Sunday, December 30, 2012

Reverse-engineering Pimsleur to enhance short-term memory stability

For a mental impression to become a long-term memory, it goes through the following life cycle:
Momentary mental impression BECOMES short-term memory BECOMES stable short-term memory BECOMES (over time) long-term memory.

For the sake of outlining the argument, here is the rough life-span of each memory "species":

Momentary impressions: a few seconds
Short-term: a few minutes
Stable short-term: a few hours/a few days
Long-term: many weeks, months, years, etc.

Momentary to Short-Term
Daily we have many mental impressions, not all of them being valuable and worth "capturing." But ANYTHING we wish to remember starts out as a momentary impression on the mind; whether or not it goes any further is up to the individual. When you are introduced to a cute member of the opposite sex, you mentally treat their name different than someone you are not really concerned with. Rather than quickly discarding it, you think about the name and link it with that person. Because you desire to know more about that person, you concentrate and put forth effort to remember their name. Thus, a momentary mental impression has now become a short-term memory.

Short-Term to Stable Short-Term
Short-term memories become more stable as you review them over a few hours. Often you can tell if a memory has gone from "short-term" to "stable short-term" by the end of the day. If you have tried learning a second language, you have no doubt found that certain words or phrases you tried to learn during a study session stuck while others did not. I believe this is because if a short-term memory does not become "stable," it quickly falls the way of the "momentary mental impression" route and perishes with the other static we dismiss during the day.

Stable Short-Term to Long-Term
This transformation is what SuperMemo, Anki, etc. adequately take care of.

My question is this: Just as there is an algorithm to ensure long-term stability of memories, is there also an algorithm that turns a short-term memory into a stable one? If you could do this efficiently and easily, learning languages (Among many other things) would become easier and more motivating. I think I know what might aid us in finding this out: Pimsleur.

Having recently finished Pimsleur Hindi (Comprehensive), I noticed something: while I do have a couple of complaints about the selection of vocabulary that was taught, I was surprised at how easily the taught vocabulary and phrases were memorized. Regardless of how difficult ANYTHING sounded when I initially heard the opening conversation, the Pimsleur audio course ensured that my short-term impression of any word or phrase became a stable short-term impression. As I went over each lesson, I put the new phrases into SuperMemo and have had no problems recalling them since.

I began to think about this while driving to and from work for a few days: It isn't an accident that this happened. At the Pimsleur language company there is not a person working there that speaks dozens of languages and uses his mysterious ways to teach you a language. At Pimsleur they don't care what language they are teaching you; they have an algorithm, and when they want to teach a new language they simply plug something different into the same algorithm.

Granted, they have to make a good selection of material (Which can vary from language to language), but once the material is selected, it's just a matter of listening to the lessons as you go about your day. This process seemed mysterious to me at first, but like anything else, Pimsleur's language teaching process must have a solid, scientific basis. Once that basis is figured out, you can use it for other things.

In an attempt to figure out their process, I've made a transcript of Pimsleur Japanese lesson 1, and I created a time stamp each time a new phrase or word is introduced, and each time the student is quizzed over something. The transcript is in the previous post. The time stamps stand for when the speaker concludes his statement, be it a quiz or an introduction to a new phrase or vocabulary word. Also, because many of the words you learn in the lessons are later used in phrases, there is some degree of overlap which I tried to keep straight.

I think that using this transcript, we can figure out the algorithm the Pimsleur company uses to make information "sticky" and memorable. Please give me suggestions, or fool around with this data yourself; if this yields positive results, it has the potential to be very useful to us language learners!

Pimsleur Japanese I Lesson 1 Transcript


Conversation 1:
すみません、英語がわかりますか?
いいえ、わかりません。
日本語がわかりますか?
はい、すこしわかります。
あなたはアメリカ人ですか?
はい、私はアメリカ人です。

Vocab 1: すみません。
Vocab 2: 英語(が)
Vocab 3: わかります(か)
Phrase 1: 英語がわかりますか?
Vocab 4: いいえ
Vocab 5: わかりません
Phrase 2: いいえ、わかりません
Vocab 6: 日本語(が)
Phrase 3: 日本語がわかります。
Phrase 4: 日本語がわかりません。
Vocab 7: 少し
Phrase 5: 少しわかります
Phrase 6: 日本語が少しわかります
Vocab 8: アメリカ人
Vocab 9: あなたは
Vocab 10: です
Phrase 7: あなたはアメリカ人です。
Phrase 8: あなたはアメリカ人ですか?
Phrase 9: 私はアメリカ人です。
Vocab 11: はい

