I love SRS (Spaced Repetition System) applications. Primarily I use this site and Twinkle.

Recently I took about 3 weeks off from reviewing kanji on this site. (see Chilling out)

Now that I'm back to reviewing I don't think I'm doing much worse (if at all) than before. Which is leading me towards thinking that as long as keyword->story relationships as well as the story itself are strong then you don't really need to review much at all. Admittedly I have been using this site for about a year so I do have a fair few reviews under my belt already which may have something to do with it.

However there are some cards that I've got to the fourth stack through repetition and "just remembering" but in the long run I want to steer away from this.

Now I'm spending more time during review clarifying my images or changing stories so that I'm working with more concrete images. Personally I find mnemonics can be really effective, e.g. mattoc_g's "Yo mamma so obese, her flesh affects compasses!" for obese.

Otherwise I think that my stories need to be visual i.e. actually see an image in my head as opposed to simply having a story.

My point being that once I get a decent image where seeing the keyword leads me towards that image then reviewing becomes a lot less necessary.

Contrasting SRS Heisig with using an SRS to drill grammar points I find that my success on grammar nuggets falls more when I skip reviews.

I think astridtops was the first to coin the phrase "working dirty". This worked pretty well for me, allowing me to seperate the cards where my images were working with those where I had to work a bit harder on the story/image. So I don't think it's necessary to spend a long time on images, just letting them come seemed to work pretty well a lot of the time.

Anyway, how advantageous do you think drilling from keyword to kanji actually is for long term retention?

synewave Wrote:Personally I find mnemonics can be really effective.

Really? That's a pretty interesting discovery you got there: perhaps you should write a book about it.

On a more serious note, the stories are simply a framework for learning and analytically viewing kanji. Over time, the stories are supposed to fade away as they're replaced with a direct link between bunch of squiggles and some concept (given your considerable experience with the language, I would imagine that this would be, hopefully, somewhat rampant): drilling, no doubt, helps facilitate this process. Oh, yeah, and it helps root out weak stories.

But, yeah, I'll totally agree that there are quite a few cards in my last box that got there via rote, which totally defeats the purpose of mnemonics.

Depends what your definition of a mnemonic is I suppose!

Megaqwerty Wrote:

synewave Wrote:Personally I find mnemonics can be really effective.

Really? That's a pretty interesting discovery you got there: perhaps you should write a book about it.

What I was thinking about when I wrote that was simple verbal mnemonics where I don't actually see any particular visual image in me 'ead. As opposed to the majority of my stories where I do make use of images that I "see".

What I got from Heisig's intro was that he was championing the idea of mental images as opposed to using 2042 short verbal mnemonics.

But yeah...maybe I should write that book