Thursday, April 29, 2010

It takes 15-20 exposures to learn a word

In an article on VOA News, Catherine Snow of Harvard's Graduate School of Education gives us some numbers on how many exposures you need to a word before you learn it:
In order to have a high probability of learning a word, you need to encounter it fifteen, twenty times.

Labels: ,

Tuesday, April 27, 2010

Getting to Grammar: Learn grammar through an ad hoc spaced-repetition system

Extracted from the current manuscript of the book, to the right you'll find the meat of how my preferred method for learning grammar works, in convenient flow chart format.

As with learning any piece of knowledge, you'll learn a grammar rule best through spaced repetitions. As such, through much trial and much (much, much, much...) error, I've found that combining a wide variety of repetitions works best. Although the repetitions do not have any systematic spacing based on a forgetting curve as spaced-repetition systems are supposed to, there should be enough repetitions here to get the rules in your head.

Let's take a walk through that flow chart, after the jump.

Read more...First, let's look at the big picture. The first three steps get you started in developing an understanding of the grammar. These steps are front-loaded in the process and can typically be taken care of in a few weeks (if you're going to the language zone, they are also good things to do before you get there). The next two steps move onto exposure, which is by far where you'll be spending most of your time. Finally, the last four steps allow you to refine your grammar to cover things that haven't already clicked, but you only need to proceed to those steps as necessary. The flowchart then always turns you back to exposure.

The above graph is an approximation of how you'll be spending your "grammar time". So what exactly do I mean by "grammar time"?

No, not quite.

It's the time you spend getting repetitions of grammar. When you're just getting started, most of those repetitions—and thus your grammar time—will come from you reading and working through the grammar, with some exposure to the same from the other things you're doing in the language. From there on, most of your reps will come from exposure, and you'll also spend a gradually decreasing amount of time refining your grammar, until finally you're spending virtually no time at all focusing solely on grammar.

That's the big picture. Now let's run through each of the steps.

Reading through your grammar is just what it sounds like. Get the grammar in front of you and read through it. This should be a relatively quick read-through; don't get hung up if you don't get something right away. Try and get an idea of the rule and move on. You shouldn't spend more than, say, 5-10 hours doing this (and you can probably get away with doing even less). This read-through will give you your first repetition—a relatively weak reading rep—of all the grammar rules.

Then we get to outlining. What exactly do I mean by outlining? Taking the grammar you see in front of you (another reading rep of the grammar rules), trying to make sense of it (an analysis rep, which will cause the rules to more readily stick in your head than mere reading reps), and then outputting it into a Word document or the like (a writing rep) in a way that makes sense to you (be sure to include plenty of tables, charts, etc.). The document you produce should be a raw and bare explanation of the rules, using example sentences and the like only when absolutely necessary.

The outlining step will typically take some time, but you should be able to get through it in several weeks (if you're studying the language full time, make that 2, maybe 3, weeks, assuming of course that you're doing other stuff as well; if you dive into an outline non-stop, I'll bet you can kill it off in less than a week, although I've always preferred mixing things up a bit more than that). If it's taking you a lot longer than that, you should reduce the level of detail you're putting into the outline. For anything that you ponder for, say, 10 minutes, and still don't get, mark it as a question in the outline and come back to it after you've covered everything else.

If you still don't get it after your second pass at it, it's time to ask a native speaker or a language-leanring forum. I often throw tricky questions to native speakers on Lang-8, and it's also common practice on italki. Some good forums include How to Learn Any Language and WordReference. These won't be able to do everything in every language, but there are of course plenty more forums to be had through a few simple searches. And feel free to ask any native speakers that might be nearby, as well. Once you get an answer from any of these sources, update your outline accordingly.

That takes us through getting started, and now it's on to exposure. This just means reading, listening, writing, and speaking the language. (Here's a little process for doing all four of those things using free, online tools.) This will get you exposure to the grammar in use, both passively (reading, listening) and actively (writing, speaking). While this requires little explanation, it will be by far the place where you get most of your grammar reps, with the goal of it ultimately being pretty much the only place where you get grammar reps.

