In 1974, Baddeley and Hitch
extended Atkinson and Shiffrin's work
on the multi-store model and
developed a much more sophisticated
understanding of short-term memory,
the working memory model.
Logie et al. (1989).
Baddeley and Hitch,
consists of three
The Central Executive (THE CONTROLLER).
This controls and
operation of the
It has a limited capacity.
The Phonological Store (THE SOUND SYSTEM).
led to the proposal of
two sub-components of
the phonological store:.
The Articulatory Loop (INNER VOICE).
This is a limited-capacity verbal
rehearsal component, which is
used to prepare speech and to
think in words, as when doing
mental arithmetic or
memorising a phone number.
The capacity of the
articulatory loop is
determined by how long
it takes to say something
and not simply by the
number of items.
If the articulatory loop has a limited
capacity determined by the length of
time it takes to verbally rehearse
information, then we should be able
to remember more single-syllable
names than triple-syllable names,
given the same amount of time.
The Primary Acoustic Store
This is a limited-capacity
system which receives
sound information from
The primary acoustic store
also receives auditory
information from our own
internal speech via the
articulatory loop, E.g.
When we 'hear' in our
head what we are thinking.
Evidence for two separate
subsystems in the phonological
loop comes from experiments
which involve measuring the
blood flow in participants' brains
when they are carrying out
different types of memory tasks.
Paulesu et al. (1993)
The Visuospatial Scratch Pad (INNER EYE).
This is a visuospatial
where we can image
and manipulate visual
and spatial information.
The best way to understand the
visuospatial scratch pad is to shut your
eyes and imagine the layout of the room
around you. Imagine standing and
making your way to the door. According
to Baddeley (1997), the visuospatial
scratch pad helps us to monitor where
we are in relation to other objects as we
move around our environment.
The scratch pad can also be used
to store visual information that has
been encoded from verbal stimuli,
such as words. E.g. If someone
says the word 'beach', you might
conjure up an image of a sandy
white beach with palm trees.
Due to the limited capacity
of the visuospatial scratch
pad, it is difficult to perform
several visuospatial tasks
at the same time, as any
learner driver will testify.
The central executive is the key
component of the working memory
system (Eysenck and Keane
2000), acting more like a system
which controls attention processes
than as a memory store.
The central executive
enables us to selectively
attend to some stimuli and
ignore others. It also plays a
role in retrieving information
from long-term memory.
In our everyday activities,
the central executive helps
us to decide when and
how to act (Matlin 2002).
Baddeley (1986, 1999) has
compared the central executive to a
company boss making decisions,
selecting strategies for dealing with
problems. It also integrates
information from assistants (the other
components), and calls on
information held in a large database
(long-term memory) (Matlin 2002).
If the central executive is
heavily involved in
controlling one task, it is very
difficult for it to do another
job at the same time.
Robbins et al. (1996).
As the role of the central
executive is very broad, the
model helps us to
understand the link between
the different cognitive
processes, such as memory,
perception and attention.
Working memory is a much more
flexible alternative to the fixed-capacity,
short-term store of the multi-store model,
and it sees memory as active rather than
passive (Eysenck and Keane 2000). It
considers how we use our short-term
memory for everyday activities.
The theory has
for the assessment and
treatment of people with
Problems with the phonological loop
system may be responsible for
difficulties in learning to read
(Baddeley et al. 1998). Children who
have difficulty reading often perform
badly on tasks which use the
phonological loop, such as deciding
whether or not two words rhyme.
Problems experienced by patients
with brain damage can be explained
using the working memory model. E.g.
If only the visuospatial scratch pad is
damaged, then performance on tasks
which rely on the phonological loop
will remain unimpaired.
The working memory
model explains only the
short-term memory and
makes no attempt to explain
long-term memorising. as
such, it does not provide an
overall theory of memory.
We know relatively little about the
central executive, even though it
is the most important component
of the model (Baddeley 1997).
to develop and
refine the working
He has recently proposed the
existence of another sub-component,
the episodic buffer, that enables
communication between working
memory, long-term memory and
present experience or consciousness.