I’m trying to understand why the default behavior is to separate the meaning review and reading review from one another. I vaguely get this might be so you can independently recall whichever one depending on what you currently need… but I mean I always think of both when I see a kanji or vocab item anyhow. I don’t go past the one until I’ve tried to recall the other and I don’t see any problems when applying what I’ve learned.
So what’s the point in not making them follow each other directly? Less re-reviewing items that you already half-answered in the last session.
(Yes, I know you can use a script to make it behave that way, I’m asking about the default behavior here.)
I think the main point is the idea that recalling information that belongs to a certain entry will help you remember that entry better in the long term. If you have the reading and meaning question back to back, you only really think about the item one time, where you recall both its meaning and reading and answer the questions. If they are separated, you have to recall this information twice which theoretically helps you remember it better long term and learn faster. Of course, it also makes your reviews take longer which makes you learn slower. I have no idea if one of these effects is significantly stronger than the other. I would say you should do whatever frustrates you less, because frustration is much worse for your learning than any possible positive effect from a specific lesson order is beneficial.
One thing that @gijsc1 didn’t mention in their post, is that there’s this phenomenon, where if you learn similar or related terms about the same time, you might not remember some of them properly, instead having to use the rest as a crutch. A common example is that if you’ve learned colours together, there’s a chance, that if you need to remember green, you’ll just go “ah, what was that one I learned after red?”. The way wk is asking these questions theoretically helps with this, while the existing limit of at most having 10 partially done items helps keep this in check, so you don’t ever have to lose a ton of progress if you have to stop halfway.
This doesn’t exist anymore. If you partially do a review, it will disappear the moment you refresh the page.
Now that I detailed how great this non-back-to-back approach is… I very much use back to back scripts. It’s just way easier for me, and I want to be able to do as many reviews in a short time as possible.
I think the initial reason comes from studies that say (paraphrasing) “interleaving improves memorization”. Using that as a reason for separating reading and meaning is probably an over-generalized interpretation. About half of the neurons in your brain are related to sequence memorization, so it’s like having half your brain tied behind your back if you don’t do them back-to-back (preferably in a consistent order, either reading-then-meaning, or meaning-then-reading).
What @Gorbit99 said is true for things that aren’t inherently related (like, you won’t always see ‘red’ and ‘green’ together). But reading and meaning are inherently related. They are two aspects of the same thing (the kanji or vocab), so it makes sense that you’d get a valuable boost to memorization by always doing them back-to-back.
However, here on WK there has been a lot of anecdotal evidence that doing reading and meaning back-to-back has a potential down side for some (but not all) people: By doing them back-to-back, you’re cutting the number of times you see an item in half, and some people found that their memorization suffered. But those same people are still able to gain the benefits of back-to-back by simply recalling both the reading and meaning no matter which one WK is asking for. (In other words, they still get two separate reading and meaning questions, but they say both in their head each time, and only answer the one that WK is asking for.)
So, if you don’t suffer from the negative effect on memorization, doing back-to-back is a fantastic way to save time (which can be used for other things, like grammar and reading). If you do suffer the negative effect on memorization, just do the “recite both” trick. It will take you just as long, but your memorization will improve significantly.
Well, that’s the thing. I don’t think this necessarily happens. There are cases when you answer the first half of an item with confidence, but have trouble with the second. Doing the extra mental exercise while reviewing of “recite both” is something that likely will take you more time during reviewing. It’s a good idea, but I also don’t think it happens all by itself or that you necessarily take time to do it fully unless you consciously do so.
I think it’s the same while reading and encountering vocab unknown to me: you might be able to guestimate the reading here, but still be unsure about the exact meaning. Or the opposite might be true, you might have a good idea about the meaning, while being unsure about the reading. This can happen during reviewing on WK as well.
Just some thoughts. I don’t have a strong opinion which is best. But, I don’t mind the default option of WK.
That’s what I meant: Consciously forming the habit of reciting both reading and meaning, then looking to see which one WK is asking for, then answering with the right answer.
Initially, yes… but I think recall often improves enough to offset any additional time spent. As soon as I started doing back-to-back, the meaning and reading both started popping into my mind simultaneously, so I spent a lot less time trying to think of reading and meaning separately. I’m not sure if that’s what others experienced, though. (sorry, I’m falling asleep while writing this… hopefully I discover that it still makes sense in the morning!)
My opinion on the matter is, listening, reading and speaking / typing / writing, need different degrees of association; but not to disagree that to recognize Kana first is helpful.
For less common words, recalling the sound first would be helpful, but not a requirement. Still, this much is helpful for cementing Kanji readings. Still, there are plenty of chance of encountering Kanji-based words not in Wanikani. Either Kana or Kanji would help guess the meaning depends. For listening, it’s Kana plus some pronunciation quirks. For reading, it’s usually Kanji, but also sometimes Kana in Furigana. For speaking / typing, both Kana and meaning must be cemented to make it effective, but probably Kana first, anyway.
But for more common ones, it remains better to have recall reading first.