji and vocabulary. I am exposed to many new kanji and new vocab, and they are shown to me at regular intervals.
My accuracy is pretty high without the ignore button. As I’ve said, when I have to use it, I make a special effort for that particular item. I will of course have sessions where I get everything right and don’t use the ignore button at all.
So you’re saying there’s no benefit to WK unless you’re getting things wrong? It’s a structured curriculum for learning the kanji, so of course it has benefit. I’m just ignoring the part where it downgrades items if you get them wrong - instead, I try to ensure I learn them on the spot.
1 Like
But my feeling is that I learn the items pretty much just as well if I use the ignore button.
Unfortunately, how well we feel we are learning is a bad predictor of how well we are actually learning. In fact, effective learning often feels kind of crappy and like you aren’t actually learning the material as well as you’d want.
It’s an easy trap to fall into thinking that we can trust our own judgment about how well we are learning. It’s one students fall into in University every semester, which is why student evaluations of courses fail to correlate with the quality of learning: studies have consistently shown that performance in subsequent courses in a series either fails to correlate with positive student evaluations or is actually inversely correlated with positive evaluations of prior courses in the series, which suggests that students are poor judges of whether they are learning the material well enough to use that material effectively in the future (see for example p. 428 of this this study. or Section 4 on p. 807-809 in this analysis of some empirical research on the subject).
You’ve said that Wanikani isn’t a religion, and that’s certainly true because the Spaced Repetition system it uses is based on scientific research into memory and ‘forgetting curves.’ So, even more specifically, the empirical evidence on learning of this type strongly suggests that the method you are taking is suboptimal for learning those Kanji.
EDIT: One last addendum to address more of your post. First, you claim that:
My accuracy is pretty high without the ignore button.
But if this is true, the the benefit you get in terms of saving time by using the ignore button is marginal. But if the time saving benefit is marginal, then it doesn’t seem like a strong reason to use the Override script to bypass the SRS system. On the other hand, if the time you are saving is substantial, then it simply can’t be true that your accuracy is pretty high even without the ignore button. The only way it saves you a substantial amount of time is if it makes a substantial difference in your accuracy.
Second,
I’m just ignoring the part where it downgrades items if you get them wrong - instead, I try to ensure I learn them on the spot.
Learning simply doesn’t work that way. In order to improve recall, you have to practice recall. Learning the item ‘on the spot’ is not practicing recall. This kind of study has been shown to be ineffective for long-term learning. The book Make it Stick is a good overview of why this is and how recall is best practiced and which methods are poor ways of practicing recall (among other kinds of learning).
You can use Wanikani how you like, of course! And all things considered, if you’d lose motivation to use it if you didn’t use the Override script the way you do, then it’s better than nothing. But when you claim that you are learning items ‘pretty much just as well’ as if you followed the SRS system, the empirical evidence suggests that you are most likely wrong.
37 Likes
I do use it, it’s pretty neat. But hey, most of the time it’s only useful when I type in wrongly. I do not abuse of it with the kanji, I think they’re really important, and I don’t really care much about getting wrong some vocab. If I Ignore wrong vocab answers, it’s because I must have written a very similiar answer but not the one WK wants. So I say, the hell, it’s literally the same thing.
But hey, If you say you are feeling pretty well about remembering them for the long haul, then keep it up. I feel the same, too.
You need to keep in mind that it is okay to get some kanji wrong. After all, you do not need to ace them all in order to level up.
Kind of crazy to me that people would want WK to not count their mistakes, since to me that’s the whole reason for using it. There’s 9 stages (4 apprentice, 2 guru, 1 master, 1 enlightened, 1 burned), so if you ignore all your mistakes you’ll only see each kanji/vocab 9 times. And if that’s enough for you to remember all of them, - I really envy your memory.
Should I have such memory I’d just read vocabulary or something.
19 Likes
My first thoughts were in line with other commentors. It seems to me harmless at first, but a pernicious habit in the long run. However, your comment about how you are using WK just like everyone else just skipping the downgrading part is interesting.
