Flashcards: configuring the "show" card - what's most effective?

[etm001]

Hi,

 

Whether using Anki, Pleco, etc., you can configure what information is displayed on the "show" side of a flashcard during a test, including:

• The Chinese character(s)
• Pinyin
• Definition
• Audio pronunciation
• etc.

The most basic test would be having the Chinese character(s) on the "show" side of the flashcard, with the definition on the "reveal" side of the flashcard. And my question is simply this:

• Beyond the simple "show/reveal" configuration just described, is there another configuration that would lead to improved retention beyond this basic configuration? (Assume that SRS is used for card selection).

By "other configuration", I mean specifically what information is included on the "show" side of the flashcard during testing. For example, in Pleco you can dynamically configure what information is displayed on the "show" side of the flashcard. This is called "configure score subjects". Here's a configuration that I have used before:

• New card: show characters, pinyin, audio pronunciation
• If card score >= 200: show characters and audio pronunciation
• If card score >= 400: show characters only
• If card score >= 800: show definition only
• (If card score >= 1200: audio only)

(The last score level I rarely use, given the sheer number of homonyms or near-homonyms in Chinese)

 

Essentially, the higher the score for a given flashcard, the less information that is displayed on the "show" side of the flashcard. With increasingly less information displayed on the "show" side, it becomes increasingly difficult to answer the card correctly. I realized that while this may be true, this approach may not necessarily lead to better learning/retention beyond the simple approach I described earlier. That is, I have no empirical evidence that learning/retention is better with this dynamic approach as opposed to the simple approach. In fact, I wonder if the dynamic approach may have a negative impact. There's no doubt that my scores are lower using the dynamic than they would be if using the simple approach. In short, I've started to wonder if the dynamic approach has an actual benefit over the simple approach, and if anything whether it might be an impediment to effective learning/retention.

 

If anyone knows of empirical evidence from studies that focus on this form of learning, and which might shed some light on my questions (i.e., the simple configuration is fine; the dynamic configuration does not offer any real benefits over it, etc.), that would be great.

[querido]

I don't think you'll get an authoritative answer such as one from "empirical evidence from studies" and even if you did it would remain arguable due to individual differences and preferences outweighing small theoretical differences. I say find something comfortable and non-distracting so you can forget about methodology and just keep working; try not to spend too much time on questions like these.

[imron]

I have 3 profiles configured in pleco.

 

The first profile is for new cards and it is configured to show everything on the front (and no back) but only shows cards once.  This is my warm up queue for newly added cards that I use to go over and learn new words fully before flashcarding them normally.

 

The other two profiles are 'revision' and 'pronunciation'.  Revision has pinyin only on the front and Chinese word + Chinese meaning on the back.  I use this to test whether I can actively recall how the word is written.  I don't bother memorising the full definition in Chinese, I just think of the general concept and make sure it matches what I thought.

 

'Pronunciation' has the characters on the front and then Pinyin and Chinese meaning on the back.

 

When reviewing, if I get something wrong (or didn't get something right fast enough), I'll also stop and spend a bit of time going over the bit I messed up to consolidate things and make sure that it won't happen again.

[Wurstmann]

Mine look like this:

front:

 

back:

 

It's an easy format so the creation and review of the cards doesn't take too long. I add words I encounter while reading books or articles on the web.

Anki is set to show me 20 new cards every day.

[Friday]

Piotr Wozniak, the creator of SuperMemo, the earliest SRS software, has some good articles on how to design cards. I have not found any outside research studies supporting his claims, or further studies on SRS of much value to language learners, but from experience, found his advice very valuable. I wasted much time ignoring his advice and made little progress.

 

The most important of his points, is to keep the cards as simple as possible. Put only one tiny fact on the answer side. It is better to have dozens of cards with the answer broken into parts instead of one card filled with much information. Make the cards easy for you. You'll go through them much faster. Even with the data all spread out amongst cards, your mind can still make the connections between them.

[etm001]

Thanks to everyone who replied. 

 

 

 

Piotr Wozniak, the creator of SuperMemo, the earliest SRS software, has some good articles on how to design cards. I have not found any outside research studies supporting his claims, or further studies on SRS of much value to language learners, but from experience, found his advice very valuable. I wasted much time ignoring his advice and made little progress.

I'll search around and see if I can find some articles he wrote. I'll report back if I find anything that answers my initial question.

[imron]

They're all on the super memo site. There are loads of them, and all very interesting.

He also recommends doing extensive reading and cloze deletion.

[stapler]

My set up changes depending on what I am focusing on.

 

I can't carry on a conversation in Chinese nor can I even say something more complicated than "不用跑!" or "好吃“. While listening I often rely on trying to pick out key words to get the meaning of a sentence. Sometimes I get the meaning/intention very wrong, even though I know what topic is being discussed. So because my listening is awful (and I can't really practice speaking by myself) my flashcards are built around trying to improve that at the moment. I drill sentences that have nothing but audio of a spoken sentence on the front then the hanzi on the back. For sentences I don't have real recordings of I have pinyin on the front and the hanzi on the back. Reading the pinyin is much harder for me than reading the hanzi; it's much harder for me to read a beginner's textbook  in pinyin than it is to read a Chinese novel. So I use flashcards with a sentence written in pinyin on the front in the hope that it will focus my mind more on the 'sound'. I feel like reading hanzi is cheating because it makes recognition so much easier with the visual clues. At this point I'm even thinking that flaschards are a waste of time for practicing listening. You can't drill your way to "audio fluency" (is that a thing?). So I might just start trying to watch TV or listen to audiobooks again (I always give up really quickly. I find it too difficult or boring doing listening practice). In fact the only reason I probably use flashcards right now is because I am too lazy/scared to actually commit myself to seriously engaging with Chinese. The cards give me a sense that I'm practicing and improving even though it's clearly not doing much at all for me, and clearly wouldn't be as effective as actually trying to listen to real conversations.

 

My reading abilities are fairly strong. I consider myself largely literate and can read most things I encounter so the only thing I flashcard now is an expired Skritter account that has 4000 characters on review (takes me just a few minutes a day knock them off) to keep my character recognition skills up to date (particularly for the more obscure characters). I also use the default flashcards in Pleco. I add new words as I encounter them. My main goal is to remember the definition though I will also mark them wrong if I forget the tones (though that's rare). I review these two or three times over the course of a week or two then delete them all and continuing adding new words I encounter while reading. Most of the words I add are fairly obscure (not rare, but not daily/basic conversation stuff) so I'm not too fussed about commiting them to my long term memory. It's enough that I can recognise them on reading rather than being able to actively use them.

[imron]

The cards give me a sense that I'm practicing and improving even though it's clearly not doing much at all for me

Something I've been saying for quite sometime now is that you should practice what you want to get good at. If you practice drilling flashcards then you'll get good at drilling flashcards, and that will not be necessarily be useful for real world activities.

To get good a listening to TV shows you should watch TV shows (yes it will be difficult at first) and to get good at reading you should do reading.

Flashcards can help as a revision aid, but they shouldn't drive the learning process. There're too many things they don't drill you on - especially with regards to ability to continually process information at speed over a sustained period, which is a vital part of being able to use the language.