Working Memory

addictedswimmingΤεχνίτη Νοημοσύνη και Ρομποτική

24 Οκτ 2013 (πριν από 3 χρόνια και 7 μήνες)

60 εμφανίσεις


At the time, short term memory research was largely devoted to the study of
an acoustic, temporary, limited capacity verbal store, and was typically
measured by a simple digit span task where subjects were asked to repeat
varying series of numbers. If prevented from rehearsal, short term memories
vanished in a matter of seconds.

This chapter directly addresses the hypothesis from chapter 1: That an
enhancement in working memory capacity powered the appearance of the
modern mind.

By the 1950’s, psychologists had come to a more or less general
agreement that people relied on two distinct kinds of memory; short term
and long term.

By the early 1970’s, experimental psychologists including Alan Baddeley
recognized limitations in this two component model of memory.

Competing models attributed this to either fading of the electrical trace of the
memory or because later perceptions interfered with them.

Both of these models failed to account for significant experimental results. For
example; it was unable to explain why some simultaneous tasks interfered with one
another (memorizing a list of words while reciting a list of numbers) but others did
not (remembering the colors of presented objects while reciting a list of numbers).

In 1974, Baddeley and his colleague Graham Hitch proposed a more
comprehensive cognitive theory that accounted not only for the standard
operations of short term memory but also how memory is enjoined and directed,
and how it is related to long term memory.

The initial Baddeley and Hitch model included an attentional, panmodal controller
or central executive, and two subsystems: the phonological loop and the visuospatial
sketchpad but recently Baddeley expanded the central executive’s functions by
adding the episodic buffer.

This episodic buffer serves as the memory component of the central executive and
integrates and temporarily stores information from the other two subsystems.

The phonological loop contains two elements: a short term phonological store of
speech and other sounds and an articulatory loop that maintains and rehearses
information either vocally or subvocally. Baddeley viewed its primary purpose as
evolving for language acquisition and comprehension.

The visuospatial store was hypothesized to involve the maintenance and integration
of visual (“what” information) and spatial (“where” information) elements and a
means of refreshing it by rehearsal.

Baddeley and Hitch’s model of working memory

With some modifications, Baddeley and others currently view the central
executive either as a unitary system or multiple systems of varying functions
including attention, active
inhibition, decision
making, planning, sequencing,
temporal tagging, and the updating, maintenance, and integration of
information from the two subsystems.

In most current models, working memory not only serves to focus attention
and make decisions but also serves as the chief liaison to long term memory
systems, and to language comprehension and production.

Baddeley adopted an attentional control system called the Supervisory
Attentional System (SAS), originally proposed by Norman and Shallice (1980)
as the basis for his central executive.

The SAS takes control when novel tasks are introduced, when pre

habits have to be overridden, or when danger threatens and task relevant
decisions must be made.

The Central Executive

How and why does the central executive make its decisions?

Executive attention is the critical component of working memory, whose chief
function is the active maintenance of appropriate stimulus representations relevant to
goal attainment in the face of interference rich contexts.

Miyake and Shah (1999) proposed that the attention and decision making qualities
of the central executive may be an emergent property that arises as a function of the
dynamic interplay of the multiple and interrelated systems associated with working

Barkley (2001) also favored an evolutionary perspective to explain executive
functions. He viewed them as a biological adaptation resulting from interpersonal
competition in groups. He saw executive functions as a useful social self defense
against resource theft and against interpersonal manipulation. He also saw them as
advantageous in social exchanges and in imitating and learning from others.

He proposed that these executive functions evolved in gradual stages over a period
of at least a million years

the ability to attend to relevant stimuli, filter out irrelevant
stimuli and to make quick and efficient decisions was favored over static processes.

Neural Structure

The executive functions appear to result from the interplay of diverse cortical
and subcortical neural systems as well as the dorsolateral prefrontal cortex,
the orbitofrontal prefrontal cortex, and the anterior cingulated cortex. There
has also been evidence presented for the basal ganglia and the cerebellum.

The dorsolateral circuit is generally associated with the classic executive
complex problem solving, decision making, verbal fluency, and
some of the operations of working memory.

Gazzaniga et al (2002) attributed its attentional functions primarily to the
anterior cingulated gyrus.

Hazy, Frank, and O’Reilly (2006) proposed a complex model, called PBWM
(prefrontal cortex, basal ganglia working memory model) which accounts for
the mechanistic basis of working memory, the central executive, and its
executive functions. They view the prefrontal cortex as critical in maintaining
representations of an individual’s perceptions in the broadest sense which are
dynamically updated and regulated by reinforcement learning systems based
on chemical neurotransmitters (primarily dopamine) activated by the basal
ganglia and the amygdala.

