Comment by s_m_t | Hacker News Reader

Comment by s_m_t | Hacker News Reader

After using what was essentially this technique to learn to read Japanese I'm not entirely convinced it is actually all that useful compared to just reading. I found myself spending so much time in anki that sometimes I wouldn't do real reading at all... And for what? So I would have a chance at being sort of familiar with a character when I came across it? So called "flashcard blindness" is also a real thing. Context is so important when learning, sometimes it was like I had to learn two words, the word I 'learned' in anki and the word in the wild. Even though they were obstinsably the same word somehow they couldn't connect.

If this technique did help at all I think the effect was mainly psychological. Looking at a page with a bunch of words you don't know is stressful and demoralizing. Even being able to think that you have seen a character somewhere without knowing what it means can be comforting. It sounds silly but any amount of stress can severely inhibit language acquisition. You might be better off meditating for 15 minutes... Or taking drugs

The only thing that seemed to be 100% effective was readimg content that I was completely engaged in. It didn't matter if said content was close to my skill level or way beyond my skill level. As long as I am sufficiently engaged it doesn't matter how many times I have to reference the dictionary or comb google for grammar explanations.

I had never heard the term "flashcard blindness." But, wow -- it fits my experience.

I think any time I mistakenly use SRS expecting that it represents the learning process rather than it facilitates learning, I'm prone to this syndrome. More explicitly, the learning process involves active use and problem solving, creating a web of connections between concepts. SRS merely changes conceptual recall times. For a CS analogy, it moves the memorized concept to main memory instead of residing on a slow disk or over the network. Cumulative slow seeks frustrates learning, so this is useful. But, having something in fast memory isn't useful if you don't have an index to it -- the absence of which (contextually) is flashcard blindness.

> "flashcard blindness" [...] sometimes it was like I had to learn two words, the word I 'learned' in anki and the word in the wild. Even though they were obstinsably the same word somehow they couldn't connect.

I've also felt this problem a lot too, but haven't found a good alternative for self study. Anki/flash card needs to be a support, that uses the minority of your study time, and links closely to the other parts. This is easier when following a class, than when doing self-study.

It is so easy to build elaborate flashcard strategies and get lost in implementing them.

Also, I added example sentences to the flashcards (early version of the tool didn't have them) and it seemed to help with this problem. Still, all valid points and I appreciate your input :)

I've definitely experienced what you've mentioned. Learning a word is a two-step process, first you memorize its definition, then you see it in context and its meaning sinks in. But I think both parts are useful. If I just read without flashcards I forget words too quickly, and the burden of looking up words again and again slows me down a lot.

I'd say you need a mixture of methods, and the ratios change with progress.

Flash card-based methods are very fast, especially in the beginning. Reading goes further, but is too slow for the first maybe 1000-2000 words (depending on the language).

Reply on news.ycombinator.com