Memory

 The multi-store model of memory: sensory register, short-term memory and

long-term memory. Features of each store: coding, capacity and duration.
 Types of long-term memory: episodic, semantic, procedural.
 The working memory model: central executive, phonological loop, visuo-
spatial sketchpad and episodic buffer. Features of the model: coding and
capacity.
 Explanations for forgetting: proactive and retroactive interference and
retrieval failure due to absence of cues.
 Factors affecting the accuracy of eyewitness testimony: misleading
information, including leading questions and post-event discussion; anxiety.
 Improving the accuracy of eyewitness testimony, including the use of the
cognitive interview.
Multi-Store Model of Memory
AO1 –
Atkinson and Shriffin (1968) proposed this model which suggested
that memory was comprised of 3 key components: sensory register,
short term memory and long term memory
According to model, memories are form sequentially:

 - Info enters via sensory register (coding raw and

unprocessed information). SR has an unknown but large
capacity and a very limited duration period of 250
millisecond.
 - If the information is paid attention to it is transferred to the
STM, which has a limited capacity of 5-9 items of information
(Miller, 1956) and a limited duration of 18sec (P&P, 1959).
This information is coded acoustically (Baddeley, 1966)
 - Then, to be transferred to LTM, this info enters a rehearsal
loop in which it is maintained. If it is rehearsed enough, it
enters LTM which has an unlimited capacity and a lifetime’s
duration (Bahrick et al 1975). Information here is coded
semantically (Baddeley, 1966) and can be retrieved from LTM
to STM when required.

MSM has been investigated and supported by a wealth of

researchers. Like Miller who concluded from a literature
review that STM has a capacity of 5-9 chunks of information
or Peterson and Peterson who concluded from preventing
rehearsal of a series of nonsense trigrams that STM has a
duration of 18 seconds, with only 3% of his sample
remembering the trigram after 18 seconds of counting back in
3s.
AO3 – Evaluation of support AO3 – Support from CW
evidence Clive Wearing, who suffered from
e.g. Bahrick et al a viral amnesia, could only
remember new information for 20-
While his evidence has high 30 seconds; however, he was able
ecological validity as the sample to recall information from his past
of university graduates were (like his wife’s name).
shown asked about their own life It has been concluded that Clive
experiences, this provides a Wearing was unable to transfer
more meaningful and accurate information from STM to LTM but
representation of the human able to retrieve information from
memory. LTM.
However, confounding variable This supports that idea that
may be uncontrolled and memories are formed from the
therefore influence the results. sequential linear passage from
Individuals may have seen their store to store and damage to any
friends since then and therefore part of the MSM can cause
retrieved and maintained the memory impairment.
information since.
AO3 – Influence on Psychology
AO3 – Patient KF (Shallice and This model revolutionised how
Warrington) psychologists viewed memory, as
PF, who was injured in a a dynamic and active process.
motorcycling accident, was able
to recall stored information from Raised the status of psychology
his LTM but had issues with his and encouraged others to
STM. He was able to transfer research into memory which has
visual images into his LTM, but lead to major developments in our
not acoustic information. understanding of recall and
improving eyewitness testimonies
This study suggests that there
are at least 2 components within However, this model does not fully
STM, which are not outlined in explain different types of STM.
the MSM. This makes it an overly Thus, Baddeley and Hitch
reductionist model and a level of developed the working memory
valuable complexity is lost in model as an explanation for the
this. Indicates that more complexity of STM and a way of
research should be conducted to explaining some research findings,
understand the complexity of the like the results of their dual-task
STM. studies and Shallice and
Warrington’s findings with KF,
But KF was one individual who which are not accounted for in the
endured an intense head MSM. The additional complexity of
trauma. As he is one individual WMM suggests that the MSM may
with a confounding variable and be an incomplete model of
these findings are very memory and more research is
uncommon, caution must be needed into the action of memory
taken when generalising this to formation.
everyone’s memory

Coding Capacity Duration

 Raw/unprocessed
Very limited (approx.
SR information (from ALL 5 Unknown, but very large
250 ms)
senses)
Limited (7+/-2 ‘chunks’
ST Limited (20 secs)
Acoustic (sound) Baddeley of information) Jacobs
M Peterson & Peterson
Miller
LT Semantic (meaning)
Unlimited Lifetime/years Bahrick
M Baddeley

Baddeley – semantically similar is difficult to store in LTM

Types of Long-Term Memory

 - Declarative
 - Knowledge – factors and
concepts
 - Starts as episodic but not
‘time stamped’ so become
Semantic semantic
 - Strength determined by the
strength of emotion when coding
 - Associated with the temporal
lobe + left pre-frontal cortex

