@fuzzytipsy made me wonder something that I thought could be worth a discussion in itself, so here I am.
The thing with SRS is that it continually challenges your long-term memory by spacing out repetitions of an item. On the other hand, the more often you see an item, the more it is engraved in your memory, that’s how immersion works.
If you use WK to learn vocab/kanji and, in our example, KameSame at the same time (set to make you review every guru+ item you have on WK), your long-term memory will be less challenged because KameSame will make you review the items WK would have you ignore for a while. KameSame has other advantages (EN-JP) but for the sake of the argument we could also imagine WK + an anki deck with WK items in the same order.
The question is : is it better to bombard your brain with twice the reviews (JP-EN and EN-JP, or why not even twice JP-EN), or does it undermine the long-term memory capacity WK is trying to build ? (in which case reviewing only burned WK-items would be preferable)
I’m guessing this has to do with how memory works and since I’m no expert, I wanted to have everyone’s opinion
In my opinion, more practice could never hurt. I do understand what you’re saying, though, and I agree to a certain extent. I believe that using multiple SRS systems at once would only become a problem if they, in some way, permitted you to take some sort of “easy route” to getting all of the items to a ‘learned’ status without actually learning them. I personally only use WaniKani, though, and I think some more input from people who use multiple SRS systems regularly would be beneficial to the conversation.
That’s why I set Kamesame to review words at the same time that I learn them in WK, so the intervals are usually about the same. However, I don’t think there’s a big issue with reviewing words more often. Also, if you’re doing a good deal of reading, you’ll be coming across a lot of words there anyway.
I think for max realism, you should see the word about as often as you’re going to see it in real life. Trying to avoid seeing it for four months isn’t necessarily a good test of reality. I don’t know, maybe that could be the case for some of the words, but I hope not the majority.
No idea but it is interesting, I think a major point is not just memory but recall. A lot of us are able to remember a large number of items, but trying to recall it when you need it is hard. I don’t know how many times I forgot something on wanikani and clicked the button to go “Oh yeah of course, I knew that”.
I wonder whether the spacing helps train recall as well as memory (so more reviews for things you know less and vice versa). Though things like immersion may increase the chance of burning the memory some other way (always remembering a certain kanji because you saw it somewhere and now its second nature).
I just want to echo what everyone else is saying. I remember from somewhere that it is WK’s recommendation to start reading as soon as possible to put the stuff that you know to use; seeing that stuff outside of WK is in fact the recommended approach. Of course, seeing them in another SRS might not be as “natural” as reading and not what was originally intended, but I still think that the principle holds.
As an aside, I am greatly envious of people who are able to juggle other SRS tools with WK. Doing 250 reviews every day is hard enough for me as it is
hard to answer this question as I’m no expert, but i’ll chime in with my personal experience.
While it’s probably not true to say it always has a bad effect, there are some times it can.
For example, while studying the core 6k deck alongside wanikani, I’ve had times where I’d incorrectly answer a mature (roughly equivalent to enlightened) card on anki, in others words I’d forgotten it. Then it might come up as a review on wanikani in the next few days as enlightened, and I would burn it. Had it not been refreshed a few days earlier through anki, I most certainly would have got it wrong. Obviously it’s different if you’re seeing the vocab in the wild and cementing it through exposure, but if you’ve just seen it through another srs system, it seems to me that the overlap between each srs could negate the intended time intervals – possibly levelling up vocab you otherwise wouldnt.
There have been times where the same thing has happened to me, but it was in conversation/reading out in the real world. I never would have remembered those certain words just in WK, but by encountering them irl it cemented them even more than an extra WK review would have
So although irl usage is probably the best, I think supplemental SRS has a similar effect
SRS challenges long term memories and strengthens them, which is awesome, but imo it’s still best viewed as a tool for retaining the information until it’s viable that you would encounter it in the wild.
That doesn’t mean don’t go out into the wild.
You’re trying to learn a language; you’re not trying to learn how to memorize things.
Do you think the act of remembering that you forgot it will cause you to remember it for sure now? I.e. Did the experience burn the word for you ? That could be a bonus of multiple SRS exposure. I personally don’t have time for more than one SRS what with all the other learning and using the language that needs to be done.
I’m not memory expert either, but I do know that within the context of WK itself the SRS isn’t going to be exactly avoiding exposure to kanji between intervals anyway: The second you guru a kanji 1-5 vocab lessons including that kanji appear so you’re constantly exposed to a guru’d kanji you wouldn’t be seeing for a week in a perfect SRS system. Yet I feel the vocab reinforces it more than anything, so I wouldn’t change that at all.
In long term: I have had leeches that have lingered for long enough to cause me to burn a related kanji/vocab. and I can never be 100% sure I would have burned the kanji/vocab if I hadn’t been constantly exposed to the related leech.
I love the concept of SRS being geared for efficiency (minimal exposure, maximum memory) but practically, it’s impossible to keep perfect intervals between exposure unless you hide yourself from the language you’re trying to learn entirely…which is of course impractical. imo an SRS system is a healthy starting platform; a little life jacket to keep learners from drowning in the ocean of language.
The point of the SRS spacing is to minimise the amount of time you spend doing reviews, not to make you remember things better!
More exposure definitely won’t hurt, but you get less benefit for the amount of time you spend (more exposure outside the context of SRS can be very helpful though).
Besides, the algorithms should work anyway (what happens for me, is that reviews on one platform space out, but stay frequent on the other until I’ve actually memorised it).
Disclaimer: I’m not in anyway an expert and am going off a couple of papers on memory acquisition and retention I read a while back.
I am using 3 SRS in parallel, but only because they have different purposes:
WK for recognition and reading capacity
Anki desk for active vocabulary (unrelated to WK)
Anki desk for kanji writing
And that’s already a limit of time I can spend on SRS daily. Whatever I do else is just reading or analysing/practicing grammar topics, or listening - i.e. actually trying to use the language.
I do not think using several SRS spoils the system - they just sort of rob you of your real language time. And they should go away as soon as you can progress without them.
Thanks for all the very interesting answers. I find particularly relevant the point according to which several SRS are always better but subject to diminishing returns since it steals time that could be invested in another aspect of the language.