Finally, we come to refining your grammar knowledge. If there's something you see or hear and don't understand, or if there's something you keep screwing up in writing or speaking, it either means that the grammar rule is new to you or that you've forgotten it. In either case, the refinement process will help you get the additional reps you need to burn the rule into your brain.

The first step is to simply review the problematic grammar in your outline (a reading rep). If you've already got the correct rule in there, review and get back to getting exposure. This will typically be the solution when you've merely forgotten the rule and just need a refresher.

On the other hand, if the rule was never in your outline, or you didn't really get it right the first time, then it's time to edit the outline. Add or fix the rule, as necessary, and get back to exposure.

If you don't get the rule, or think you've got it but still seem to be getting it wrong, it's probably time to turn to others for help. Just like above, native speakers and forum participants should be able to answer your questions.

As an absolute last resort, you can actually memorize problematic grammar. This means making grammar items in your spaced-repetition system. This should generally be unnecessary, but if you've found that you keep going through the refinement cycle without really getting the rule, this can help you finally nail it down.

And that's it. At the end of all this, the process should lead you to getting just about all of your grammar reps from exposure, with a very occasional dive into the refinement process.

Labels: , , ,

Sunday, April 25, 2010

A single workflow to make use of online language-learning tools

There are so many language-learning resources out there on the web, it's kind of tough to figure out how to make use of them all. In looking at how I'm using these tools myself, I put together the following little process to incorporate many of the language-learning tools I've been using into a single workflow:


Oh, and this workflow is completely free.

Let's walk through this, after the jump.

Read more... Start with reading and/or (but preferably and) listening to something in the target language. LingQ is all about content with both text and audio, so that's a good place to start looking, but you're hardly limited to LingQ; any recordings you can find with transcripts, unabridged audio books (including children's books), etc., will do the trick.

To the extent there's anything you don't understand in the text or audio, look it up and add it to your spaced-repetition system. Anki is my current SRS of choice, but some other popular choices are Smart.fm and Mnemosyne.

Then write something about what you read or listened to in the target language. Try to make use of whatever you needed to look up and add to your SRS, and to the extent that you need to look up anything else, add that to your SRS as well.

Then get that writing corrected. There are a number of ways to do this, but Lang-8 is my standing favorite, and italki recently implemented this feature. Again, if the corrections include things you need to look up, add them to your SRS system.

Once you've got the corrected text, record yourself speaking it and get that recording corrected by native speakers. I use Cinch and Lang-8 to accomplish this.

You've now written and read that writing. Now it's time for some plain old talking. Making use of everything you've learned thus far, record yourself saying something about the running theme and get that corrected in the same way you got the recording of your text corrected. Once again, if the corrections give you any thing that needs to go into your SRS, add it.

At this point, you should have everything you need to get in your SRS. Now go over to RhinoSpike and get native speakers to record the pronunciation of each of those words. Take those audio recordings and add them to your SRS system. From there, you just need to review your newly added items as part of your regular SRS review.

You've also got two things that you've recorded yourself: your corrected text and some plain old talking. Go to RhinoSpike again and get a recording of both from native speakers. Once you've got those recordings, add them to a playlist on iTunes and listen regularly. I'd recommend just throwing all of these recordings into a random-order playlist and listening to them in the background while doing other things. This will provide a review of all of the above.

This entire workflow can be tailored to your level. At the most basic level, you can even use children books; my kids have plenty of books that come with audio CDs in all three of their languages. But you don't necessarily need to dumb the text down; you can also just keep it short. For example, if you're just starting a language but want to read a news article, you could limit yourself to just the first paragraph. This will likely take a while, but it won't be insurmountable.

If you've got a way to make this workflow, I'd love to hear it!

Labels: , , , , , , , , , , , , , , , , , ,

Friday, April 23, 2010

This is your brain on languages.