 - Declarative
 - Personal experiences – e.g.
trip to the dentist
 - Complex – context, emotions,
time, place, details
Episodic  - Strength determined by
emotions when coded
 - Associated with the
Hippocampus + right pre-frontal
cortex

Procedural  - Implicit
 - How to perform tasks
 - Required repetition and
practice
 - Often acquired while young
 - Hard to verbalise
  - Associated with cerebellum
and motor cortex

AO3 – Brain scans

Tulving et al found that
when participants engage in
AO3 – HM
different activities engaging
Suffered severe epilepsy and
with different types of LTM
underwent surgery which involved the
they showed up in different
removal of his hippocampus. His STM
places on a PET scanner.
remained intact but was unable to
With the episodic being
transfer certain types of information
found in the right of the PFC
into his LTM.
and semantic being in the
left. Highlights the physical
Milner discovered that HM could lead
distinctions between the
implicit tasks but not explicit tasks as
types of memory.
he gradually improved on a block
building task but could not remember
But Blood Oxygen Level
doing it. His procedural memory was
Dependent signals
working but his episodic was not. This
assumption of brain scans
clear distinction furthers the idea that
make the assumption that
the different types of memory are not
activity in the brain equates
only distinctive but also localised to
to oxygen location. It cannot
different areas of the brain.
be established that the area
of the brain that is
associated with episodic and
procedural memory is
actually the location
AO3 – PM AO3 – issues with case studies
A cellist who suffered from Confounding variables Individuals
viral amnesia. Like HM, his Not generalisable to general
explicit memory was population
impacted but his implicit
procedural memory was not. Much of the research conducted by
He could play through whole cognitive psychologists into LTM is
songs on the cello, which using case studies. Though they
demonstrated how his provide idiographic and detailed
procedural memory was still descriptions of individuals memory,
working. A clear distinction the unstandardised and complex
is provided between nature of these studies limit there
different types of memories. ability to tell us about the general
processes of human memory.
Nomothetic laws cannot be supported
 through single studies.

Working Memory Model: Baddeley and Hitch (1974)

AO1 –
Basic assumption - STM is an active process
Central executive – directs attention to specific tasks and
determines which resources are allocated. Directs slave system.
small capacity and this cannot attend to many tasks at once
Visuo-Spatial Sketchpad – used for planning tasks e.g. going from
one room to another. Makes sense of visual information (what
things look like) and spatial information (how they relate to one
another). Within this is the inner eye consisting of the inner scribe
and visual cache (storage).
Phonological loop – deals with auditory information and preserves
the order of information. Made of two subcomponents: the
phonological store, which holds the words you hear; and the
articulatory process which is used for words that are heard or seen.
These are silently repeated in a form of maintenance rehearsal via
the inner ear and inner voice.
The Episodic Buffer integrates the information from other
components as well as LTM. It is a general store of STM. Limited
capacity but relatively larger than other stores.
AO3 – Dual Task Studies
B+H conducted dual task studies
in which participants had to
complete two tasks
In C1 – participants were required
AO3 – KF
to complete 2 acoustic based
PF, who was injured in a
tasks In C2 – participants were
motorcycling accident, was able
required to complete one auditory
to recall stored information from
and one visual task.
his LTM but had issues with his
Found that when required to
STM. He was able to transfer
complete two tasks with the same
visual images into his LTM, but
type of stimulus (e.g. condition 1)
not acoustic information.
performance was worse in
This supports the WMM as it
comparison to the when it
indicates that there are two
required different types of
distinct types of processing for
stimulus.
acoustic information and visual
and spatial information (the
Further supported through PET
phonological loop and the visual
scans which indicate different
spatial sketchpad).
areas of the brain are active when
doing tasks which require
However, this is only one case
processing from PL and VSS
study of a man who endured
significant head trauma. Perhaps
This can be explained through the
this was simply the way that his
WMM as parallel processing of
STM worked or perhaps his
visual and acoustic information
trauma led to the distortion of
can occur as the limited capacity
memory processes. This
of the slave systems are not being
idiographic study cannot be
competed over for attention.
generalised to create nomothetic
Support B+H’s distinction of the
laws of memory to all.
PL from the VSS.