0:10 conversation 1 begins
0:39 vocab 1 introduced
0:56 vocab 1 repeated
1:21 vocab 1 quiz
1:36 vocab 2 introduced 
1:49 vocab 2 quiz
1:50 vocab 2 & phrase 1 quiz
2:17 vocab 2 & phrase 1 quiz
2:32 vocab 1 quiz
2:48 vocab 2 quiz
3:02 vocab 3 introduced
3:35 vocab 3 quiz
3:50 phrase 1 quiz
4:10 vocab 3 & phrase 1 quiz
4:22 vocab 1 quiz
4:29 vocab 3 & phrase 1 quiz
4:39 vocab 2 & phrase 1 quiz
4:56 phrase 1 introduced
5:20 phrase 1 quiz
5:32 vocab 4 introduced
5:42 vocab 4 quiz
5:48 vocab 3 quiz
6:00 phrase 1 quiz
6:17 vocab 4 quiz
6:26 phrase 1 quiz
6:37 vocab 3 quiz
6:47 phrase 1 quiz
6:57 vocab 3 quiz
7:04 vocab 3 quiz
7:12 vocab 1 quiz
7:21 phrase 1 quiz
7:37 vocab 4 quiz
7:51 vocab 5 introduced
8:16 vocab 5 quiz
8:28 vocab 5 repeated
8:55 vocab 5 quiz
9:04 vocab 2 quiz
9:12 vocab 5 & vocab 2 quiz
9:28 vocab 3 quiz
9:38 phrase 1 quiz
9:54 vocab 5 quiz
10:02 phrase 2 quiz
10:19 vocab 5 & vocab 4 quiz
10:31 vocab 3 quiz
10:39 phrase 2 quiz
10:49 phrase 1 & vocab 5 quiz
11:02 vocab 6 introduced
11:27 vocab 6 quiz
11:39 vocab 2 quiz
11:39 vocab 6 quiz
12:09 vocab 3 quiz
12:19 phrase 3 introduction
12:37 phrase 3 quiz
12:49 phrase 4 introduction
13:10 phrase 3 quiz
13:14 vocab 1 quiz
13:27 phrase 3 quiz
13:35 vocab 7 introduction
13:53 vocab 7 quiz
14:04 phrase 5 introduction
14:25 phrase 3 quiz
14:39 vocab 7 quiz
14:46 phrase 6 introduced
15:12 vocab 3 quiz
15:19 phrase 1 quiz
15:30 phrase 2 quiz
15:41 phrase 6 quiz
15:59 vocab 2 quiz 
16:15 phrase 3 quiz
16:33 phrase 2 quiz
16:45 phrase 3 quiz
16:56 phrase 6 quiz
17:24 vocab 8 introduction
17:46 vocab 8 quiz
17:57 vocab 9 introduction
18:18 vocab 9 quiz
18:29 vocab 10 introduction
18:45 vocab 10 quiz
18:55 vocab 9 quiz
19:10 phrase 7 introduction
19:51 phrase 8 introduction
20:15 vocab 11 introduction
20:45 vocab 10 quiz
20:58 vocab 11 quiz
21:11 phrase 7 quiz
21:21 phrase 9 quiz
21:40 phrase 7 quiz
21:53 vocab 1 quiz
22:02 phrase 3 quiz
22:14 phrase 1 quiz
22:26 phrase 8 quiz
22:50 vocab 11 introduced
22:55 vocab 11 quiz
23:01 phrase 3 quiz
23:19 vocab 11 & vocab 7 quiz
23:33 phrase 6 quiz
23:45 phrase 8 quiz
23:58 vocab 11 & phrase 9 quiz
24:45 conversation 1
25:04 vocab 1 quiz
25:19 vocab 3 quiz
25:33 phrase 3 quiz
25:52 phrase 1 quiz
26:11 vocab 5 quiz
26:25 phrase 3 quiz
26:43 phrase 5 quiz
26:58 phrase 6 quiz
27:11 phrase 8 quiz
27:30 vocab 11 quiz
27:36 phrase 9 quiz
27:57 vocab 3 quiz
28:20 vocab 7 quiz

Thursday, December 27, 2012

Seek Intellectual Stimulation, not Emotional Stiumulation

One principle that I try to keep in mind when seeking mental input, whether it be reading an article/book, watching a video, or going somewhere: am I seeking emotional or intellectual stimulation?

A great deal of emotionally stimulating material  exists today. Youtube videos, web pages, blog entries, games, etc. While a number of things out there are intellectually stimulating, a majority of them appeal merely to (usually shallow) emotions. Even "news" that merely gives the details of celebrities' lives, or focuses on the detailed aspects of a murder or extramarital affair appeals to emotion but disguises itself as intellectual appeal. Beware of being tricked in this way.

Just as a diet of mostly sugar is unhealthy for the body, a constant diet of mental "junk food" is bad for the mind. Junk food is not a problem, as long as it is eaten in moderation.

If someone has unhealthy eating habits, they must first recognize it and then monitor what they eat. If you (As I did in the past) realize that your mental diet of input is mostly junk food that doesn't appeal to the intellect, recognize that you should change something and then monitor everything that goes into your mind. Do you really want to listen to that pop music station? Maybe instead you could listen to the news, or a download a lecture onto your phone, or maybe try to learn Spanish in your car...?

Like eating vegetables, adjusting what you mentally "eat" can be challenging at first, but over time you will prefer it over junk food.

Wednesday, December 19, 2012

Institutions ≠ Education (Not always)



Interesting video sent to me by a friend. I'm not anti-school, but while education is good, being institutionalized doesn't always mean one is educated. Balance is always important; use what opportunities come your way, but never lose sight of your long-term goal.

Tuesday, December 18, 2012

Evernote: Online Incremental Reading without SRS

The other day I was looking for incremental reading alternatives for the iPhone or iPad, and I came across Evernote, a platform that lets you archive information and create notes that are searchable for later use. While the purposes of Evernote and Incremental Reading are different, they do overlap quite a bit.

Evernote helps you create a "scrapbook" of things relevant and interesting to the user. Phone numbers, interesting quotes, ideas you read about, etc. are all placed on digital "index cards" that become searchable for later use. You can also create to-do lists based on that information.


The difference between Evernote and Incremental Reading is that with Incremental Reading, all of your "index cards" are placed on a conveyor belt that forces you to review and evaluate the information, and choose to do something with it (Remember it, delay it, delete it, read as much as you can/want). The end result of Evernote seems to be accumulating a useful collection of searchable information, Incremental Reading creates a useful collection of remembered information.

To put it another way, Incremental Reading is an internal, offline version of Evernote where notable information is eventually turned into flashcards.

But unlike incremental reading, Evernote can be used on an iPhone, iPad, the internet, etc.

So here has been my recent workflow with using Evernote with Incremental Reading on my iPhone:
1. Find interesting article.
2. Read in Instapaper.
3. When I find an interesting piece of information, copy into Evernote.
4. Once a week, copy things from Evernote into SuperMemo via Incremental Reading and process the information in the days to come.

I'm not sure why I never thought of doing this in the past, but it almost completely takes care of my need for portable Incremental Reading. There are not very many steps and it has been working out wonderfully. Maybe it will be helpful for you, too.

Thursday, December 6, 2012

Using Pimsleur (Or any audio language course) with SuperMemo

Language acquisition is not a simple, cut-and-dry matter. A great deal of blood and sweat is required, and no single language course is a "silver-bullet" for becoming fluent in a language. Each language course has strong points and shortcomings. Pimsleur is a language course that has strong points and also shortcomings, but I believe that the strong points outweigh those shortcomings, especially when you are starting out with a language.