The image you see here is a visualization (which is obviously not comprehensive) of how a given piece of information in a language might get lodged into your brain. The piece of information could be anything: a vocabulary word, a grammar rule, pronunciation, a character, etc.

Every one of those lines emanating from the piece of information connects with one kind of exposure. The more exposures you get, the more connections your brain draws to that piece of information. The more repetitions of a given kind of exposure, the stronger that exposure becomes (imagine the lines getting thicker with each exposure). The stronger and more plentiful your exposures are, the more likely you are to remember the piece of information.

Exposure to a language can be largely divided into reading, listening, writing, and speaking. It doesn't matter if an exposure is via reading/listening (i.e., input from an external source) or writing/speaking (i.e, output to an external target). These traditional ideas of "output" and "input" are both input as far as your brain is concerned.

Output is input.

Labels: , , , , , , , , , , ,

Saturday, April 3, 2010

Get Cramberry (spaced-repetition app) for iPad and iPhone for free (U.S. residents only)

I've mentioned before that my current go-to spaced-repetition system is Anki, but there are a lot of other options out there, including well-known systems such as Smart.fm, Mnemosyne, and SuperMemo.

Another contender in the field is Cramberry. What's kept me from making more use of Cramberry is that you can only study 30 cards per day in the free version of their web app. That said, they're doing a promotion right now that will get U.S. residents their iPad apps for free, and the first 50 people to download the iPad app can also get their iPhone app for free. And I'm guessing that those apps, which currently cost $2.99 (iPad) and $4.99 (iPhone), don't have the any study count limitations, even when you're getting them for free.

Get your free apps, after the jump.

Read more... From an email that went out to Cramberry's mailing list earlier today:
To celebrate the launch of our new Cramberry app for iPad, we're giving away free copies of Cramberry for iPhone and iPad. Here's how it works: Send us an email (contact@cramberry.net) telling us how you use Cramberry, and we'll send you a coupon code for a free copy of Cramberry for iPad. Once you've downloaded the iPad app, send us your iTunes receipt, and we'll send you another coupon code for Cramberry for iPhone. You must download the iPad app to receive the iPhone app. You can download the app from iTunes on your computer; you don't need an iPad to participate in this offer.

Note: this offer is only available to U.S. residents. Sorry!
After I sent them the required email, stating that "I use Cramberry for language learning, of course!", they wrote back to me with further details, including this one that they should have probably had in the first email:
Only the first 50 people to download the iPad app will receive free copies of the iPhone app, so act quickly!
So while the iPad app appears to be available for all, the iPhone app only goes to the first 50 to get in line.

If you're not already signed up for Cramberry, you probably should sign up and give it a whirl before contacting them. With a little luck, I'll be checking out a free copy of their iPhone app shortly (don't have an iPad, nor any plans to get one, so checking that out will have to wait, despite the fact the download is already sitting in iTunes).

This post was updated a few hours after the original post to reflect that only the first 50 downloaders of the iPad app can get the iPhone app for free, which was only revealed by Cramberry in the email in which they sent the promo code for the iPad app.

Labels: , , , , , , , ,

Monday, August 17, 2009

Can you help me improve my language-learning routine?

I thought I'd share with you what's shaping up to be my language-learning routine. I'd love it if you could take home a few good pointers from my routine, but I'd love it even more if you could give me a few good pointers to improve my routine.

My days are, predictably, dominated by Japanese and English. I try to maximize my use of Japanese because of my need to use it at work, but there are two places where I use English as a matter of course. The first is with my kids; I only use English with them, and my wife and I speak English to each other whenever we're in earshot of them, in order to maximize their exposure to English. This is of course a direct trade-off between my Japanese and their English, but one I'll take to prevent them from speaking Engrish. The other place I use English regularly is of course at work when I need to do any of the various things a lawyer might need to do in English.

My language-learning day gets kicked off with my morning alarm; I awake to the sound of Japanese podcasts giving me today's news. Breakfast with the fam is largely in English, although my wife always speaks to the kids in Japanese and the nanny speaks to all of us only in Chinese, so that'll be floating around as well. My mother typically joins us for breakfast via video chat, so once in a while she and I will use some Italian when we don't want anyone else to understand.