RWA – cannot text and drive and

studying (no listening to music
with lyrics)
AO3 – vague and untestable AO3 – relies on lab studies
concepts Research into the WMM often
The role of the central executive displays experimental
 reductionism, as it attempts to
examine the complex nature of
human memory through isolated
variables operationalised in a lab.
This is boiling memory down into a
is vague and unclear, which has
purely cognitive process, the
led to criticism for the WMM.
psychodynamic aspect is removed
Baddeley even acknowledged
(repression).
this, saying ‘the CE is the most
important yet least understood
Not only is the study of memories
component of working memory.’
in laboratories lacking mundane
This is moving away from
realism due to the artificial nature
Popper’s idea of falsifiability.
of the tasks (e.g. dual processing
Removing
tasks where one must memorise
If the CE cannot be
sounds and images) which do not
experimentally measured, it is
reflect the stimulus of everyday
unfalsifiable. This limits the
STM; but the experimental
scientific basis of the WMM –
reductionism also means that the
making this paradigm widely
complex and dynamic process
rejected by science.
must be reduced to
operationalised variables.

This produces unrealistic and

unreflective results.

Explanations for forgetting: interference

AO1 –
Forgetting occurs in LTM when we cannot get access to memories
even though they are available. Interference is when these two
pieces of info are in conflict with one another.
Proactive interference – occurs when an older memory disrupts a
newer one. Old telephone means you cannot remember your newer
one Keppel and Underwood
Retroactive interference – occurs when a newer memory disrupts an
older one. Not being able to remember your first phone number as
your current one interferes with retrieval
Baddeley and Hitch
Interference is worse when memories are similar –
PI – previously stored information makes new info more difficult to
store RI – new information overwrites previous memories which are
similar McGoech and McDonald researched the effects of
similarities. In this study, participants were asked to learn a list of
words to 100% accuracy and then learn a list of one of seven
experimental conditions ranging in similarities from synonyms to
numbers. Found that level of similarity influenced the recall of
individuals
AO3 – Credibility
We can view these situations in
AO3 – Research support
everyday life – example. Further
McGoech and McDonald
supported by research which
researched the effects of
examines this in a real setting,
similarities. In this study,
giving it high ecological validity
participants were asked to learn a
(Baddeley and Hitch). B+H found
list of words to 100% accuracy
that the number of names rugby
and then learn a list of one of
players remember was
seven experimental conditions
associated with the number of
ranging in similarities from
games that they played.
synonyms to numbers. Found that
Provides evidence for retroactive
level of similarity influenced the
interference in a real life
recall of individuals
situation rather than a lab study
which assesses individual
variables through stimulated
artificial settings
AO3 – Lab studies issues
Artificial and contrived setting of
lab studies tells us little about the
effects of forgetting in real life
situations.
In real life experiences of
interference stimulus more like
people’s faces or birthdays, but
not lists of words (like that of
McGoech and McDonald’s study).
This use of artificial stimulus
increases the likelihood of
interference taking place. An
overemphasis on this theory may
have arisen as a result of this.

Moreover, Experimental
reductionism limits the factors
involved in order to establish a
causal relationship or association
between two variables. But this
may lower the validity as factors
which can influence memory are
controlled.

Explanations for forgetting: Retrieval Failure

AO1
Lack of cues can cause retrieval failure – when info is placed in
memory, associated cues are stored at same time. If cues are not
present at time of recall, then retrieval may not be possible
Encoding specificity principle – Tulving suggests that cues can help
retrieval if the same cues that are present at time of encoding are
present at time of recall
Some cues have a meaning link to memory while others do not:

 - Context dependent forgetting – when memory retrieval is

dependent on external environmental cues e.g. the weather
or the size of the room
 - State-dependent forgetting – when memory retrieval is
dependent on internal state of mind e.g. feeling upset or
being intoxicated

AO3 – Research support (CDF) AO3 – Discussion of evidence

Godden and Baddeley (1975)
18 divers took place in a study Control variables
which used being on land or G+B had little control over other
underwater as contextual variables in the study. For
external cues for examining example, the pools in which the
context dependent forgetting. divers completed the conditions
Found that in the conditions in in were different and there was
which place of encoding and of no standardised time that they
recall matched (e.g. being both were conducted in – providing
underwater or both on land) on different contextual cues. This
average was 40% higher than in influences the results as the
conditions when variance in
they did not. environmental contexts may
This supports Tulving’s have caused a difference in the
conclusions about ESP and CDF as ways that the memory is coded.
the information was not as readily
accessible in situations in which Experimental design – RM
the context did not match This may have led to demand
 characteristic or order effects.
E.g. recall improved or fatigue
effects lowered their scores
AO3 – Research support (SDF)
Cater and Cassady (1998)
AO3 – Discussion of evidence
By using anti-histamines, C+C
Support from similar research,
produced an sedative effect
like Goodwin et al who examined
which allowed individuals to learn
the influence of level of alcohol
and recall a list of words in one of
intoxication on SDF. Found
4 experimental conditions.
similar results to C+C. further
Similarly to G+B, when the states
emphasising the importance of
matched memory was
the role of the state in memory
significantly better than when
they did not
Artifical stimuli not reflective of
everyday memory
This supports the ESP provided by
Tulving
AO3 – applications
The concept of ESP and its influence on memory has high
explanatory power of real world situations of memory. By returning
to the context and emotional state of a situation we are more able
to remember things. This basic concept of retrevial failure makes up
a key aspect of the cognitive interview which has revolutionised the
effectiveness and accuracry of EWT by rooting it in psychological
principles. The stage of context reinstatement the EW recalls
contextual factors – like what they were wearing and what the
weather was like – as well as internal factors – like their emotional
state – to minimise the effects of CDF and SDF

