Magical Number Seven

The Magic Number 7 ±2

Miller in 1956 (“The Magical Number Seven, Plus or Minus Two: Some Limits on Our Capacity for Processing Information”) Miller showed a number of remarkable coincidences between the channel capacity of a number of human cognitive and perceptual tasks. This was the number seven plus or minus two (7±2).

Read the original article

Much has been quoted about Millers Magical Seven, before we explore the basics on this page you may wish to read the original article: Miller’s Magical Seven

The magic seven in training and learning

The use of the magic 7± has been used as fact by many trainers, instructors and other ‘educators’ without understanding, indeed it is thrown as a fact rather than as a piece of a model of learning or cognitive thinking capacity.

Magic Number in everyday life

Some parts of our world seem to have taken the magic number seven to mean a limit or structure, indeed if you look in much of the IT world many of the menus and options are grouped in seven’s or near sevens. This is just a fallacy which appears to have entered the ergonomic designers world – and not we are all inflected with it in inappropriate times and places.

The research was never intended for this application, nor indeed to limit functionality.

The Basics – Miller’s Magic Number Seven

In 1956, George Miller’s study identified that the amount of information which can be remembered on one exposure is between five and nine items, depending on the information.

This range is conveniently the number seven, which has long held ‘an interest’ for people. Applying a range of +two or -two, the number seven became known as Miller’s Magic Number Seven (7±2). This is the number of items which is thought can be held in short term memory at any one time.

The Limits
Miller himself stated that his magic number was for items with one aspect or attribute.

Millers work was based on subjects listening to a number of auditory tones that varied only in pitch. Each tone was presented separately, and the subject was asked to identify each tone relative to the others she had already heard, by assigning it a number. After about five or six tones, subjects began to get confused, and their capacity for making further tone judgements broke down.

Miller found this to be true of a number of other tasks. However when more attributes were included, then we can remember more, depending upon our familiarity and the complexity of the subject (Remember in Miller’s original research, there was only one aspect, the tone).

We remember phone numbers by their aspects of 2 or more groupings. We don’t really remember “seven” numbers. We remember the first group of three and then the other grouping of four numbers. If it is longer then we add an area code. So we actually remember 10 numbers by breaking it into groups of three. National Insurance numbers, bank account numbers etc work on the same principle; xx-xx-xx-xx (four groups of numbers).

Remember

The Magical Number Seven is only true with information with one memory attribute or function – we CAN work with a lot more – just look at the world memory Olympics – there people remember and recall large amounts of data – but they do chunk them up and add references that naturally were not there in the first place!

Good for remembering numbers – but not good for other factors!

Applications – Chunking

Believing that a limited number like 7 is valuable is one thing – applying it is another. Chulking is a term which has been developed to to use the limits of the human short term memory.

Chunking refers to a strategy for making more efficient use of short-term memory by recoding information.

In addition to its short term usage, Herbert Simon has used the term chunk to indicate long-term memory structures that can be used as units of perception and meaning, and chunking as the learning mechanisms leading to the acquisition of these chunks.

A chunk can then be defined as “a collection of elements having strong associations with one another, but weak associations with elements within other chunks” (Gobet et al., 2001, p. 236).