I am not sure I should write this post in English as far as I am neither HR nor Law specialist and the language is not my native. Still I have a perso...
December 4, 2014
Learning How to Learn
This article is here to provide several extremely important methods and techniques to master new learning skills. The article is based on "Learning How to Learn" training course by Dr. Barbara Oakley of "UC San Diego". The course is provided and distributed by Coursera on a regular basis. Please be aware that this article is a brief overview of some parts of the course and you are strongly advised to complete the full course yourself.
We are going to target the following topics throughout this article:
Basic brain physiology
Brain in sleeping
Before starting any self-improvement the one needs first to understand the systems and mechanisms he is going to improve. It is obvious we are talking about the Central Neural System (CNS) especially the brain as applied to learning skills. You are most likely aware that your brain is an enormous network of interconnected neural cell (neurons). Each neuron in the network is able: 1) to receive some input from one or many other neurons; 2) to process the input with some fixed internal rules; 3) to generate a SINGLE output to be "read" by the following neuron (neurons). There are several basic things to remember here. The "rules" inside the neuron are immutable - they do not change with respect to input or situation. There is only one output signal generated by a neuron and the signal has binary value. For example, a neuron can not fire with 70% power. The output value is always True or False.
You can see that a neuron is not that complex and it is definitely easy to make a working model of a neuron now. The bad thing here is uselessness of a neuron model to simulate the brain. Same as a single transistor doesn't make a computer. The first computers in 1950s had about 200 interconnected transistors. In 2014 we have the latest iPad A8X chip with 3*10^9 (3 billion) transistors. This sounds quite a lot and not that far away from the brain of a human adult having about 10^11 (100 billion) neurons. Most parts of the brain can not generate new neurons, so the number of neurons since birth is permanently decreasing.
What IS important is the structure. While CPU (Central Processing Unit) of a computer has the transistors interconnectivity fixed by design, the brain is always in the process of restructuring connections. A single physical connection between neurons is called a "synapse" and there are 10^14 - 10^15 (quadrillion) of them in a human brain. These synapses define the "workflow" of a "thinking" process.
Well, now you know some very basic physiology of the brain activity and should understand the importance of effective brain restructuring. As soon as scientists found out these basics they studied the process of new synapse generation and have noticed that the rate of restructuring long-term memory (to be discussed later) is most active during sleep. This means that if you have learned something new and do not "sleep it over" the knowledge will be hard to recall and very fast to forget. You may think of an analogy to disk fragmentation with the only difference that the disk will not have fragmented data overwritten but the brain could. Defragmentation helps significantly lower access time and increase reading speed both for brain and for disk. The illustration shows the schematic state of memory before and after defragmentation.
The other importance of a good sleep is the cleaning of "toxic" substances, which are generated during wakeful state. It happens that during sleep the neurons shrink a little and let the Cerebrospinal fluid (CSF) wash out the toxins which can harden new synapses forming. There are plenty other reasons for a human to sleep enough but these ones have the most significant influence on the process of learning.
You may have noticed already that we use several important analogies with computer mechanics in this article. Well, this is because the author is a strong believer in scientific description and explanation of every nature process. It simply takes some time for the humanity to find out this or that. And now we are going to talk about memory, especially about short-term (Working) and long-term memory. These can be compared to Primary memory usually known as Operative memory or Random-Access Memory (RAM) and Secondary memory (hard-drives and other storage) correspondingly. Many of the memory properties and behaviors are very similar in computer and brain, so we shall understand the basics of both.
What is good about Working memory? CPU (or brain) has extremely fast direct access to it, so almost everything you are working with "right now" is loaded into Working memory. As far as all the memory is part of the brain itself we are talking about different segments corresponding to the thinking (or processing) and memorize/recall processes from the storage (long-term memory).
What is bad about Working memory? Both machines and humans have very limited amounts of it compared to long-term storage. An average adult brain has only about four "units" of short-term memory. Actually, this can not be compared to bytes (or gigabytes, or smth.) in computer RAM because human memory is not binary. Although the neural signal is binary, the many-to-one connections make complex networks and the unit of memory itself is a small part of this network. To visualize one of the four "units" in your Working memory you may think of them as of simultaneous ideas currently in your head. It is obvious that to get any new idea "loaded" to the short-term memory you have to release some of the old ones first. The following picture demonstrates the brain as octopus who can access long-term memory through several (four) holes - "units" of your working memory.
Still the size matters. The stronger is the memory in your "storage" the simpler is it actually in the neural network. The synapses are organized more effectively, they use less neurons and are much faster to access. Therefore, the knowledge is easier to load to your Working memory and takes less space. For example, when you know some favorite meal recipe very good and have cooked it thousand times, you can easily cook, watch TV and talk to your friend using hands-free simultaneously with no risk of cooking something unpalatable.
Here goes a simplified workflow of great importance when learning something. Each time after you recall the knowledge it is "marked" for reconstruction. The next time you go to sleep the brain optimizes it a little and thus strengthens in your long-term memory. Actually, this is almost the only way to remember something "for long". Special attention is required to the fact that not "WRITE" to memory works but "READ" from it.
The following conclusions come here:
You should not reread something. Recall it!
You should not practice one thing many times in a row. Optimization the following night will run only once no matter how many times you recalled the knowledge the previous day.
Recall something you have learned the next day, then after several days, then a week later and several times more after time has passed. This gives the best optimization for strong memory.
The schema represents how the strength of some knowledge grows in your brain with multiple recalls.
I hope this article helped you get a better understanding of the basic processes in human brain and discover useful techniques of effective learning. The feedback is always welcome.
(c) 2014, Nikolay Grishchenko
List of materials
"Learning How to Learn" course (materials of Dr. Barbara Oakley), 2014.
Octopus of attention and diffuse tentacles, images. (c) Kevin Mendez, 2014
Brain during sleep, images. (c) Kevin Mendez, 2014
Neural pinball images and patterns. (c) Kevin Mendez, 2014