The Psychology Book (54 page)

embraced—focused on the study of

Miller’s main interest was in the

delusions of persecution.” Despite

mental processes, such as memory

field of psycholinguistics, stemming

the whimsical nature of his title

and attention. At the same time,

from his work during World War II on

COGNITIVE PSYCHOLOGY 171

See also:
Hermann Ebbinghaus 48–49 ■ Bluma Zeigarnik 162 ■ Donald Broadbent 178–85 ■ Endel Tulving 186–91 ■

Gordon H. Bower 194–95 ■ Daniel Schacter 208–09 ■ Noam Chomsky 294–97 ■ Frederic Bartlett 335–36

by the recurrence and possible

accurately assigning numbers to

significance of the number seven;

each of them, but above seven

“sometimes a little larger and

(give or take one or two), the

sometimes a little smaller than

results deteriorated dramatically.

usual, but never changing so much

In another experiment, by

The persistence with

as to be unrecognizable.”

Kaufman, Lord,
et al,
in 1949,

The first instance of the

researchers flashed varying numbers

which this number

“magical” number came from

of colored dots on to a screen in

plagues me is far more

experiments to determine the

front of participants. When there

than a random accident.

span of absolute judgment—how

were fewer than seven dots,

George Armitage Miller

accurately we can distinguish a

participants could accurately

number of different stimuli. In one

number them; when there were

experiment cited in Miller’s paper,

more than seven, participants were

the physicist and acoustic

only able to estimate the number

specialist Irwin Pollack played a

of dots. This suggests that the

number of different musical tones

span of attention is limited to

to participants, who were then

around six, and caused Miller to

speech perception, which formed the

asked to assign a number to each

wonder whether the same basic

basis for his doctoral thesis. This

tone. When up to around seven

process might be involved in both

led him to take an interest in the

different tones were played, the

the span of absolute judgment

growing field of communications,

subjects had no difficulty in

and the span of attention. ❯❯

which in turn introduced him

to information theory. He was

particularly inspired by Claude

Shannon, a leading figure in

communications, who was

investigating effective ways of

turning messages into electronic

signals. Shannon’s communication

model, which involved translating

ideas into codes made up of “bits,”

underpins all digital communication.

Miller was inspired to look at mental

processes in a similar way, and to

establish the ground rules for the

modern field of psycholinguistics

in his 1951 book,
Language and

Communication
.

Seven categories

Miller took Shannon’s method of

measuring information and his idea

of “channel capacity” (the amount

of information that can be processed

by a system) and applied it to the

An experiment into the span of attention
presented

model of short-term memory as an

participants with random patterns of dots flashed on a

information processor. This was

screen for a fraction of a second. Participants instantly

when he began to be “persecuted”

recognized the number if there were fewer than seven.

172 GEORGE ARMITAGE MILLER

The tones and dots in these

took this idea of channel capacity

experiments are what Miller calls

a stage further, applying it to the

“unidimensional stimuli” (objects

model of short-term memory.

that differ from one another in only

William James first proposed the

one respect); but what interested

notion of short-term memory, and

Miller is the amount of information

it had long been an accepted part

The process of memorizing

in speech and language we can

of the model of the brain as an

may be simply the

effectively process, and items such

information processor, coming

formation of chunks…

as words are “multidimensional

between the sensory input of

until there are few enough

stimuli.” He looks to later studies by

information and long-term memory.

chunks so that we can

Pollack in which the simple tones

Hermann Ebbinghaus and Wilhelm

recall all the items.

were replaced by tones that varied

Wundt had even suggested that

George Armitage Miller

in six ways (such as pitch, duration,

short-term memory had a capacity

volume, and location). Surprisingly,

limited to around seven items

despite the apparently larger amount

(seven, again). Miller believed that

of information, the results still

what he called working memory

pointed to a differential limit of

had a capacity that corresponded

seven, plus or minus two. The

to the limits of absolute judgment

difference is that as more variables

and span of attention.

are added, accuracy slightly

that by the same principle,

decreases. Miller claims this allows

Bits and chunks

working memory organizes “bits”

us to make “relatively crude

In terms of our ability to process

of information into “chunks,” to

judgments of several things

information, if working memory is

overcome the informational

simultaneously.” It may explain how

limited to about seven elements,

bottleneck caused by our limited

we are able to recognize and

there is a potential bottleneck

spans of absolute judgment and

distinguish such complex things as

restricting the amount that can be

short-term memory. A chunk is not,

spoken words and people’s faces,

put into long-term memory. But

however, just an arbitrary grouping,

without having to process the

Miller suggested that there was

but an encoding of bits into a

individual sounds or features.

more to the correspondence than

meaningful unit; for example, a

Miller sees the human mind as

just the number seven, no matter

string of 21 letters represents 21

a communication system: as the

how magical it appeared. The

bits of information, but if this can

input information increases, the

multidimensional stimuli of

be broken down into a sequence

amount transmitted to the brain

previous experiments could be

of three-letter words, it becomes

also increases initially, before

seen as composed of several “bits”

seven chunks. Chunking is

leveling off at an individual’s

of related information, but treated

dependent on our ability to find

“channel capacity.” Miller then

as a single item. Miller believed

patterns and relationships in the

Supercalifragilisticexpialidocious

Super

cali

fragi

listic

expi

ali

docious

Miller’s theory of chunking
says that by building

up or breaking down long streams of numbers or letters

into memorable chunks, we increase the amount of

information we can hold in working memory.

COGNITIVE PSYCHOLOGY 173

Binary code
is a way of recoding

information into ever-more tightly

packed parcels (through multibase

arithmetic). Miller claims our chunking

process operates in a similar way.

powerful weapon for increasing the

amount of information we can deal

with.” It effectively stretches the

informational bottleneck.

The study of memory

Miller himself moved away from

the subject of memory in his later

George Armitage Miller

research, but his theory prompted

George Armitage Miller was

others to examine it in more detail.

born in Charleston, WV. After

Donald Broadbent argued that the

graduating from the University

real figure for working memory is

of Alabama in 1941 with an

probably less than seven, and this

MA in speech pathology, he

bits of information. To someone

was later confirmed in experiments

earned a PhD at Harvard in

who does not speak the same

by Nelson Cowan, who found it to

psychology, working in Stanley

language, the seven words might

be around four chunks, depending

Smith Stevens’ Psychoacoustic

be meaningless, and would not

on the length and complexity of the

Laboratory, with Jerome

constitute seven chunks, but 21 bits.

chunks, and the age of the subject.

Bruner and Gordon Allport.

Miller’s theory was backed up

In the conclusion to his paper,

During World War II the

by earlier experiments by other

Miller is dismissive of the

laboratory was asked to help

psychologists. In 1954, Sidney

significance of the number

with military tasks such as

radio jamming.

Smith conducted experiments in

that originally prompted it. He

In 1951, Miller left Harvard

memorizing a sequence of binary

concludes by saying: “Perhaps there

for Massachusetts Institute

digits—a meaningless string of

is something deep and profound

of Technology (MIT), then

ones and zeroes to anyone

behind all these sevens… but I

returned to Harvard in 1955,

unfamiliar with the binary system.

suspect that it is only a pernicious,

where he worked closely with

Smith broke the series down into

Pythagorean coincidence.” ■

Noam Chomsky. In 1960, he

chunks, at first into pairs of digits,

cofounded the Harvard Center

and then in groups of three, four,

for Cognitive Studies. He later

and five, and then “recoded” them

worked as a professor of