(The most up-to-date version of this document, and MemAid itself, can be found at: memaid.sf.net)

Back to Index

Effective learning - repetitions

Copyright (c) 2003 MemAid dev team
Everyone is permitted to copy and distribute copies of this document.

PROBLEM: forgetting

Well, first and foremost. we all forget.
We forget PIN-numbers, phone numbers, things we learn in school, at university, etc.
It's obvious, isn't it? There is nothing we remember forever.
That is, even if you do remember something throughout your life it's only because you repeatedly recall that fact/information.
If something is important to you, you will think about it from time to time -
these *repetitions* will reinforce your memory of this fact.
We repeat some things while not being fully conscious (e.g. while dreaming).
But let's make it clear: we would forget even own names without repetitions.

While learning anything (e.g. study/school) we have the same problem.
What is the solution for forgetting?
Yeah, I heard you said "repetitions". Correct.
But hey, there is another problem:
we want to learn new things, but we have to repeat older things so as not to forget them.
The problem is: how much time should we spend on repetition?
If we spend too much, we will repeat things we already remember well
(wasting our time). If we spend too little time, we will forget some things..
Not good...

SOLUTION: optimum spacing of repetitions

(Retention = proportion of remembered knowledge)

Yeah, good news.
People have studied human memory for years.
As far as we know, the first big step was done by Hermann Ebbinghaus,
in 1885 he published the manuscript "Memory: A Contribution to Experimental
Psychology" [http://psychclassics.yorku.ca/Ebbinghaus/].
He observed how and when we forget and researched it futher and ...
well, let's skip the boring details, the most important thing here is that we actually can,
more or less, estimate the optimum time of next repetition!
Pretty cool, isn't it?

The next milestone was achieved by Dr Piotr Wozniak.
He developed an algorithm that estimates the next optimal repetition time of an item
("item" is a simple fact). He wrote a great software package - SuperMemo [www.supermemo.com], which is very similar to MemAid.
(Well, MemAid uses some SuperMemo ideas - the main idea for such a program -
but MemAid uses a totally different approach (an Artificial Neural Network) to compute optimal intervals).

Supermemo or MemAid help you to manage your learning.
By monitoring your learning progress, they ensure that you remember new things, but don't forget existing things.
Repetitions are suggested only when necessary, so it's very optimal...

Active recalls

Basically, there are two kinds of repetitions:

Active recalls are MUCH, MUCH more effective, because while you are trying to recall any information, at the same time you are also effectively reinforcing your memory of this information!

Sometimes it's called the "power of question", because active recalls are also when somebody asks you something and you recall that information without any external help.

MemAid and SuperMemo use active recalls.

How effective is it?

David has been using SuperMemo for about a few years, then he developed MemAid (mainly because of a lack SuperMemo on Linux) and now he uses it everyday.
"It helps me a lot.
My first SuperMemo collection was "Advanced English97" - I have learnt then about 10000 items of knowledge (mainly English words I have never known before) in 6 months spending about 30 minutes a day. And I still remember most of them - thanks to optimal repetitions which are now very rare (few words a day to repeat). Whenever I now learn something, I use MemAid and I don't have to worry that I will forget what is in my learning process in MemAid. Nice feeling, really.

Happy learning!

Back to Index

© Copyright 2003 MemAid dev team