What I am curious about, but not knowledgable enough to weigh in myself, is how this practice changes the way you think about learning. For example, when I get an item wrong (especially if its because of a small sound difference or a typo) I have an emotional response, I get frustrated. I wonder how much this emotional response plays into remembering the kanji the next time. Or if my mind passively mulls over the kanji that I ‘just barely missed.’ Actually, I feel this is even more important for items that you just barely get wrong, which are the ones you are likely to use the override button for.
While I dont necessarily recommend that you do, if you choose to continue this path I would be curious to see if your theory of skipping the downgrades while retaining the knowledge holds up as the levels progress. If you continue, you should make an update post in a few months or so letting us know what you find.
3 Likes
nice, i didnt know there was a wkstats using the v2 api.
1 Like
I know, right? But hey, also do remember that even level 60 people forget even burned items, no matter if they leveled up accordingly. I guess this whole deal of using the script doesn’t really matter if you accompany it by side studies. Like a grammar book and then reading…
1 Like
No, this is not true. You will see it when you first encounter it in your lesson. For me, I often look at the lesson ahead and then don’t do the quiz until a bit later (unless I happen to already know all or most of the items). So you will see it in your lesson once or twice and be asked to recall it in the lesson quiz, and then see each item an additional 9 times (but wrong answers get seen multiple times in the same review). Then I’m getting it reinforced through dealing with regular Japanese materials.
1 Like
WK is a well-designed system for learning, but it is also suboptimal, because we don’t know what optimal is, even if there are studies pointing one way or another. Native Japanese people have also learned their kanji in a suboptimal way, as have all fluent foreigners.
Yes, the time savings might not be that substantial because I’m not getting tons wrong, but still - I prefer this approach of getting to review an incorrect answer and trying again. The vast majority of the times it’s incorrect is due to a true typo or my brain on autopolite on stuff I truly know. But still, I do sometimes get stuff wrong.
Most people who have successfully become literate in Japanese did not do it through WK. I am lucky that I have the great WK curriculum and interface and community to learn Japanese. I am making a slight iteration that I can update at any point. It’s really a small thing - I had no idea people here would think it’s such a big deal.
WK will continue to evolve into a more optimal learning platform and the people who did it in the past haven’t had an inadequate learning experience just because it wasn’t as optimized. They just had a different journey - we all have a unique journey to Japanese. We’re all using different tools and such - it’s not the biggest deal in the world.
We’re all gonna get there, guys! Use the Ignore button or don’t use it - it’s gonna be OK! 
But when you claim that you are learning items ‘pretty much just as well’ as if you followed the SRS system, the empirical evidence suggests that you are most likely wrong.
I do think I’m learning WK just as well as anyone else doing the system. Just a moment ago I finished my reviews and paid extra attention and I was able to 100% it with zero mistakes and zero forgotten readings or definitions. But it makes it less enjoyable to have to be uptight about it. I’d rather know that if I slip up (mostly by accident), it’s no biggie. I won’t see an error showing that my level is degraded because I didn’t choose the right English phrasing that WK wants me to say.
Studies show that we learn better in a relaxed state.. It’s not relaxing to see your level going down.
1 Like
This is actually a great point. We know that we learn better in a relaxed state. I think most people here have probably seen their WK performance might change a bit depending on their mood on ocassion.
I like using the Ignore button because it’s kinda frustrating to see the level go down, especially for some dumb typo, or maybe you know the English meaning quite well but can’t recall WK’s particular choice of English meaning. Or even when something is wrong and you realize you didn’t do a good job of assigning a mnemonic so you assign a new one right there and you consistently get it right thereafter.
Feeling good and relaxed is going to make your studies stick better and make them more enjoyable, which is a big reason why I make ample use of the button.
I don’t have the script installed on the PC but this override button is a feature of the iOS app I’m using to access WaniKani on the go. I really try not to use the button but I sometimes do end up using it if I’m ABSOLUTELY sure that my mistake was due to some typo or fat fingers type of thing.