When I first started learning Japanese (And before I started using SuperMemo), I listened to Pimsleur Japanese I, doing one lesson a day. Although I read reviews criticizing Pimsleur's Japanese course, I found it very useful in acquiring a few phrases and getting rid of my American accent. After I was using Pimsleur for a couple of weeks, a Japanese friend (Who is very strict at correcting my Japanese) called me, and I picked up the phone and said "hello" in Japanese, and my friend said "I thought I accidentally called a Japanese person." I say this not to toot my own horn, but to show that Pimsleur language courses have at least SOME bit of usefulness.

If you've never heard of Pimsleur, they are a company that makes (Expensive) audio-only language courses. What makes them effective is the "fill-in-the-blank" structure, which requires that you produce responses frequently AND the smart use of graduated interval recall to aid in short and long-term retention (It helps short-term retention like nothing else I've ever used, personally).

For years I have not been using Pimsleur, but after deciding to learn Hindi, I have started using Pimsleur Hindi. This time, I am going through the course with the aid of SuperMemo.

This has been my routine, and it has been very useful for me:

1. Listen to a lesson of Pimsleur with SuperMemo turned on.
2. If I hear a new phrase or word used, pause the lesson and make a flashcard of that phrase.
3. If I have trouble producing a phrase they ask for, pause the lesson and make a flashcard of that phrase.
4. Keep doing this until lesson is finished.

Two things I've noticed so far:
This whole process takes about 35-40 minutes to complete.
The Pimsleur lessons take care of short-term retention while SuperMemo can take care of the long term retention. If you do more than one lesson per day relying ONLY on Pimsleur, you risk messing with their spaced repetition algorithm. Using SuperMemo ALONGSIDE Pimsleur removes this barrier, allowing you to complete more than one lesson per day.
You can also go days or weeks without doing a lesson without losing progress. Typically Pimsleur courses are best done one lesson per day, every day. Again, SuperMemo acts as the safety net you would otherwise be without.

I'll create another post if I find anything else notable as I go along this course. I suspect I will be finished with Pimsleur Hindi in the next couple of weeks, mainly due to various demands of my time.

Wednesday, November 14, 2012

Flashcard selection based on core values and abstract patterns


Committing things to memory can be beneficial or detrimental to one's life. For example, one could memorize the entire phonebook and waste a great deal of time. On the opposite end, you could memorize how to do the Heimlich Maneuver and save someone's life. So one important thing is the SELECTION.

So the next question is: Which knowledge is most beneficial? It depends on a number of things, but the most important one is this: What do you want to do with your life? Do you want to understand Indian culture? If so, learning the history of bubble gum might not be a good idea. If you were trying to compete in the bubble gum industry, though, learning the history of bubble gum seems much more useful now. So much depends on your CORE PRIORITIES.

This is what guides my overall intellectual life: "Learn something about everything and everything about something." It is my goal to learn a great deal about a few subjects, and learn a bit about many other subjects. The few subjects that I want to really know well are (Broadly speaking): Foreign languages (And language acquisition), history and neuroscience/social psychology. When it comes to these subjects, I crave new knowledge like water.

Other subjects are like soft drinks; I drink them every now and then, but they don't receive as much attention as my core subjects.

So when you look at something you could memorize (Math equations, jokes, quotes from books), the only one that can make a proper judgment call on this is yourself; "Do I want this in my mind forever?" If so, make a flashcard out of it. It will only require two or three minutes in your lifetime, so the long-term "damage" of committing a few extra seemingly trivial things to memory is very slim (As long as you don't commit a great deal of trivial things to memory).

Certain information might not seem very useful, but it becomes EXTREMELY valuable later. What is the value of having JUST the right joke or anecdote to say at the dinner table with friends? What about having JUST the right romantic quote to say to a pretty girl? Or a succinct metaphor that contributes to a discussion on a subject? While many factual flashcards have a definite value (Foreign language vocab words, test material from medical textbooks, etc.), certain things are more abstract but are also very valuable; saying the RIGHT THING at the RIGHT TIME is a great feeling for yourself and others.

Success in most fields boils down to recognizing and properly anticipating patterns. If you know what kind of situations you are going to encounter, you should commit to memory enough patterns of knowledge relevant to that field. The more abstract those patterns, the more creative you must be to articulate and anticipate them. If you can hit the sweet spot by anticipating the right pattern at the right time, good things can happen on a small or large scale; I am convinced that most breakthroughs in various fields (Science, computer design, etc.) is a result of this process. Like Wayne Gretzky said, "I skate to where the puck is going to be, not where it has been." If you can do this in whatever your field of study, the time investment in SuperMemo has more than paid off.

This serves to reinforce that SuperMemo (Or any other flashcard software) is only a shovel to be used not for the sake of using it (To dig holes), but to find buried treasures of knowledge beneath the many patterns we encounter in life.

Thursday, October 11, 2012

Incremental Reading: What is of Value?

Imagine you are an archaeologist exploring a cave that potentially has valuable fragments of a lost civilization. As you explore the cave, you find many rocks of little value (If any). However, as you find artifacts that might be relevant to your field of work, you pause, examine the artifacts, place them in bags, label them for future study, etc.

Incremental reading is difficult to describe; like traditional reading with books it involves exploring a line of thought, but unlike traditional, linear reading, you are encouraged to examine and constantly assign value to the material under consideration, much like an archaeologist exploring a cave. What information and articles are valuable can be very subjective, as people have many different motivations, hobbies, fields in which they are ignorant, etc. (After all, to a geologist, the same "valueless" rocks that were discarded by one might potentially be more useful than a valuable artifact at that same site!).

As you examine any text, you need to first ask yourself, "What information is valuable to me?" To do this you must first have a set of core values (Stated or otherwise). To illustrate, consider an article I just finished reading from Time magazine: "Red Truth Blue Truth."