Read more... Whenever I'm walking around (such as to, from, and in train stations) or standing around (such as on trains when I can't get a seat), I use my iPhone to listen to podcasts and to review vocabulary with iAnki. My first iAnki/podcast stint every day is from the time I leave my apartment until I sit down on the train to work.

Once seated on the train, the podcasts continue, but I typically break out my computer and try to get stuff done that often doesn't involve a foreign language—doing actual work, responding to emails, working on the book, or preparing these blog posts. When I arrive at the station at which I get off, I return to iAnki/podcasts until I get to my office.

Once in my office, I switch from listening to podcasts on my iPhone to listening to them on my laptop quietly in the background, and I keep them playing in my office the entire time I'm there. I also run a screensaver that shows selected vocab on my laptop screen while I work from the firm-supplied computer. You do end up glancing at it from time to time, and it's especially useful for getting extra exposure to things you've been struggling wtih.

Although I end up doing much of my work in English, I get exposed to plenty of Japanese over the course of the day. Once people figure out that my Japanese is passable, they typically stop using English with me whether via email or in person (and I of course encourage this by using Japanese as much as possible). I also regularly have to deal with Japanese-language documents, websites, etc.

All of these serve as founts for vocab to feed into iAnki and from there into my brain. As I come across words and phrases that I'm unfamiliar with over the course of a day, I quickly note them down in an Excel spreadsheet. Before I leave the office each day, I send the Excel sheet I made over the course of the day—which typically has somewhere between 15 to 30 items in it—to my personal email. When I get home each night, I look up all the words, get example sentences, and add them to iAnki.

Whenever I write Japanese, I get it corrected, review the mistakes, and make any new items for iAnki that might be necessary (by first adding them to that Excel spreadsheet). My secretary helps to correct any Japanese I put together for work, but I've been submitting everything else to Lang-8 for corrections—totally gratis. On Lang-8, native speakers of the language you are learning will correct your writing (and you're expected to reciprocate). Response times are impressive, and I've rarely waited more than a hour for corrections, and certainly never more than a day.

As for other languages I encounter at work, I treat them the same way I treat Japanese. As I'm part of the China Practice Group at my firm, I regularly get exposure to Chinese. I've also had to review documents in other languages, such as Spanish and French, and there have been phone calls to Latin America, so any words I've had to look up have ended up mixed in with my mostly Japanese iAnki reps.

Whenever I get the chance, I'll revert to podcats/iAnki, e.g., on a walk to the bank, which is maybe 5 or 10 minutes away from my office. And whenever I get a little bit of time in which I can't effectively do anything else—such as if I'm on hold on a phone—I'll quickly pull out my iPhone and do a few reviews on iAnki. Even if I only have 30 seconds, I can probably get through at least 10 reviews in that short a time period.

On the way home, it's back to iAnki/podcasts. I typically can't find a seat until maybe halfway through my ride home, so this is typically the period each day in which I spend the most time reviewing vocabulary. Once I do find a seat, I break out my laptop and do the same kinds of things I do on the morning ride, while continuing to listen to the podcasts. And, once again, the walk from the train to home is more iAnki/podcasts.

Once home, I add the new items from the Excel spreadsheet mentioned above to iAnki and see what I've managed to do over the course of the day. Typically, I'll get through somewhere between 300 and 500 reviews in a given day. I'll then make any changes necessary to the items in iAnki (such as adding example sentences to things I'm struggling with), as well as updating the vocab words in my screensaver.

It's also at night when I do thing like read news in other languages, although I don't spend as much time doing that as I'd like to.

And that's pretty much my routine as it currently stands.

I am looking to make a few changes, however. One thing I've been puzzling how to do efficiently is bring in languages other than Japanese in a more systematic manner. I think I'm going to do this by assigning a time percentage to each language and then listening to podcasts in each language accordingly. Ideally, I'll be able to find podcasts with transcripts and then review those as well, and then put the vocab into iAnki.