I try to go really fast during reviews to test whether recognition is becoming “automatic” and sometimes I end up typing a different letter than what I intended. Anything along the lines of “Oh yea the rendaku I forgot about that” or 「にんせい」instead of「じんせい」does not get an override from me. For instance, I’ve had the word 「明日」sitting in my apprentice queue forever cuz I always confuse it for 「朝日」for some reason. The first Japanese word I ever learnt just can’t seem to click for me yet.
You remind me of myself in French class. I’d write a word and forget where to put the accent or put the opposite one or screw up some other way. I’d say to myself yea dude that’s easy I’ll definitely remember that in the exam and so I just brush over them. Come exam time, I was a few marks away from my A 
.While there’s no exam in self-studying Japanese, it’s a real handicap to brush over the little things and that’s what abuse of the override button is.
6 Likes
Studies show that we learn better in a relaxed state.. It’s not relaxing to see your level going down.
First of all, the study you are linking to was of eight people and it measures theta waves and not relaxation. Can you point to more robust studies that show a substantial impact on recall made by being in a relaxed state?
Second of all, there is the question as to the relative benefits of relaxation on learning compared to the used of a spaced repetition system. If the benefits of the latter are more robust, then the benefits of relaxation don’t outweigh the benefits of following a spaced repetition system. It would be a lot more relaxing for my students if I gave them easy tests (or didn’t test them at all) shortly after learning material. But it turns out that they would not learn the material as well as if I gave them more stressful tests that are challenging and occur longer after the initial exposure. Presumably this is because whatever positive benefits relaxation has for learning, those don’t outweigh the positive benefits that using long intervals to force more difficult retrieval has (even though doing so is more stressful). Indeed, the improvement in the above study was the difference between ~60% correct recall and ~40% correct recall.
In fact, in general, difficulty has been shown to enhance learning (this article lists quite a few studies and it’s also the thrust of the book I mentioned, Make it Stick). But difficulty is often experienced as stressful, so that suggests that the benefits of having difficult but stressful learning outweigh the benefits of easier but less stressful learning. Naturally, if the stress is not coming from difficulty that is related to the learning (like if you’re stressed because of outside of other factors), it will be better to be relaxed. But it would need to be shown that relaxed people learn better even when the stress in the comparison is coming from difficulties that are known to enhance learning.
Lastly, there’s another solution: learn not to get stressed out by mistakes. Then you can follow the spaced repetition system and be relaxed, reaping the benefits of both. The fact is, allowing yourself to make mistakes enhances learning, so it’s probably good to become comfortable with doing so.
19 Likes
The person who can read better, is better off. In this story, it’s mostly likely the 700 day person.
There’s no shortcuts in life.
I’m super super strict myself. I have two conditions I use for the override, and the above one is not good in my mind. If they ask vocab I need to remember the vocab version. I’ve done that exact same mistake.
My use for the override:
1: Clear typo, I meant to write correctly but I misspelled. I use this maybe once a day.
2: I know the meaning properly, but the one I wrote was same intention but just not listed. So far I’ve used this twice I think. Yesterday was one: 度 I know very well, and I wrote “times” when it needed degrees or occurrences. I know many words where they use that so I settled to myself that times = occurrences.
11 Likes
I’d say cheating on Apprentice items isn’t too bad, except if completly blanking on something, that makes for less reviews, guru 1 maybe, depends on the mistake, higher is a fine line, that shouldn’t be crossed often, can’t remember something ? well 残念, and a burn should not be overriden.
Typos should always be corrected, except when you’re dumb and press buttons too fast, and validate a typo, DON’T ASK ME WHY I’M BEING SPECIFIC
Also, investigate, investigation, research, examine, examination, exam
5 Likes
The person who can read better is most likely the person who spends more time reading content intended for natives.
Exactly, which is why I make sure to spend time reading Japanese daily, in addition to keeping up with my WK reviews and lessons.
1 Like