What information is valuable to me? I have no political affiliation, I consider myself neutral when it comes to politics. While one might cut the top part of a weed every few years by electing a new face for the government, the roots of corruption, greed, etc. stay the same. Thus, while one political party might advocate a policy I feel is good, the corruption runs so deep on every level, the decaying institution is either beyond reform or very close to it. I'm not advocating that you adopt this viewpoint, I'm simply stating my core values that came into play when evaluating this article.

If I was a Romney supporter, I might be more inclined to look for knowledge and facts that would be detrimental to Obama's administration and something I could use to win an argument with a co-worker. If I was an Obama supporter, I might be just as inclined to look for things detrimental Romney's policies and track record. Because I don't have a stake in either side, going into the article I was looking for mainly historical tidbits and a more complete picture of the tattered state of political discourse.

(It is also good to remember that many statistics are greatly subject to change. To memorize them would be just as valuable to memorize the stock price of a company on any given day.)

I read the article, one paragraph at a time. If I didn't find anything enlightening, surprising, etc. in one of those paragraphs, I would delete that paragraph and move on to the next one. If I found something interesting in that paragraph, I would analyze what was valuable in that paragraph and try to condense each point into a single sentence (Very important!). After doing this, I would extract that sentence, delete the paragraph and move on. I kept this up until I finished reading the article.

Among the interesting points I found in the article were:
-Lies and mud-slinging is as old as US politics itself; it goes back all the way to the very first election involving John Adams and Thomas Jefferson.
-The quote "We’re not going to let our campaign be dictated by fact checkers," is telling about politics in general.
-Those that subscribe to a political party are more likely to dismiss or downplay the lies said or endorsed by their own party (Basically "fanboy-ism," which works with technology, movie franchises, etc.).
-I never knew Ronald Reagan claimed that 80% of pollution came from trees and vegetation.

Other than the above points (Along with a few more), the rest of the article contained nothing extremely surprising or enlightening. Still, 6 or 7 good flashcards came from an article that took 10-15 minutes to read, so I'm satisfied with the result.

Tuesday, October 9, 2012

Incremental Reading Principle: Priority


(This is in response to a question asked in the comments section of the previous post)

I had been using Incremental Reading on-and-off over the past few years, but only recently do I think I have hit my stride when it comes to Incremental Reading. I am having fun reading more than a dozen recreational books, articles from various journals (Time, New Yorker, etc.), long blog posts, articles on TV Tropes (It's the only way I can read that site), reviews of movies, video games, etc. Obviously different texts receive different treatment. A fluff blog entry isn't analysed to the degree a book on string theory is.

Here is a rough way I mentally classify articles/books/etc. that I read:

HIGH PRIORITY (Methodical)
-Read every 1-5 days
-Create flashcards as soon as I extract noteworthy information
-Do not advance the article until I understand every piece of  information
-When I encounter difficult parts, endeavor to understand them ASAP

MEDIUM PRIORITY (Casual)
-Read every 10-30 days
-Extract topics, make flashcards out of them whenever I get to it
-Skip difficult parts that require further study

LOW PRIORITY (Passive)
-Read a paragraph or two every 50-150 days
-Extract topics, but I don't really care about them unless I'm getting rid of old topics that clutter my database

So when reading a high priority article, I will methodically break down every paragraph and sentence that contains something noteworthy, creating flashcards as soon as I think I can create one adequately. If I find a piece of information that I do not understand but understanding it is necessary for progressing through the article, I put the article on "pause" (Delay it for a few days) while I endeavor to understand that piece of information, doing the same steps as mentioned above. Once I am done, I resume reading the original article.

When reading a medium priority article, I will casually read it and freely jump around from one section to another as I see fit. I don't really care about deeply knowing every thought contained. If, in order to understand the article, I must first consult a different source to grasp a concept alluded to in the article, I may or may not confront that gap of knowledge. If I choose to, I do so "when I get around to it."

When reading a low priority article, I am almost completely passive about the information I'm reading. Aside from a few interesting things that the article might potentially contain, I don't care very much whether the article lives or dies. It might be delayed for many months or years (Or deleted), and it doesn't take up any "mental RAM."

Some articles might start out as a 'medium priority' article, but as I read it I realize that it has very valuable information contained within, and it goes to 'high priority' for a few days. After I extract the most useful bits of knowledge from the article (If I do not delete it), it might go back to 'medium' or even 'low priority' for a few months until I put it out of its misery.

As you can see, how you incrementally read articles changes as your priorities change. What you consider valuable one month might not seem so valuable later. Although I never explicitly say to myself "This article is 'high priority' now," my thirst for the knowledge contained in the article compels me to gobble more and more of it, while an article that doesn't excite my knowledge apatite are delayed (and eventually discarded if they don't prove their worth).

As the SuperMemo web site says, Incremental Reading is more of a reading management technique than anything else. And while it has taken me a while to get used to it, it is one of the most compelling features of SuperMemo. Once you hit your stride, you can quite comfortably read hundreds or thousands of articles at the same time, although I didn't think that it was possible at first.

Monday, October 8, 2012

Reading books incrementally

Reading Books Incrementally

I love reading books, but I enjoy incremental reading too much to read books, blog entries, etc. in the linear fashion they were intended to be consumed. The negative aspects of linear reading will be covered in another post soon. Until then, to make it easier to read books using Supermemo, I have found this to be particularly helpful:
There are many sites that allow you to create text documents of .epub and .mobi files. Here is one that I have been using: Whoosh.


After I first acquire the epub or mobi file and convert it to a .txt file, I copy and paste 10 or so pages into Supermemo (Too much text slows Supermemo down). I use the "picture article" template so I can put a picture of the cover of the book on the opposite side of the page (I have always been fond of book covers, album covers, etc.). I delete the text that has no value to me, highlight and extract sentences or paragraphs that might contain a flashcard, and keep doing this until I read the end of the text I copied over. After this, I copy another 10 or so pages, and then continue on.

Right now I am reading at least 10 books (Two of which are fiction), and I have thoroughly enjoyed this way of reading. If you haven't tried it, it is a very fun way to read scientific and historical books (Which is what I enjoy reading the most).


To summarize:
-Incremental reading is superior and more fun than traditional reading.

-Convert an acquired epub or mobi file into .txt.