And, of course, I'm sure you might have some tips for me as to how I can improve this routine, so please drop them in the comments below!

Labels: , , , , , , , ,

Monday, August 10, 2009

Database of Spanish sentences

Ramses of Spanish-Only.com is a man who likes his sentences. While I tend to butt heads with him on grammar-learning methods, I do agree with him wholly on the value of sentences in language learning. I use them (although not solely, as he does) to learn vocab, and finding good sentences can be something of a challenge.

Well, Ramses has made it a little easier for you Spanish learners with this database of Spanish sentences and their English translations (hat tip: Babelhut).

There's nothing quite like context to help you learn a language, whether for vocab or catching grammatical rules in practice, and sentences give you bite-size chunks of context good for just that. They won't be quite as useful for you as they were for Ramses (who got them from their original context), but the database can save you some work if you're looking for a ton of sentences that already have translations conveniently located nearby.

My one request at this point is a way to easily incorporate these into a spaced-repetition system. Currently, as far as I can tell, you'd have to pluck out the sentences one by one, but hopefully that is something that will change over time as it continues to be a work in progress.

Labels: , ,

Wednesday, August 5, 2009

Review of iAnki for iPhone: Not very elegant or convenient, but gets the job done

As I mentioned before, I've been looking for a spaced-repetition system to use on my iPhone that I can (1) sync with a desktop app and (2) use when there's no internet connection (because I spend time on a subway line every weekday from which I can't get online).

So far, it looks like Anki's solution for an unbroken iPhone, iAnki, is the best option available, although it's far from ideal. The software is testy, something of a challenge to get working, and syncing with Anki on the desktop can be a headache, but its core study functions by and large work fine and, in the end, you do get two-way syncs with Anki's desktop application.

Let's get into the nitty-gritty, after the jump.

Read more... Installation. With iAnki, you're quite likely to get off on the wrong foot because installation is a pain. As if needing to deal with some quirky work-around to a true iPhone app isn't enough (iAnki is not an iPhone app—more on that below), the instructions on the iAnki page are less than clear. I've got a few nerd credentials, and it took me some time to get things working, so I feel for the even less nerdily inclined. Nevertheless, after somehow bumbling through the instructions, I eventually managed to get everything up and running, so hopefully you will be able to as well.

Syncing. But I couldn't jump into using iAnki without first having a syncing headache; the syncs were just stalling out and clearly nothing was happening. I put this question up in iAnki's forum, but before I could get a reply I simply deleted everything and started from scratch. I must've done something different, because the second time around I got the sync to work.

In terms of syncing, Anki and iAnki work together in a bit of a strange way. First of all, iAnki is not an iPhone app at all, but rather a bookmarked web page that you use Safari to run. To sync with Anki (and to set up iAnki on your iPhone in the first place), you need to download a plug-in called iAnki Server. iAnki Server runs over your local network to sync Anki and iAnki. iAnki unfortunately does not make use of AnkiOnline, so both your computer and your iPhone need to be connected to the same local network to use iAnki Server.

If you're having trouble keeping Anki, iAnki, iAnki Server, and AnkiOnline straight, you're not alone. Here's an image of how the set-up works for me that'll hopefully clear things up a bit:


So my laptop syncs to AnkiOnline and then to the iPhone via iAnki Server on the local network hosted over our Time Capsule. There's no way to sync directly between the computer and the iPhone, as far as I can tell (I suppose there might be a way to set up your laptop as a server, and if you can explain how to do that, I'd love to hear from you in the comments).

At the same time—and somewhat inexplicably—iAnki can't sync up directly with AnkiOnline. As I mentioned above, iAnki is just a bookmarked page in Safari, so I don't see why it can't sync with AnkiOnline instead of needing to deal with the hassle of using iAnki Server on a local network. (As an aside, AnkiOnline works reasonably well on an iPhone—especially in full-screen mode (it's in the left-hand column of AnkiOnline)—but as I need to be able to study when I can't be online, it's not a solution that will work for me.)

