Miller’s article explains how a person can store 7±2 chunks of information in short-term memory. The word “chunks” is used quite deliberately as each chunk can contain different values of information, from a single digit to a rich word that represents an entire concept. The value of each chunk doesn’t affect short-term memory – but the number of chunks does.
Bruner explains that, while we seem to be able to do little to increase the number of chunks available to us, we can increase the value of each chunk by enriching our language with words and symbols that represent bigger concepts. This is the most compelling reason to learn new words, pattern names and domain-specific notation you will ever be given.
So if you have a seven-chunk representation of a concept and you can roll it up into a single chunk by creating a word that represents it, you have a system that will give you perhaps seven times the ability to recode information as it slides through our short-term memory.
In programming, design patterns were created to represent entire designs in one or two words – so if you and your team learn what a “decorator” or “factory” is, you get the benefit of entire concepts in a single chunk – and programming terms like “encapsulation” or “abstraction” perform a similar role.
So whatever domain you are working within, discovering and understanding the notations and language that represent important concepts will lead to more effective communication, more effective thought and better decision making.
Toward a Theory of Instruction – Jerome S Bruner (1966)
The Magical Number Sever, Plus or Minus Two – George A Miller (1955)