A New Method of Learning [Experimental]
Tuesday, June 30, 2015 at 19:02 I don’t know how many of my readers have used Spaced Repetition System (SRS) software to learn facts. They are most commonly used for foreign language vocabulary, but can be used for any type of fact that you wish to learn.
Personally I have always found them to be quite effective, but they suffer from some severe negatives which in the end have always lead me to abandon them after a period. Because they put the emphasis on the facts which you are having difficulty learning, you tend to end up with a huge number of difficult words which you have to plough through each day. The list tends to get longer and longer until, if you are not careful, you find yourself ploughing through vocabularly at almost every spare waking moment.
That’s frankly not the way I want to spend my life.
So how about a gentler and easier method which is even more effective?
You will probably think the method I’m going to propose is crazy, but I’m finding it has worked very well so far. I haven’t been doing it for very long, but I’d be interested in the results that other people get if they are bold enough to try it out too on an experimental basis.
Like all my systems I’ve designed it for paper and pen. I suggest if you want to try it that you use paper and pen at first, and then only if you find it works start worrying about how to make an electronic version.
For the purposes of the explanation I am assuming that you are an English speaker wanting to learn French vocabulary.
The system is based on spaced repetition, but with the difference that all the vocabulary items are revised at each interval.
The intervals are:
On the day of entry
The following day
One week later
One month later
One year later
I use a loose-leaf binder with a sheet for each day’s vocabulary. All I have to do when I’ve finished revising is put the date of the next revision at the top of the sheet and re-file it so all the sheets are in date order.
THE PROCEDURE
The First Day
On today’s sheet collect vocabulary as you come across it in the traditional two-column format. That is to say, French in the left-hand column and English in the right-hand column. Make no attempt to learn it until you have finished collecting it for the day.
Then you go through two phases: 1) pre-learning 2) learning.
Phase 1. Pre-learning
Cover up the right-hand column (the English) and test yourself on the meaning of the French words in the left-hand column. DO NOT CHECK YOUR ANSWERS. If you can’t get any answer, just go on to the next word. Then cover up the left-hand column and test yourself whether you know the French for the English words in the right-hand column. DO NOT CHECK YOUR ANSWERS.
Phase 2. Learning
Do exactly the same, but this time move the covering card down after attempting to answer each question so you can see whether you got it right. Do it both ways as in Phase 1. That’s all. You only do it once. Don’t repeat it, regardless of how many you got wrong.
This Phase 2 on the first day is the only time in the entire process that you check your answers.
Subsequent Revisions
These are all carried out in the same way as Phase 1 on the first day. In other words you test yourself without checking the answers.
CONCLUSION
Although the process may sound crazy, it is in accordance with the most recent findings on how we learn. A pre-learning test increases learning ability. Not checking one’s answers makes the brain work harder so that it remembers better on subsequent tests.
Reader Comments (22)
Thank you, Mark! Even if it should turn out, that it doesn't work: this sounds like fun!
On the subject of learning languages, do you know the Birkenbihl method (named after Vera F. Birkenbihl, the German lady who developed it)? I'm intending to try it out for starting to learn Italian. You can find an English summary here:
http://www.tigweb.org/action-tools/projects/download/16947/BirkenbihlApproach.pdf
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3990058/
<< I recently read the book "How we learn" by Benedict Carey. >>
I designed this method immediately after finishing that book.
<< do you know the Birkenbihl method >>
I have come across it, yes. But I've never used it in its entirety. I use the chorus method a lot. I don't find "wording" as it's called (having the word-for-word translation) necessary for Germanic or Romance languages since I'm pretty aware of how they work, but it's vital for languages like Chinese or Homeric Greek!
<< Seems to make use of the incubation effect? >>
Very deliberately so, yes.
<< Doesn't this have a big disadvantage in that it's easy to misremember the meaning of a word and not realise you've done so? >>
No, the fact that your mind has misremembered a word on one occasion improves the chances that it will remember it correctly on subsequent occasions. Not checking the answers makes your memory work harder.
And don't forget that you are testing your vocabulary both ways (i.e. French to English, and English to French).
In any case it's best to try it out, rather than theorize about it.
Spaced repetition works well, but it does take time and you have to motivate yourself to do it. At one point, when I had a lot of free time, in less than a year (3 months is easily doable though) I learned how to write the 2000+ commonly used Japanese characters on http://kanji.koohii.com/ which combines the book they mention (a way to create mnemonic stories to remember how to write the characters), and the Leitner box system https://en.wikipedia.org/wiki/Leitner_system , and a way for the users to share good mnemonic stories with each other.
Mark's right; it is time consuming, and getting the motivation to do your designated repetitions for the day is pretty hard. Even if you only add 20 items to the system each day, you still end up with 300+ items to review. (With Mark's system, if you added 20 items a day, you'd still be reviewing 100 items a day after a year.)