Since iAnki doesn't sync online, it means that if I study on my iPhone on the way to work and then use AnkiOnline during the day, I'll be reviewing the exact same things, since one won't know that I've already done it on the other. However, I can live with that; I just do all my reps during the day on my iPhone rather than making any use of Anki or AnkiOnline.

Basically you've got to remember to sync iAnki with iAnki Server regularly. You can lessen this burden by syncing lots of cards so that, even if you forget to sync for a few days, you'll still have plenty to work with. However, as happened to me a few days ago, you'll probably eventually forget to sync for enough consecutive days that you run out of due cards and end up playing Labyrinth 3D instead of being productive.

And today I discovered another fun headache. When I got home, it wouldn't sync and iAnki Server was giving me some kind of unintelligible gobbledygook error message. I eventually discovered that because my Time Capsule had assigned my laptop a new IP address (ending with a .6 instead of the .7 it had before), this was causing big problems. I first fixed it on iAnki Server's end, and that got rid of the weird error message, but I needed to create a new bookmark on the iPhone following iAnki's installation instructions before I could sync.

One additional annoying aspect of the sync process is that, when syncing multiple decks, you need to click "OK" on the iPhone after each deck or else it won't begin syncing the next deck. This means that you've got to babysit the iPhone during the sync process if you want to make sure that it gets through all your decks. I can handle clicking OK once at the end after all decks are synced, but needing to repeatedly click OK is just a nuisance.

All this leaves me looking forward to seeing what the next sync headache will be.

So, bottom line on syncing, let's cross our fingers for automatic AnkiOnline or even direct Anki syncing. The latter probably requires an actual iPhone app, but the former should be much easier to accomplish.

Studying. And last, but definitely not least, is the core studying function. It is, unsurprisingly, iAnki's strong point and is stripped down but is much like Anki's. You're shown the question and a button to reveal the answer. Once you reveal the answer, you have buttons from 1 to 4, paralleling Anki's buttons, with 1 meaning you didn't get it, 2 meaning you got it but it was tough, 3 meaning you got it, and 4 meaning you got it easily. Press your button of choice, and it will move onto the next card and tell you when the card you just scored will be shown again.

I do have two complaints about the studying features, however. The first is that there is no undo button. On occasion I'll accidentally hit the wrong button and want to go back and press the right one. There's just no way to do this. To take the extreme example, if you accidentally hit 4 on a word you've gotten correct a bunch of times but have just recently forgotten, you might not see that card again for months. A simple undo button in the top left corner of the screen would be ideal.

The second is the lack of any way to flag cards. For example, if a card is just plain wrong, or is confusingly similar to another card, a way to flag it and then fix it in Anki would be a great help. This could correspond to Anki's "mark" feature, and could be implemented with the addition of a button in the top right corner of the study interface.

* * * *

So what's the final call on iAnki? I'm going to keep using it. It may have its quirks and difficulties, but it's doing what I need it to do. It syncs with the desktop app and I can use it when offline. The most important part—the study functions—are solid, and the other issues are manageable.

And, on top of that, I'm not sure that I have much choice, as I don't think there's anything else that does what I'm looking for, so if you know of any other options, please drop a line in the comments.

Links: Anki, AnkiOnline, iAnki

This post was updated twice on August 6, 2009. The first time was to delete a complaint that the font sizes are too big in iAnki. As it turns out, the font size on iAnki is completely under the user's control (via Anki). Victor, the author of iAnki, pointed this out to me in a comment below, and sure enough it turns out the issue I was complaining about was my own fault. In Anki, I went to Settings > Fonts and Colors... and lo and behold I could change the font size (in fact, I'm indeed the one who set Japanese so large in the first place). It took me a few tries to get that to sync over to iAnki, but sure enough it did and it looks much better than before. I may need to tinker with the sizes a bit until I figure out what works best, but the bottom line is that I can do just that. The second time was to note that the author of iAnki is Victor, not Damien Elmes, the author of Anki.