Phonological Loop

The phonological loop is intimately involved in language use. Basically, it
engages in various verbal and acoustic tasks and keeps these going until you
are done with them.

Baddeley hypothesized that it has two components: a brief sound based
storage that fades within a few seconds and an articulatory processor which
maintains material in the phonological store by vocal or subvocal rehearsal.

Spoken information appears to have automatic and obligatory access to
phonological storage and therefore Baddeley hypothesized that it evolved
principally for the demands and acquisition of language.

Repetition of sounds held in the phonological store, usually by means of the
vocal or subvocal articulatory processor will relegate those sounds into long
term declarative memory if there is sufficient motivation or emotional
salience. A strong motivation to memorize or an elevated emotional meaning
will increase the likelihood that the sound will be successfully transferred into
long term memory.

Phonological Loop

Sounds can be relegated to long term memory even if they have no initial
meaning. For example repeating “Won
due, era
due, muru” over and over
either vocally or subvocally will eventually transfer them to long term memory.
( phonetic sounds of 1
3 in a south Indian language)

The phonological loop’s processes also help explain why brain
patients who have lost their ability to repeat sounds vocally can still memorize
them. However, those patients who cannot create a sound or speech motor
form through the phonological loop cannot memorize new material.

Recently, Aboitiz, Garcia, Bosman, and Brunetti (2006) have noted that
phonological storage capacity represents a short term memory ensemble that
can be phylogenetically tracked to earlier homologues in hominin evolution
and to current primate brain systems. Further, they postulated that language
has evolved primarily through the expansion of short term memory capacity,
“which has allowed the processing of sounds, conveying elaborate meanings,
and eventually participating in syntactic processes”.

Plays an important role in adult second language learning, the
acquisition of native language learning in children, and memorizing
stimuli in the visuospatial sketchpad.

Phonological Loop Neural Structure

Aboitiz et al. situate the neurological epicenter of the phonological loop
in the posterior superior temporal lobe gyrus and the inferior parietal
lobes areas.

They also agreed with Fuster (1997) who noted that the dorsolateral
prefrontal cortex plays an important role with reconciling short term past
and short term future and cross temporal contingencies.

Thus insuring that one’s present speech is in accord with previous
utterances is a function of the complex interactions of the prefrontal
cortex, temporal, and parietal areas as well as interconnectivity with other
cortical area and subcortical structures.

Becker, MacAndrew, and Fiez (1999) have also identified the neural
location of the phonological store as the inferior parietal lobe of the
dominant hemisphere, particularly the supramarginal and angular
gyri. Additionally, Broca’s area has been implicated with phonological
tasks that present the items visually.

Visuospatial Sketchpad

A temporary store for the maintenance and manipulation of visuospatial
information. It also has an important role in spatial orientation and solving
visuospatial problems.

Studies with brain damaged patients, and with healthy adults, appear to
demonstrate that the visual and spatial processes may comprise separate
memory systems, although the visuospatial sketchpad is assumed to form an
interface between the two systems.

Visual information can also be integrated with sensory information such as
touch and perhaps smell.

Neuroimaging and other neuropsychological studies with a variety of brain
damaged patients and other people indicate significant involvement of the
primary visual cortex (occipital lobes), the parietal lobes, and the frontal lobes.

Episodic buffer

Baddeley initially described the central executive as largely attentional
in nature without its own storage capacity, but eventually realized it also
must have some way to store information independent of the
subsystems (how else could phonological, visuospatial, and long term
memory information be integrated?)

He thus proposed the episodic buffer as the storage component of the
central executive.

He endowed the episodic buffer with the ability to bind and integrate
the two subsystems, the phonological loop and the visuospatial
sketchpad, and also traces from long term memory via a multimodal

By attending to multiple sources of information simultaneously, the
central executive is able to create models of the environment that
themselves can be manipulated to solve problems and even plan future
behaviors and alternative strategies, so that if a plan fails another may
be chosen or generated.

A Memory Primer

Cognitive psychologists have long distinguished between
declarative memories and procedural memories.

Procedural memory is the long
term memory of skills and
procedures, or "how to“ knowledge (riding a bike, tying your

Declarative memory is a person's memory of facts or events.
It is learned information (phone numbers or the makeup of
chemical compounds).

Both types appear to use relatively independent neural
pathways; selective types of brain damage may affect one
type of memory but not the other.

Declarative memory

There are varieties of declarative memory.

An episodic memory is a coherent, story like reminiscence for an
event, often associated with a specific time and place and a
feeling signature. It is sometimes labeled personal memory or
autobiographical memory.

The other type of declarative memory is semantic

the memory
for general facts.