-Only copy 10-15 pages of text at a time to prevent slowing Supermemo down.

Sunday, September 30, 2012

Reading Books Incrementally


I love reading books, but I enjoy incremental reading too much to read books, blog entries, etc. in the linear fashion they were intended to be consumed. The negative aspects of linear reading will be covered in another post soon. Until then, to make it easier to read books using Supermemo, I have found this to be particularly helpful:
There are many sites that allow you to create text documents of .epub and .mobi files. Here is one that I have been using: Whoosh.

After I first acquire the epub or mobi file and convert it to a .txt file, I copy and paste 10 or so pages into Supermemo (Too much text slows Supermemo down). I use the "picture article" template so I can put a picture of the cover of the book on the opposite side of the page (I have always been fond of book covers, album covers, etc.). I delete the text that has no value to me, highlight and extract sentences or paragraphs that might contain a flashcard, and keep doing this until I read the end of the text I copied over. After this, I copy another 10 or so pages, and then continue on.

Right now I am reading at least 10 books (Two of which are fiction), and I have thoroughly enjoyed this way of reading. If you haven't tried it, it is a very fun way to read scientific and historical books (Which is what I enjoy reading the most).


To summarize:
-Incremental reading is superior and more fun than traditional reading.

-Convert an acquired epub or mobi file into .txt.

-Only copy 10-15 pages of text at a time to prevent slowing Supermemo down.

Thursday, August 23, 2012

Misconception of Memorization



Proper memorization is a healthy part of learning. After all, memory is required to perform seemingly basic tasks. Sadly, today all memorization is portrayed as a mindless, soul-numbing experience that has an adverse effect on creativity and intelligence as a whole. But as the Supermemo method shows us, healthy memorization can help one become more creative and less dependent on antiquated teaching and testing methodology.

Thursday, August 2, 2012

Beyond Mascots & Logos

Even if you do not think of yourself as a "visual learner," we humans tend to think in pictures.

Because of this, (just about) every company creates a mascot or logo to represent their company and/or a good or service they offer.

The more you see a picture of a mascot, logo, etc., the more familiar you subconsciously become with the company they represent.

Companies do not do this to exercise your ability of image association, but to make money.

The more familiar you become with a company and their overall brand, the more likely you are to purchase a good or service they offer.

Thus logos, mascots and images in general have a very powerful associative effect on the mind. Everybody knows this; both the companies designing them and the people consuming them.

If it is acceptable to use mascots to link a sometimes abstract corporate identity with a tangible image for the sole purpose of making money, could not this same principle be used to make abstract information more tangible and appealing for the sole purpose of learning?

I personally think so. For the last couple of weeks I have been experimenting with using various images to make subjects more "sticky," and it's working.

As an exercise I have created "mascots" for countries, scientific principles, elements of the periodic table, major and minor wars, large-scale natural disasters, and different translations of various books. While some of these exercises have been fruitless, the great majority of applications are yielding positive results. Having an image associated with something abstract makes it easy to differentiate that information from a similar concept, person, etc.

I am still experimenting with assigning mascots to stuff, and I have a feeling this will be ongoing. Just know that it is working and its application is much more broad than I initially thought.

To conclude, mascots are useful in linking abstract things to tangible things. Spaced repetition software makes forming and maintaining this link very easy. So, the next time you have trouble keeping two similarly related ideas separate, assign two cool-looking characters to represent those ideas, and use Supermemo (Or whatever SRS you use) to make sure those two characters are never confused.

I've found the "pixiv" Japanese artist community and conceptart.org to be an almost bottomless pit of original artwork and characters. "Pixiv" seems to be a Japanese version of Deviantart. While there are many original characters to work with, you have to deal with and skip over the anime tropes that are too wahjah for your tastes.

Friday, July 6, 2012

Making Historical Figures (And Others By Extension) More Exciting

I enjoy learning about history, but often I have a problem with keeping different historical figures separate in my mind. Plato, Aristotle and Socrates are three different philosophers, but for each of them I imagine a guy with a beard reading a book. Once you learn more about their contributions to history, you can help make them a bit more distinct. For example, instead of Plato being a bearded guy, you can say "Plato is the bearded guy that helped establish the teaching of the trinity among those professing Christianity." While this can be "linked" to Plato, it still doesn't give Plato any visual distinction over the other two philosophers, you only get a bit more information about him.


To fix this, you could go to Google Image search and find various paintings, statues and other depictions of Plato, Aristotle and Socrates, but you are likely to find pictures of old guys with beards with only a mild amount of variation. While this might do the trick, my goal is to keep them visually separate in my mind. If you wish to remember how Plato, Aristotle and Socrates were depicted, you could create flashcards saying "Who is this piece of art depicting? [...] Socrates." But if the goal is to keep them separate in your mind, there is something (I have found to be) even more effective to do:

About 5 weeks ago, I found a web site called ConceptArt.org. As the name suggests, it is a web site where talented people post concept art to be used in various movies, animated shorts, games, etc. While some of the concept art is used for an actual product, much of the art never goes beyond the conceptual period. Maybe the game or movie never gets enough momentum to see the light of day, or maybe the concept art is just a creative exercise. Regardless, the ConceptArt.org web site is a treasure trove of memorable looking characters, robots, landscapes, etc. Recently I have been going through their forums and saving pictures of anything remotely interesting I've found. I'm trying to organize them into different folders such as "male characters," "female characters," "robots," "villains," etc. As I encounter a historical figure that I don't know very well or cannot visualize easily, I go into one of those folders and find a piece of cool concept art and use it to represent that historical figure.

While it is certainly unrealistic to think of Plato as a cyberpunk samurai or video game protagonist, the purpose is to make him distinct in your mind. Associating some (even bizarre) sort of picture with someone will keep him/her distinct from other similar people. Also, if you prefer your mental image of someone to be more realistic, Flickr is a very good resource for simple pictures of people. Just search for "male" or "female" and organize them by date posted (You are more likely to get candid and "unremarkable" shots of people). Find a picture that you like and create a topic in Supermemo saying "Who is this guy? Plato." and look at it every day until you are sure you have established a solid enough mental "link" with the two. For me, it has taken anywhere from 3 to 7 days for any given picture-person image to successfully "link."

