I was just reading a new thread where a user was asking if they were going to slow. Given that WK is based on SRS and the intervals are meant to be those where you’re “just about to forget”, it follows that people who their reviews as soon as they are available are following the SRS “correctly”. This would also mean that they are leveling at max speed provided that they’re accuracy is high.
Just for information, I level somewhere between every 14-15 days and occasionally decide to spend 4-5 weeks on a level simply to manage my SRS distribution, so this is not a brag disguised as a question.
My question is:
What do you think? Are people leveling at those speeds getting the most benefit out of it? Perhaps there is a correlation between speed and accuracy?
I never advocate going faster than one is comfortable with, but it would be interesting toto know that by suggesting people pace-to-comfort, we might be unwittingly encouraging them to remember less well (lower accuracy).
Having said all that, I acknowledge that everyone is different and there is not reason why the set intervals are the right ones for everyone, but the mental exercise that this entails might be worth it simply to get us to think about these things.
WK doesn’t really have evidence to support the claim that their chosen SRS timeline is optimal (they just changed it this year even), even if that’s the principal behind SRS in general.
Looking at the WK dropout stats I think people should focus more on how they could achieve a sustainable speed and stay motivated (*instead of worrying too much about speed). And you definitely need to read Japanese outside of WK to train the vocabulary and kanji to reliably remember stuff, looking “only” at the SRS and its intervals is misguided in my opinion.
The only benefit of faster leveling people is usually that they can drop out faster
Like you said, different for different people - but there is definitely such a thing as too slow. Even a month per level would mean by the time someone got to level twenty to thirty (which is when most people start to try real reading), they might not have seen some items from the first few levels for well over a year. If they hadn’t been getting a lot of outside exposure alongside WK, they may well have forgotten that earlier stuff when they then attempt to read ‘real’ material.
In general though I agree with @acm2010 - the danger of dropping out is far more real than anything else, so any speed that prevents that is good.
Oh pardon me, I re-read your response and it turns out that I am an idiot who misunderstood you. My apologies. You are indeed correct, sitting on lessons would mean they don’t level as fast.
I think the exact intervals are less important than the number of times you review, the fact that the intervals do grow over time, and the system ensuring that wrong answers reset the level somewhat and thus result in more reviews total. That’s really the benefit of SRS, not the exact timings - which can never hope to be one size fits all anyway. The time of day and proximity to other life obligations could just as easily affect retention as the time elapsed. Maybe I retain better right before I go to bed, while you retain better right after waking up. Seems plausible to me.
So I don’t necessarily think someone leveling at max speed is inherently getting more out of the SRS. Of course I haven’t read academic studies of SRS so take my comments with a grain of salt, as I’m not an expert.