This post was updated again on October 10, 2009, to note the need to click OK on the iPhone after each deck when syncing and the need for undo and flagging features when studying and to clean up the previous changes.

Labels: ,

Saturday, August 1, 2009

Correcting small children's language use

Steve Kaufmann and I have been going back and forth on the correcting of small children's language mistakes in the comments of this post on Steve's blog The Linguist on Language.

Here's the play-by-play. In my earlier post on the best age at which to learn a language, I wrote:
Kids often get corrected by the adults around them...
Steve replied:
[I]t is not my experience that infants are corrected in their use of language.
And I elaborated:
Although some people correct children explicitly (we do), the usual route is that they're corrected indirectly when the adult repeats the phrase in some way back to them. Think something like a child saying, "Eat apple!", and the adult saying back to them, "Oh, do you want to eat the apple?" It's more subtle than directly correcting, but it's correcting nonetheless.

And that sort of thing is absolutely helpful when learning a language. When I'm not quite getting my point across, then finally I manage to grudgingly do so and the person says the equivalent of, "Oh! So you mean that you want to eat the apple!", I've got the correct way of saying it right there for the taking.
After I posted this comment, I went back and read Steve's earlier comment where he wrote "[kids] hear it and they imitate it", and began to wonder if this was a to-MAY-to / to-MAH-to thing where I'm calling it "being corrected" and he's calling it "imitating".

In any case, Steve continues:
I simply do not buy it. You cannot possibly correct enough errors to make a difference. Children and most good learners correct most of their mistakes on their own. The brain gradually corrects itself as the patterns of the language become clearer.
OK, I'll draw a line in the sand in response to that. My position, after the jump.

Read more... While I generally agree with Steve on most things language learning, the utility of being corrected is one place where we definitely disagree. When Steve touched upon this issue earlier, he wrote:
The idea of perfect strangers correcting my use of language . . . strikes me as just rude, and certainly not helpful.
In contrast, perfect strangers correcting my use of language is always extremely welcome but sadly doesn't happen enough. I definitely find it helpful and I think the difference comes down to how the learner takes the corrections. I've got a post on just this issue in the oven, but for now I'll just flag this disagreement as I think it may add some color to our views on correcting children's language use.

Turning to Steve's comment, I would say that the brain will gradually correct itself as the patterns of the language become clearer, but why wait? My first-hand experience with my daughter has shown me that waiting doesn't pay off. A recent example of a mistake of my daughter's that I corrected was "buyed". I don't recall the exact phrase, but if she had said, "I buyed it", I'd've probably said, "No, you gotta say, 'I bought it'" and, used to the routine, she'd just repeat, "I bought it", without missing a beat and the conversation would continue. Typically she'll use the word again soon in the conversation, e.g.: "What'd you buy?" "I bought..." So, while the brain will gradually figure out the rule from passive exposure (listening), she just had two repetitions of active exposure (speaking), which I always find to be even better for getting something in your head; once you can use it correctly yourself, you've burned some pretty good paths in your brain to that piece of knowledge.

Sometimes I'll even explain the rule to her. When she was having trouble with "geese" being the plural of "goose", I told her that that word is "weird" because, instead of adding an "s", it changes to "geese". After telling her that, the next few times she encountered the word she paused to tell me the rule and then said it correctly. Now she just knows it. In fact, she now actively asks why something doesn't fit into the rules as she understands them (just today I was asked why "fish" doesn't become "fishes" in One Fish, Two Fish, Red Fish, Blue Fish).

I think in any of the examples above it would have taken her a lot more time to get enough exposure to "bought", "geese", etc., to have figured the rules out on her own, although I don't doubt at all that she eventually would have.

