I’ve ended up in roughly the same position as you. I started out really liking flashcards, and I still think it’s an extremely good way to build basic vocabulary quickly, especially for reading. Then I didn’t use flashcards at all for a while and just read and listened a lot, but then I picked them up again. The reason is of course that there are certain areas where flashcards do really well, especially for things you need, but might not need that often (specialised vocabulary and handwriting, to name a couple of examples).

The problem with rare vocabulary is of course that the amount of reading and listening one needs to do to maintain knowledge of them organically simply is too big for people who don’t have that much time to learn and maintain their Chinese (this includes me, too). Then it’s a matter of finding a flashcard setup/solution that works for one’s particular circumstances.

It’s never too late to pick it back up again! 🙂 And yes, it’s so much easier to find reading and listening material these days. Low-intensity listening and reading (but especially listening) is not very demanding, so if you can find some materials you enjoy, listening to it while doing specific other activities, that might be a good idea (maybe when you’re out for a walk, commute, run, knit or whatever you fancy)!

I agree that this is an interesting area and probably not something that has been studied very closely. I think the best way to approach it is to be as aggressive as possible with settings and/or answer grading if available. In Skritter, for example, always picking “easy” if you do know something even though you’ve been taking a break. I Anki, there are other ways to be more aggressive (if you find the source you mentioned, that would be interesting).

Personally, I think the SRS apps I’ve used work reasonably well after coming back from breaks (this would be Anki and Skritter). Obviously, the initial retention rate drops quite a bit, but I don’t have the feeling that subsequent scheduling has been messed up, at least not enough to cause noticeable problems.

Then again, this is only what I’ve experienced, which highlights the issue here: Initial learning is rather similar compared to coming back after a break. It doesn’t seem unreasonable that retention when learning something completely new is more predictable than coming back after a break. There are so many different situations the latter could occur and each might be different.

This might be an idea for an article for you, Scott! You’ve written so much good stuff about learning, but what about relearning? That would be interesting to read more about!

I don’t remember where exactly I saw this right now (YouTube perhaps), but I found someone who had their Anki set up to address this.

Say you forget a 30 day card, it then pops it back to the “new” category for a couple days. Get both of those right and it would go back up to a 20 day interval instead of resetting entirely.

Your comments on flashcards are something I’ve thought a lot about myself. My Chinese deck grew enormous during my initial phase (maybe 16k old cards?), before I put it off to spend more time with actual reading/listening materials.

Thinking back now, I wouldn’t mind getting back into the flashcards to brush up. Especially given I’ve noticed some weaknesses in some foundations. Yet, going back to the old decks is often a tricky thing with SRS. The options built-in tend to consider forgetting an all-or-nothing event, so there isn’t a good system for forgetting due to longer-than-optimal spacing.

I suspect there’s probably a better algorithm for dealing with missed cards than the one Anki actually employs, but I haven’t seen it discussed so much. It seems clear to me that a failed card that was one you got correct a bunch of times in the past is simply a different kind of memory episode than a failed card which you’ve never encountered before. Not to mention the old queue of cards where failure means going back to the beginning again is hardly a great solution motivationally.

-Scott