Learning with Spaced Repetition

Hey there, Doubters! It’s been a while, I know. Where have I been, you ask? Just languishing in post-doctoral unemployment and general malaise, enjoying my time off, doing some travelling, and doing unpaid labour to prove my dedication to a perverse system happily working on some research and post-doc applications.

But enough about me. Today I’d like to talk about something completely different; something that often gets ignored or maligned in our current age of google, smart phones, and the constant deluge of information they provide us; something that, pedagogically speaking, is just about as un-sexy as you can get: rote memorization.

Memorization has gotten a bad rap lately. It’s often held up as the example of what we shouldn’t be doing in the classroom. Process over formula! History is more than just dates! Facts can always be looked up, it’s how we treat them that matters!

But the truth is, more often than not there is simply no substitute for actually knowing things, and we can’t always absorb everything we need to know by osmosis. Sometimes you just have to sit down and, y’know, memorize that shit.

This is especially true in language acquisition, at least once you’re past childhood. Since I have a lot of free time lately (ah-hem), I’ve been working a fair bit on my Japanese. I’ve been focusing much of my effort on learning the kanji, which is why memorization is on my mind. Kanji have a bit of a reputation among students of Japanese, since they require a whole lot of rote memorization: there are currently 2,136 characters officially in common usage, most of which can be read in at least two ways depending on their context. If that sounds daunting, it is. It’s also crucial barrier for learners (and particularly self-learners) to overcome, because the inability to read at least the 1,006 grade-school level kanji leaves one functionally illiterate and totally unable to engage with any written media intended for people over the age of 7.

I learned about 200 kanji during my time taking Japanese classes, but most of my efforts at self-study in the time since were less than successful (read: not at all successful). Then I started using a wonderful online tool called WaniKani, which operates primarily through the use of a Spaced Repetition System that is slightly gamified in order to keep students motivated. Spaced repetition (for those who didn’t click the link) essentially works by increasing the time intervals between each recall of a given bit of information, on the principle that these items are more successfully transferred from short- to long-term memory if they are recalled just at the point that they might otherwise be forgotten. A correct answer will move the item up a level (and increase the interval), while an incorrect answer will drop the item down into a more frequent rotation.

WK adds a level system to the SRS, thereby giving users a sense of accomplishment and progress as they memorize more and more kanji (each level has around 30-40 characters, plus radicals and lots of vocabulary items that employ the characters’ various readings). It also provides mnemonics to help with initial memorization, but I must admit that I personally very rarely find mnemonics helpful as learning tools since I often have trouble remembering the mnemonics (is that meta-forgetfulness?).

I’ve been using the system for a little over three months now (with a longish break), which has put me at level 6 out of 50. I must say that it has impressed me on two counts.

The first (and perhaps the most impressive) is simply that I am still using it after all this time. I have long been skeptical of self-study in language acquisition, because I’ve found that without the structure of a class or a teacher providing external motivation to study I tend to just let things slide for weeks and/or months at a time. For some reason, though, the mere desire to progress through the levels has kept me doing my lessons and reviews every day. It helps considerably that I can do them on my phone from anywhere in the world (at least with an internet connection), but the same possibility with other study resources had not produced such sustained interest.

I am also impressed with the program’s actual effectiveness. Since I was already an intermediate-level student, it took me quite a while to start getting to kanji and vocabulary that were new to me (and this was, admittedly, a bit discouraging). Now that I have mostly passed that phase, however, I am continually impressed by how well I am remembering the new characters, and how much easier it is compared to the traditional method we employed in my Japanese classes. The rate of acquisition is even slightly faster (30-40 kanji every 9-12 days with WK, vs. 12-16 or so in every chapter of Genki), but the spaced repetition ensures that characters don’t fall through the cracks after you learn them.

As valuable as I think WaniKani and spaced repetition are, however, it’s definitely true that the method is not a cure-all for difficulties in self-study nor it it sufficient on its own to arrive at any level of true fluency (for which I am still convinced that, after a certain point, immersion is required).

There is of course also a difference between recognizing something under controlled conditions and mastering its use in the wild, though this is a weakness shared by all flashcard-style learning systems. WaniKani also deliberately ignores stroke order, which might bother some but does seem reasonable in an age where few of us write by hand in any but the most informal circumstances.

In all, though, I am extremely impressed by the effectiveness of SRS as a learning tool, especially when it is paired with gamified incentives. Gamified SRS systems would not only make for great textbook supplements to help students recall key facts and vocabulary, but I think would also have great potential in other arenas.

For example, I often have trouble remembering exact sources and their authors (a trait that was decidedly unhelpful on my comprehensive exams). An SRS system with authors, titles of articles, and a brief summary of their main arguments would have been invaluable to me. One could even maintain such a system for an entire academic career! Armed with a tool like that, I might even have a chance of one day equalling the encyclopedic knowledge of my thesis advisor! Well, one can dream, anyway. 頑張って everyone!