Some might raise to the objection "Shouldn't you imagine your own image of a person? Why do you have to be unoriginal and use someone else's work?" While creating your own image of someone is a possibility, I have recently begun viewing artwork, music, movies, etc. as potential mental "shortcuts" that I can remix as I see fit in my memory. Someone else has already captured a really neat idea (Picture, music, etc.) through many hours of effort and revision. When you're trying to use established concepts as mental anchors, however, the priority shifts; no longer is originality the most important thing. Instead one's focus becomes the efficient remixing of established concepts in the mind; the "cognitive extension" of the "remix culture" we find ourselves in.

Your past experiences and memories are an essential part (Some would argue the most essential) part of who you are. If you're going to efficiently keep information that makes you a better person (Whatever one considers "culture"), you might as well have fun while acquiring and storing that information by using other people's work.

In conclusion: That guy in red? That is my mental image for Plato. Why? Because it's useful and makes thinking more fun.

Tuesday, July 3, 2012

Making "Unsticky" Information Memorable (Aside From Mnemonics) - Pseudo Context Involving Novelty and Familiarity

I have found that often I will often get frustrated at information that isn't "sticky" enough to last one or two repetitions in Supermemo. If this ever happens, the first step should be to create a mnemonic, however basic or silly. But what if a mnemonic simply is not feasible? For example, what if you are trying to remember a certain point of foreign language grammar that doesn't easily conform to a mnemonic system? I think I found the answer: meaningful (pseudo)context.


When I wasn't studying Chinese very seriously, I had a hard time recalling certain grammar points. This was partly because I wasn't putting much effort into studying the language, but I was also unfamiliar with the language itself (Individual words and the "flow" of conversation). But a certain piece of grammar cemented itself in my head while I was reading the novel "After Dark" by Haruki Murakami. Although I stopped reading After Dark halfway through, I can tell that Murakami does a very good job at capturing a particular atmosphere and drawing you into it. The characters of After Dark felt like they had depth.


Later on in the book, a Japanese girl that speaks Chinese has a brief conversation with a Chinese girl. While many Chinese example sentences have slipped my mind over time, After Dark's brief Chinese conversation stuck with very little effort. In fact, those same sentences could be in a Chinese textbook and be just as forgettable as any other sentences. But once they are part of an interesting novel, those words suddenly stick in my mind. Why is this? Because I was emotionally and mentally invested in the characters that spoke; thus their words had added weight in my mind. They weren't just sentences, they were words spoken by two characters with a back story.


So how can you convert something forgettable and "unsticky" but important? After all, there will not always be a movie or book that directly addresses the thing you are trying to learn. You are left to create your own context. This is what I've been doing (With Chinese grammar or any unsticky information):


Step 1: Have a folder of assorted neat pictures (Photos/ paintings/ digital artwork/ etc.).
Step 2: Select one picture to go with that flashcard.
Step 3: Passively review that flashcard every day until the word-picture pairing feels familiar.
Step 4: Create a flashcard, making sure that you do not see the picture when being asked the question.
Step 5: Give the flashcard a few weeks to be reviewed, evaluate if the memory link is still there or not.


Why wait until you are given the answer to reveal the picture? If you saw the picture before you tried to recall the answer, there is a risk in you associating the picture with the correct answer rather than the context of the flashcard. That is one of the negative aspects of the Rosetta Stone language software. When sentences become complicated, you associate the answer with a picture rather than what you should be learning. While picture-answer flashcards are useful for nouns, for more complicated or abstract material it doesn't appear to be as effective.


What is the goal of repeatedly viewing a (seemingly) unrelated picture alongside unsticky information? Establishing context, "legit" or not. As you repeatedly view the word-picture pair, you might even create a plot surrounding the information in question.


One ancient proverb says "to the making of many books there is no end." Especially today, information is very cheap. But as Einstein said, "information is not knowledge." Having lots of information at your fingertips doesn't mean anything unless it is meaningful to the reader and is applied in one's life. Therefore, instead of procrastinating by saying "I can't learn this language until I buy this awesome "Sentence Pack ABC" that I've been told I must buy" or "I cannot learn about subject ABC until I go and spend money on a book about it," simply use one of the many free resources online and start dissecting it! Using pictures (And no doubt other media such as movies or music, although I have only tried pictures), information can (and should) be as interesting as it is important. If it doesn't seem interesting, brainwash yourself into thinking it is interesting!

Thursday, May 17, 2012

(Theory) The Key To Memory Priming - Decontextualized Passive Review

The reason I began to investigate this methodology is mainly for one reason: English is my native language, but I found it very difficult to learn new vocabulary in English. I was able to learn many foreign language words with little trouble, but not so with English. It is not due to a lack of understanding of  the English language, so why would these words not "stick"? I think I've figured out why, and how to make them (Or any other knowledge) more "sticky."


The following is pure theory; but even if I am a bit off base, I think I am onto something.

Regardless of how good you are at remembering numbers, if you are reading this blog (You have a computer, access to the internet, understand a bit of English), you likely know that the first three digits of pi are 3.14. Why do so many people know this? Obviously we are not all as equally adept at remembering numbers.

What is the answer? Through sheer passive repetition our minds have been primed to remember it. Before you had to remember it for a test, you likely encountered the number many, many times at school, among your friends, on TV, etc. By the time you were quizzed on what pi is, because of so many passive encounters with the number, your memory had been perfectly primed to recall it; regardless of how adept you are at remembering numbers.

Thus, if you want to remember something that you typically aren't good at remembering (A historical date, a number, foreign/native language word, etc.), how do you prime your memory to remember it?

Decontextualized Passive Review


What do I mean by decontextualized? Let's say you have a list of information you want to remember; a list of foreign-language vocabulary words or maybe a list of important years in Chinese history. You can scan down the list and look at each item, but the order that you are looking at them will never change. It would be better to write each individual piece of information on an index card and shuffle the index cards and THEN review them. As you review each item, they will not be sequential or in a predictable order. Thus, you cannot mentally "coast" through the information as easily as you might if it were in a sequence. Obviously you do not want to write each piece of information on an index card, but in Supermemo a Topic is a basically a digital index card that is reviewed passively.