The end result of all this is that instead of spending more time trying to learn these passively, she'll get them more quickly and move on to something else. And, although far short of a scientific control, there's actually some basis for comparison with my daughter. My wife, who speaks Japanese with my daughter, corrects her about as much as I do in English. However, we've had Chinese nannies and babysitters teaching her Chinese for most of the time she's been speaking, and they are the generally less stringent about corrections. While living in the States when her exposure to Japanese and Chinese was roughly equal, her Chinese mistakes would linger much longer than her Japanese mistakes. So, while she was moving ahead in terms of grammar and vocab in Japanese, she was progressing more slowly in Chinese. I think the amount and efficacy of the corrections she was receiving was one factor in this.

Based on the above, I obviously don't think that we can't "possibly correct enough errors to make a difference". Beyond the results we've seen, this is really no different than any other spaced-repetition system. If she makes the same mistakes over time, she gets a repetition of that piece of information. If the typical spaced-repetition system requires five or six repetitions to generally learn something, why would kids be any different?

In sum, I feel like I'd be doing her a huge disservice by not correcting her and just letting her "brain gradually correct itself as the patterns of the language become clearer". That'd happen, for sure, but I can't see how it'd be the most efficient way to go about it.

Labels: ,

Tuesday, July 21, 2009

Seeking recommendations for a spaced-repetition system that syncs between your iPhone and your desktop

I recently got an iPhone with one of my main reasons for doing so being to productively fill the time I have on the Tokyo subways when I can't grab a seat and break out my laptop. One of the things I intend to do with that time is using a spaced-repetition system ("SRS") to help expand my Japanese vocabulary. So I'd like to see if anyone out there has any recommendations for such a system.

There are a few features in particular that I'm looking for.

Read more... The most important is the ability to sync between, on the one hand, a desktop SRS app, such as Anki, Mnemosyne, or SuperMemo and, on the other, whatever I use on the iPhone. The idea here is that if I use the desktop app for a minute while on hold on a call at the office and then study on the iPhone on the way home from work, when I get home and sync with my desktop app I'll have my most up-to-date learning data. Accordingly, simply importing from a desktop app is insufficient for my needs.

The second feature I need is that I have to be able to use it on the iPhone while not connected to the internet. Part of my commute to work is underground (and it's also the part of my commute where I'm most likely to not get a seat), so needing an internet connection will not work. Accordingly, I can't simply use one of the SRS websites out there.

Finally, it's gotta be easy to enter the things I want to learn. And that means that I won't have to do it on my iPhone.

So... any recommendations?

I've so far only scratched the surface in my own research (and I'll of course report back when I have more), but I've been considering using Anki's iPhone system, which isn't an App Store app but somehow works nonetheless. I've also been trying out StudyArcade (and considering its $4.99 pro version), but I've yet to fully explore how it syncs up with Anki.

Labels: , , , ,

Thursday, March 26, 2009

Spring and spaced-repetition systems

What do spring and spaced-repetition systems (SRS) have in common? They both seem to be in the air this month.

There's been a lot of buzz in the language-learning blogosphere about SRS lately. See Mastering Mandarin parts 1, 2, and 3, Ramses on Free Technology for Teachers, Tower of Confusion, Flashcard Aficionado, and The Linguist on Language.

After the jump, a few highlights.

Read more...For the uninitiated, here's a brief definition, courtesy of Wikipedia:
Spaced repetition is a learning technique in which increasing intervals of time are used between subsequent reviews, rather than studied frequently for a short time...
Although from last year, it's worth pointing out that All Japanese All the Time has a pretty in-depth series on how to make the best use of SRS.

The consensus appears to be that the big players in SRS software are SuperMemo, Mnemosyne, and Anki. Mastering Mandarin has a brief overview of these, Cunning Linguist has a slightly longer one, and Nihongo Pera Pera has one that's much more detailed. I'd fault them all for leaving out Smart.fm, f.k.a. iKnow.co.jp.

Everyone seems to love their SRS, but Steve Kaufman goes against the grain, arguing that SRS might not be the most efficient way to learn, at least in part because it lacks "resonance".

If you're interested in implementing an SRS system in a classroom, Ramses of Spanish Only has some useful ideas for you.

Labels: