I know the topic of "kana>kanji, or kanji>kana, etc..." has probably already been debated to death. But I know ya'll like a little bit a debate/discussion once in a while , so I was wondering if anyone had found success with Khatz' image hack. By that, I mean where you put a picture to help remind yourself of the topic of the sentence or whatever. I think I may do this when I start sentences since I want to do the kana > kanji cards. Since I've heard those can be a time-consuming pack of rabid beasts, I might do this to soften them up a bit. Has anyone else done this? For me, there's this random temptation in the back of my mind to just "play it safe" and do the traditional kanji sentences, but whatev. Has anyone found success with this? Let's hear your thoughts!

Well from what i've seen while using iknow the pictures make it much easier to recal when your not sure of the word, i've also tryed it with a couple of the more difficult sentences and it does seem to make it easier to remember as it gives you some idea if your not sure, as they say the more connections to a word the better.

Also i can see this being a great advantage when going monolinguel as even if you have trouble with the deffinition the picture will give you an idea.

The answer is an emphatic yes. Using images gives you stronger associations to the word.

There were threads earlier (too lazy to look them up right now) that talked about this. One guy drew his stories. Granted, that's for RevTK methods.

For the sentences, I think iKnow is good proof that a referential picture helps burn meaning into memory (not necessary, but it helps). It's really good if you want to start learning Japanese names. Just get photos and names from d-addicts.com wiki and throw them into Anki or iKnow.

The kana to kanji has not been debated to death. From what I've seen, most tend to agree with the concept. I altered so it's audio to kanji (with kana as reference if needed). While it was a time consumer, I found a simple way to speed up the process. I began just dictating the single vocabulary word the sentence was created for. Since the reviews went quicker, it allowed me to study new material more often.

So, I think dictation offers the best way to test your knowledge of both the spoken and written language. However, it will take the longest. Find a method that works.

I've also (thanks to iKnow), been using picture references. It too has been great. If you think too much info is being given to you think about this: you're hearing a word in a sentence , seeing a situation dealing with the word, and able to write it out perfectly from memory. Could Joe Blow off the street do that?

Okay, I pretty much agree with you guys. And I've heard Khatz' argument that Nukemarine is referring to there at the end. I pretty much agree with that, too. But...just to play devils advocate, here...what about sentences that are mainly vocabulary-ish? I'm thinking of an example from the Japanese in Mangaland here (too lazy to look it up), but it was something like "I've broken/chipped a tooth." Even for me (low-skill level), the grammar was pretty easy. And for the most part, the vocab was easy too except for the word tooth. So I guess you would call this a "vocab sentence." In this example, would it be cheating to put a picture of a chipped tooth or just a tooth or something? I mean, think about it. In essence, you're basically telling yourself half of what you should be trying to remember (the meaning). Whatever, I suppose that doesn't matter that much b/c, like Nukemarine said, my aunt sally would have no clue why there's a picture of a tootht there. But, I'm just worried that you'll weaken your connection/ability as far as the kanji for tooth. If you encountered it somewhere on it's own out in the wild, or if you had to write the kanji off of the top of your head randomly, would you still be able to do it? Like I said, just playing devil's advocate. I'm sure there's a sufficient answer for this.

I don't personally use images (if nothing else, too danged lazy), but your arguments are essentially the same as the arguments for why you should drill vocabulary by itself because in a sentence you're telling yourself half what you need to know and etc. So the answer to the argument is the same: more context is only a good thing. You can't weaken the connection by adding more connections. It's not like plugging another appliance in to drain away the power, it's more like adding another leg to a table. You're only making it more stable.
Because the goal is something else. People are often inclined to think that they need to test themselves hard in an SRS or whatever, to prove that they know it, but that's not what an SRS is for. It's not for testing. It's for learning. The test is when you see it somewhere real. Now, if you just have a single vocabulary word in your SRS, when you see the word for real, the only reference point you may bring up for it is a word in your SRS with a vague (English?) definition in your head. That.. may not be enough to go on. But what if you have the word, a sentence (or several) that put it in context, a sound file, an image, and a memory of the specific pieces of fiction (or whatever) you've encountered it in? Your chances to pass the REAL test are a lot better.

igordesu Wrote:I'm thinking of an example from the Japanese in Mangaland here (too lazy to look it up), but it was something like "I've broken/chipped a tooth."

New method:

Ok... Stay with me here... Get a hammer and a digital camera ready....

Ok, not the most popular idea...

But a really great method is to find pictures with short descriptions in Japanese already under them, found in; news articles, magazines, blogs, etc. In other words search picture -> description, not description -> picture. Going description -> pic can be incredibly time consuming and frustrating, imo. The pictures that already have descriptions to them often follow a story, are more memorable, and are easy to drop in your deck. Also, I haven't personally tried this, but taking your own pictures may also be a great way to reinforce vocab, just ignore the above joke.

Ok, here's what I do. I have a sentence that contains a noun that I learned the meaning of through a google image search. So when I make a card, I drag the image from the google image search page into my notes field in Anki. Simple, simple, simple.

Maybe I'm a perfectionist, but I've always had trouble finding images that I like, and choosing one can take a ridiculous amount of time. Simple, simple, oh crap, I just lost one hour googling. @alkys; good point, I should stick with nouns, it's stuff like "criticism" and "fall flat on one's face" that gets me snookered.

igordesu Wrote:Okay, I pretty much agree with you guys. And I've heard Khatz' argument that Nukemarine is referring to there at the end. I pretty much agree with that, too. But...just to play devils advocate, here...what about sentences that are mainly vocabulary-ish? I'm thinking of an example from the Japanese in Mangaland here (too lazy to look it up), but it was something like "I've broken/chipped a tooth." Even for me (low-skill level), the grammar was pretty easy. And for the most part, the vocab was easy too except for the word tooth. So I guess you would call this a "vocab sentence." In this example, would it be cheating to put a picture of a chipped tooth or just a tooth or something? I mean, think about it. In essence, you're basically telling yourself half of what you should be trying to remember (the meaning). Whatever, I suppose that doesn't matter that much b/c, like Nukemarine said, my aunt sally would have no clue why there's a picture of a tootht there. But, I'm just worried that you'll weaken your connection/ability as far as the kanji for tooth. If you encountered it somewhere on it's own out in the wild, or if you had to write the kanji off of the top of your head randomly, would you still be able to do it? Like I said, just playing devil's advocate. I'm sure there's a sufficient answer for this.

Again, this derives when I began to stream line how I reviewed my sentences. In the same way I began to just dictate the vocabulary word and not the entire sentence, for "grammar" sentences I didn't dictate at all. I figured now grammar sentences are not there to teach you vocabulary. If anything, you're going to get a lot of repetition of vocabulary. I can see photos being useful when you're learning grammar for verbs where direction of action is important. 来る、行く、返す、帰る for example. If the problem in your grammar sentence is a vocabulary that any other word can go, replace the word if you got to. I stopped failing grammar cards based on missing words in them not related to the grammar. But then again, a photo of chipped tooth sounds perfect for your sentence, even though it's a grammar point about passive/active verbs.

Recently though, for grammar sentences (and those only), I've been doing production (ie translation) cards and recognition cards. These are from Tae Kim's sentences which seem good for going English to Japanese.

As for out in the wild, you're going to miss some just like in reviews. I think though that after 2000 or 5000 or 10000 cards, what you miss (on things you studied in Anki, but see in the wild) begins to pale in all that you get right.

Kind of the "do you want to spend time getting 1000 cards 100% perfect, or use the same time that would get 2000 cards 80% perfect" line of thought. If images speed up the process, use them.

Good call. Thank you very much for all of your responses. You have...dispelled all of my doubts about images, lol... No, but really, thank you. Now that I think about it, it really wouldn't be cheating. It's like you guys said. If I ever did have to write the word for tooth off of the top of my head, then I guess the photo of the chipped tooth that I had seen would help remind me of the kanji or whatever. You're right, it really is just a helpful way of building more "connections." Yea!...now I'll be ready when I start sentences.