Why should it be passive? Because while you might be able to connect two pieces of information and understand the significance of both (In 1912 the Republic of China was established), one aspect of the information is not very "sticky" for your brain (In this case, the year 1912). If you were to try and memorize the information, you will likely fail to recall it a number of times and it might become a "leech" (In Supermemo, a leech is a flashcard that you have failed to remember five times. It is statistically proven that it will drain your time). But as you (daily) passively see the same information over and over again (Without the stress of trying to "get it right"), very quickly the information becomes more "sticky" to your brain. Obviously using mnemonics will hasten this process a great deal.

Review? As opposed to what? Rather than giving the information a cursory glace, information you are priming your brain to remember requires deliberate purposeful concentration. This only requires a few seconds of your time, assuming the information is not exceedingly wordy (Which would be a violation of
the "20 Commandments of Knowledge Formulation").

I am not recommending "brute forcing" information without any forethought (Like writing a Japanese character over and over again as opposed to using the Heisig system of "divide-and-conquer-through-mnemonics"). What I have been trying for the past month or so has been this: after taking difficult information and making the wording as simple as possible (English vocabulary, Russian and Hindi alphabets, creating mnemonic systems themselves, etc.), I simply look at that chunk of information every day until it no longer feels "foreign" to me. Once it feels this way, I turn it into a flashcard and leave the rest to SuperMemo.

Before I began doing this, certain flashcards would be marked as "wrong" repeatedly and never "get off of the ground" in my mind. Some difficult Japanese characters, many English words, and a number of mnemonic systems (Specifically a system for remembering numbers) were easy to understand but couldn't survive the initial few days of the first review. But once I began daily passively reviewing bits of knowledge, I've been able to learn just about anything I've thrown at it (Mnemonic systems, English words, various foreign language alphabets which I am unfamiliar, just about anything I can think of).

How do you do this? I do not know of a solution with other flashcard programs, but Supermemo allows you to create Topics, which are basically digital index cards that are reviewed passively. Simply create a topic for whatever piece of information you want to remember, and when you see it (After first concentrating on the information), reschedule it (CTRL + R) for the next day. Keep doing this until the information no longer feels foreign when you examine it.

Supermemo is designed to help you remember, not necessarily learn. But by daily reviewing certain information in a decontextualized fashion has helped me "weaken" the difficulty of the information enough that I feel that I can "capture" it in a flashcard (Hence the Pokemon picture).

Again, this is an experiment that I've been playing with for the past month, so take it with a grain of salt. But it has worked wonderfully.

Tuesday, May 8, 2012

Why Pre-Made Flashcards Are Not Useful (For Now)

Perhaps you have heard of the video game crash of 1983 (Here is the Wikipedia article about it). There were many contributing factors to the crash of the video game industry at that time, but one of them was a market lacking organization flooded with bad quality titles. How could you tell the difference between a good game and a bad one? In an age before the internet, this was much more difficult to figure out.

Right now, pre-made flashcards suffer similar problems. There are many flashcards available from many sources, but curation is practically nonexistent. Will there be good sets of flashcards out there? Yes. Are there reliably good sources that consistently publish flashcards that adhere to the "20 commandments of knowledge formulation"? Not as easy to find.

Why is this? While Spaced Repetition Software has advanced greatly, public acceptance has not; traditional and archaic learning methods prevail and show no signs of changing due to the global institutional dysfunction (Among many other things). In other words, people don't really care because they don't understand that SRS is superior to traditional learning. Also, the public as a whole has not been taught how to properly structure a flashcard. Thus, many of the flashcards that do exist are poorly constructed.

So while there might be a few good sets of flashcards out there, I've found making my own much easier than modifying poorly formulated pre-made flashcards and weeding out unnecessary ones (Which you will likely have to do). It is also easier to personalize your flashcards (Add pictures, reference memories, etc.) right from the start, thus making them more memorable.

Thursday, May 3, 2012

Supermemo or Anki?

We live in a very critical and divisive world. We are divided and angered by the most insignificant preferences, be they computers, game consoles, shoes, stickers, what TV shows you watch, the list never ends. Granted not everybody plays into this, but in a media landscape that loves to judge, criticize and point out the flaws in others (Reality TV being one big contributor to this, I imagine), it has become much more acceptable to be angry about things that you don't need to be angry about. The anonymity of the internet doesn't help, either.

With the above in mind, at times, the same kind of pointless rivalry can exist within the world of flashcard systems.

There are many different flashcard systems out there, but two of the bigger ones you will likely see (Especially if you are using it for language learning) are Anki and Supermemo. There are many more (Mnemosyne, Skritter, etc.), but Anki and Supermemo seem to jump out as being the most popular.

Here is a quick list of pros and cons that readily come to mind for Supermemo

PROS
-More features
-Most advanced spaced repetition algorithm*
-More data about your memory than you can shake a stick at
-Incremental Reading
-Sleep tracker integration
-Flexible flashcard creation (Multiple images, sounds, etc.), made to encompass more than just languages

(* By "advanced" I mean that Supermemo 15.0 has been in development for a longer period of time than all other SRS programs out there. Put simply, it has the longest track record development-wise).

CONS
-Ugly compared to the simpler UI we have become accustomed to with modern technology
-Most of the features are lost to most users due to poor explanation, lack of simpler tutorial or lack of interest on the part of the user
-Most of the data about one's memory is not cared about by "casual" users
-Difficulty handling non-English fonts (Although this seems to have changed in Supermemo 15, I can write in Hindi, Japanese and Chinese with no problems).
-Must back up manually

While I have not used Anki for any length of time (I'm already committed to using SM every day), these seem to be the pros and cons as I see them. Just keep in mind I don't know this program very well:

PROS
-Simple(r) to use
-Free of features that some users may consider pointless (Sleep tracking, task lists, etc.)
-Made with language learners in mind (Specifically Asian languages)
-No font problems
-Sync your data through the internet
-Using Supermemo algorithm on the backend (SM-2)

CONS
-Algorithm is older than newest SuperMemo algorithm (Right now the algorithm is SM-15)
-No Incremental Reading support
-Fewer features that some users have grown accustomed to (Sleep tracking, task lists, etc.)