Reminiscence will include semantic details, but its recall and
subjective experience will be psychologically and neurologically
different from the recall of the semantic components alone.

Episodic memory is advantageous in that people use their
episodic memories to simulate future scenarios. This ability has
been labeled constructive episodic simulation by Schacter and

Declarative memory

Tulving (2002) has proposed that the ability to simulate and
contemplate future scenarios has been the driving force in the
evolution of episodic memory.

He proposed the term autonoesis to refer to the ability, unique to
humans, to form a special kind of consciousness in which individuals
become aware of the subjective time in which past events happened. It
is also this ability that allows humans to travel mentally in time.

Mental time travel, by way of episodic processes, allows awareness of
not only the past but of what may happen in the future. “This
awareness allows autonoetic creatures to reflect on, worry about, and
make plans for their own and their progeny’s future in a way that those
without this capability possibly could not. Homosapiens, taking full
advantage of its awareness of its continued existence in time, has
transformed the natural world into one of culture and civilization that
our distant ancestors, let alone members of other species, possibly
could not imagine.”

Declarative memory

Baddeley proposed that greater working memory capacity
would allow for the reflection and comparison of multiple past
experiences and that this might allow an individual to actively
choose a future action or create an alternative action (based on
the success or failure of previous actions) rather than simply
choosing the highest path of probably success.

Recent neuroimaging studies now link the prefrontal cortex to
episodic memory recall.

When normal adults are asked to form episodic memories
(through experimental manipulation), the left prefrontal
cortices are differentially more involved than the right
prefrontal cortices, whereas when they are asked to recall
them, this pattern is reversed and right prefrontal cortices are
more heavily involved.

Working memory capacity

Because working memory appears to involve simultaneous attention to task
relevant information, as well as its manipulation, processing and storage,
experimental psychologists researching individual differences developed the
concept of a working memory span and more recently working memory
capacity. (synonymous although measurement varies according to the
experimental task)

Working memory capacity= simply how many items can be recalled during a

Engle and Kane have importantly noted that there is probably no pure
measure of working memory capacity, as a task must be designed in the
domain such as verbal, visual, acoustic, spatial etc. Thus it may not be directly

A variety of measures have been developed to asses working memory
capacity and these measures do have practical applications.

Working memory capacity

Working memory correlates with a number of important practical
abilities including reading comprehension, vocabulary learning, language
comprehension, reasoning, suppression of a designated event, language
acquisition and second
language learning, many neuropsychological
measures, fluid intelligence, and general intelligence.

The strong relationship between fluid intelligence and working memory
capacity is an important one because fluid intelligence is thought to
measure ones’ ability to solve novel problems and it appears less
influenced by learning and culture and more by some natural or inherent
ability to figure out solutions to problems.

Heritability of working memory

It is clear that the bulk of human nature and behavior;

predilections, predispositions, fears, personality, psychopathology, and
motivations have evolved via natural selection upon genetic mutations over
millions of years.

In 2001 Morley and Montgomery compiled a list of over 150 possible genes
that appear to influence cognition. Thus there is currently good empirical
evidence that working memory, its executive functions, and its subsystems
have a strong genetic basis.

Coolidge et al. (2000) did an analysis of child and adolescent twins as rated
by their parents and found that a core of executive functions, consisting of
planning, organizing, and goal attainment, was highly heritable (77 percent)
and due to an additive genetic influence.

Ando, Ono and Wright (2001) also found a strong additive genetic influence
49 percept) upon working memory storage and executive functions, in
both phonological and visuospatial tasks.

Heritability of working memory

Rijsdiik, Vernon and Boomsma (2002) found a 61 percent additive
heritability in young adult Dutch twin pairs on the digit span task of the
Wechsler Adult Intelligence Scale, which is one measure of phonological
storage capacity.

Hansell et al (2001) using event related potential slowwave measure of
working memory in a visuospatial task, showed solid heritability in a sample
of 391 adolescent twin pairs.


Working memory is a highly heritable trait and has received considerable

Hard to prove but it appears there may be some pure working memory capacity
that varies across individuals and this undoubtedly evolved in primates and in
hominin evolution.

The nature of this capacity has something to do with attention to task relevant
stimuli and the ability to maintain this information in active memory.

The neural substrate of working memory, its executive functions and subsystems
appears to be the frontal and prefrontal cortex, parietal and temporal cortices,
and the basal ganglia among other cortical and subcortical regions.

Studies have shown that working memory capacity is limited. When attention is
diverted more info gets in and the old info is pushed out because of this limit.
Most people average around 7 slots (plus or minus 2) and can hold info in short
term store without rehearsal for around 20 seconds.