Another example: Anki is some good spaced repetition software. It has a more complicated algorithm than Lietner where it schedules cards based on how hard they have been for you. Unlike the kanji site, it's for anything you want to learn. You can also download decks of cards that are pre-made.
Mark's method above appears to be a Leitner system, except that you never fail an item back to box 1. (Plus the specific way to pre-learn, learn, and review, of course.)
Based on what Mark said about language learning before, I assume you would also be learning the language by reading (for example reading the original and the translation side by side, as Mark does), listening, etc. so that you encounter what you're learning often.
I never did actually learn Japanese. Marc (from Brussels)'s link to the Birkenbihl Approach looks interesting too, and you could do that alongside Mark's vocab review.
All this taken together gives me an idea. (Using Spanish in my example.) If I have a book in English and Spanish, and the Spanish audiobook (I just grabbed "The New One Minute Manager" in all three versions; I like the qualify of the narrator on the audio sample on Audible, and it sounds interesting enough to read, but still short enough to handle), I could use those to:
Step 1: understand each word in context using the Spanish edition and English edition side by side. Add new words to today's list using Mark's system above!
Step 2: Write (type) a word-by-word translation (just the English words in the Spanish order). Then use this to do the active listening while reading the English word-by-word version. (I suspect this may be too much effort, and would be what causes the system to fail. Only one way to find out.)
Step 3: listen to the same section of audio on repeat while I do other things (I'm thinking very short sections). Maybe the length can be determined by when you hit 20 new vocab words.
Step 3 note for iPhone (there's gotta be a better way): It appears that in order to make this happen, I can play the audio from my computer (Audible website since that's where I bought it), use the Voice Memos built in app on my iPhone to record (with the phone by the speaker), use the app to trim the audio so it starts and stops where I want it to (if desired), save it (give it a name so you can find it easily, like "aaaaaa"), sync the phone to iTunes, then pull up the voice memo in iTunes (under music, playlists, voice memos), right click it and choose "get info", change the Genre to anything besides Voice Memos, e.g. Weather (you can leave the album name as Voice Memos), click Options and change the file type from "Voice Memos" to "Music", sync again. On the phone, find your song (aaaaaa) or look under the Voice Memos albums for it. Play and set it to "Repeat Song".
Step 4: Review vocab (from Mark's system). Optional steps depending on interest level. I'd probably try to do shadowing to practice speaking. Reading along with the Spanish version while listening to the audio for reading, and/or re-reading the parts I have already covered. Not sure if I'm interested in writing at this point.
A very good player for shadowing is WorkAudioBook.
http://www.povalyaev.com/WorkAudioBook/How-to-use%28Windows%29.aspx
The "no checking" rule feels at least for my mind like the spaced repetition version for grown-ups. High risk of failing and serious evaluation instead of boring routine and all too tiny little steps.
No tweaks needed so far.
My actual start day was July 1st.
I haven't as yet tried it with something more difficult like Chinese hanzi or Japanese kanji/kana, but I can't see any reason why it shouldn't work with them.
I might try it with something that always defeated me with SRS - remembering the USA State Capitals!
That was the system I used to learn French before we moved to France 12 years ago. Only that I wrote myself a small program that worked accordingly and I had to type in my answers, just because it was more fun this way. Took me about 15 minutes per day over two years to become fluent with the basics. I consider the Leitner system the best method to learn any bunch of stuff you have to hammer into your brain, and I only regret that I didn't knew it when I was in school. The ways you "learn" things in school are, in retrospect, those that are the least efficient.
The whole point is that not remembering the answer makes your memory work harder in order to remember it the next time. So the answer to your question is ”nothing”.
Something I think I will try out.
By the way, the original book "So lernt man Lernen: Der Weg zum Erfolg" is well worth a read in my opinion (if you can read basic German). Sebastian Leitner was a science journalist and the book is pitched at around the difficulty level of the average newspaper or magazine article. You can find used copies on amazon.de for a few euros. It covers a lot more ground than spaced repetition using a box of paper slips, including many useful hints.
Apart, perhaps, from a decent course on time management, Leitner's book stands right at the top of the list of "things I wish I knew when I was a schoolboy."
If you run across a word and you aren't sure if it's already on your list from a previous day, would you add it to today's list?
Also, I always wondered if your series of No-List systems was inspired or influenced by your experiments with this learning method. There seems to be a key similarity in the way each system engages your mind by not allowing you to rely on crutches. And this engagement exercises your memory and "wakes it up". And each system has the counter-intuitive result that we tend to remember things BETTER this way, rather than making long comprehensive lists of things that need exhaustive review.
If you can't remember if a word is on your list from a previous day, then you shouldn't either check whether it is or add it to today's list. That will force your memory to work harder to remember whether you put it on the list. Same principle as the learning method itself in fact.
It's possible that No List was influenced by this, but not consciously.
J'essaie un nouveau système de Mark pour enfin apprendre le francais correct.