Think of it this way: Supermemo is like an apple, and Anki is like a banana. Both offer nutrition. While one person might prefer one fruit over another, it is better to eat a piece of fruit than a candy bar. Regardless of which you choose, keep this in mind. The effective use of ANY kind of SRS flashcard program is better than ANYTHING the modern education system offers by way of information retention. While Supermemo might have a feature that Anki doesn't, or Anki has a simpler and eye-pleasing style, they can BOTH be effective tools for learning (Especially for languages).

My advice? If you want to learn languages, start with Anki (Or any of the other programs offered online). If you find that you enjoy using Anki and want to try out a more advanced flashcard system, give Supermemo a try. If you want to learn something other than a language (Law, medicine, etc.), start with Supermemo.

Sunday, April 29, 2012

The Hipster PDA, A Useful Flashcard Tool

Considering that making flashcards is my hobby, one consistently useful tool for me has been what is called the "Hipster PDA," which is nothing more than a stack of index cards and a binder clip holding them together.

If a subject catches your attention and you wish to research it using SuperMemo later, simply jot it down on an index card. After you look the subject up and import a few articles into Incremental Reading, you can throw the card away.

If there is foreign language word you want to remember, you can quickly make a paper flashcard and familiarize yourself with the word until you feel comfortable enough to put it into SuperMemo for long-term storage.

They are also useful in giving others your phone number, e-mail, etc. quickly. Today paper is becoming more and more scarce in favor of some sort of electronic device.

I enjoy the act of throwing a "completed card" away, it feels like I am making progress in some intangible way.

Flashcard Principle - Single Knowledge Unit

When using flashcards to retain information, it is easy to overestimate how much information you can store in a single flashcard. This mistake is often made when it comes to languages (There is also a great deal of debate in regards to this). What strategy seems new and exciting in the beginning eventually becomes tiresome months and months down the line (And your flashcards are no longer interesting because they are no longer new).

One principle to keep in mind is that a single flashcard should only deal with a single unit of knowledge. Thus, instead of having a flashcard that asks "Who was Albert Einstein?", the questions should be "In what year was Albert Einstein born? When did he articulate the theory of relativity? In what year did Albert Einstein die? etc."

If you have spent any time making flashcards, you have no doubt encountered this principle in some way. While not overloading a single flashcard with too much information seems obvious when it comes to factual information, this seems to be neglected greatly when it comes to learning languages.

For example, if you were going to learn a Japanese character, often this kind of flashcard is made:

Q: 空
A: Sky, empty; そら、クウ

So when looking at "空," what are you supposed to recall? The meaning of the character? One of the many pronunciations? Which pronunciation? Let's say you recall the meaning of the character the first time you review it, but the second time you recall one of the character readings? The algorithm is only supposed to work with a single, unchanging memory, but the flashcard allows for many different memories to make it qualify as "correct."

In other words, one flashcard is trying to do the job of three or four flashcards. Instead of structuring a flashcard in such a broad and ambiguous way, the more effective thing to do would be to create a few highly focused flashcards such as these:

Q: Write the Japanese character for "Sky" or "empty."
A: 空

Q: How do you pronounce this character when it is with other characters (ON yomi)? 空
A: クウ

Q: How do you pronounce this character when it is by itself (KUN yomi)? 空
A: そら

There are many different ways of pronouncing this single character, and this does not reflect what I believe to be an effective character learning strategy. BUT you can see that dividing and focusing on single knowledge units rather than a bunch of knowledge units at a time will result in more accurate records for your flashcard software. This will result in your long-term retention of knowledge.

A flashcard is a sniper, not a shotgun.

Thursday, April 12, 2012

Twitter Page

I've created a Twitter account, and I will answer brief questions using either this or the comments section in the blog.

Tweet me at: @Supermemofish

Tuesday, April 10, 2012

Life Pro Tip - What is the VALUE of "activity x"?


A very useful and worthwhile practice is to get better at assessing how valuable a certain activity is. Be it brushing your teeth, laying off certain foods, doing Supermemo, etc., whether you acknowledge it or not, you assign a certain value to things you do. That is why sleep becomes less valuable if your house is on fire (life > sleep) or you interrupt your book to talk to a cute girl (potential love > (possibly disposable) knowledge intake).

While not all decisions are as obvious as the above, here is an exercise to make assigning values easier:

The next time you have to do something (Clean the house, do Supermemo, attend class on time, etc.), ask yourself: If I could pay someone money to do this activity for me, how much would I pay them?

Try doing this as you think of various projects and activities that you need (or wish) to do. The better you get at assigning number values to your activities, the more objectively you can view your decisions as a whole.

This (And the things I'm going to post in the future) has been helpful in accomplishing tasks preventing me from learning more. After all, the quicker you can trick your brain into doing things you have to do leaves more time remaining for things you want to do.

Friday, April 6, 2012

Scribblings Of A Madman


Hello; it's been a very long time since a previous post. Here's why:

For a bit of time I was kicking around the idea of writing a book about memory and Supermemo, but I have since opted out of doing this. Instead of unloading a large amount of information at once, I would rather dote it out in smaller chunks. Release many single songs on a consistant basis rather than an entire album with a great deal of time in between.

In an effort to simplify my life and embrace minimalism I've been trying to come up with what amounts to as an "Operating System for Life;" a series of principles/rules that can be applied to get work done as efficiently as possible. I've had mixed results over time, but I think I've found the overall solution; I'll post about it shortly.

Also, I think I've figured out a way to study/learn languages much faster than before, and I'm fairly certain I've learned how to make any piece of knowledge more "sticky," thus making it less likely to fall off the memory wagon as knowledge tries to make the journey from short-term to long-term memory.

All of this I will post about shortly.
Sorry for the long delay.