Eyewitness Testimony: Misleading information - LQ and PED

AO1
Leading questions is a q that suggests a certain answer because of
how it is phrase. The question ‘was the knife in his left hand?’
suggests that the knife was in his left hand
Response bias explanation – the wording of a question has no
enduring effect of the memory of the event but influence the kind of
answer given Substitution bias explanation – wording of a question
does effect the memory of the event; interferes and distorts the
event

Loftus and Palmer:

45 participants were asked to watch film clips from a car accident
and answered questions about speed. Critical question ‘how fast
were the cars going when they
smashed/contacted/hit/collided/bumped?’ they found that the verb
given in the question directly influenced the speed at which they
reported the cars to be travelling: contacted – 31.8; smashed 40.5
The leading question affected the accuracy of the event with
smashed suggesting a faster speed.

Post event discussion – witnesses to an event discuss what they

have experienced, influence the accuracy of the memory recalled
Memory contraction – co-witnesses discuss a crime, mixing
misinformation from other witnesses and own memory

Memory conformity – witnesses go along with one another the win

social approval or as they believe other witnesses are right
Gabbert et al
Paired participants watched a video of the same crime, but from
different angles which meant that not everyone could have seen
the same exact event. In the co-witness variable, individuals
discussed the event before completing a self-report form.
71% of participants mistakenly recalled information from the event
that they did not see but had picked up in PED compared to no
mistakes in the control group
AO3 – RWA AO3 – Artificial tasks
By understanding the role of The artificial nature of the task
memory and how it is influenced, in L+P ang G et al’s research
important practical uses for police may mean that the event is not
officers and investigators can be applicable to everyday lives. In
implemented the events, individuals watched
For example, being careful to not the whole event through from
use LQ and take witnesses from start to finish as they were
the crime as quickly as possible to instructed on where to look by
prevent PED. Ask individuals to not Loftus.
discuss the event with others. Moreover, individuals were
Moreover, CI – reporting everything aware that they were in a safe
freely and openly about the event and secure experiment, a low
without the prompts of leading stakes environment.
questions leads to more accurate This is very different to people
information that a SPI (as who actually watch crimes
demonstrated by researchers such occur, who may not watch the
as Geiselman) whole event from start to finish
This is important as it means that and are in a sense of danger. As
 the individuals knew they were
in an experiment they may have
displayed demand
police officers will get more characteristics by which they
reflective EWT of the event and paid significantly more attention
therefore a better understanding of to the event that normal.
the crime – leading to less false All of these factors lower the
arrests ecological validity of the
findings of lab studies and
decreases their wider
application.
AO3 – population validity
In L+P the sample that was used
may be a confounding variable to
the variation in their answer.
As they were students, they would
have been less experienced drivers
and therefore less aware of how
fast looks like it is going.
This could account for the
variability in their answers more
than the question type.
Therefore the support for leading
questions influencing memory is
flawed.
EWT: Anxiety
AO1 –
Views anxiety as an influence into memory

Loftus – negative
Participants sat it waiting room believed that they were going to
participant in a lab study. Heard an argument in the next door room
followed by a loud clatter and a man crossing the waiting room,
here there were two conditions: low and high anxiety. Found
individuals in the low anxiety condition (where the man walked out
with a pen) were able to identify the man 49% of the time but in the
high anxiety (holding a bloodied letter opener) identified him 33%
This is explained by the tunnel theory of memory which argued that
the witness’ attention is directed onto the weapon in a phenomena
called weapon focus, focus on what it most scary to you.

Yuille and Cutshall – positive

Real life crime of a shop keeper shot dead in a robbery. 13
participants agreed to participant in the study they were
interviewed 4-5 months after the incident and these accounts were
compared to their initial testimony.
Found that their information remained very accurate up to 5
months after the crime and that participants who rated themselves
as most stressed were the most accurate (88% compared to 75%)
AO3 – Conflicting findings
The conflicting findings of J+S
and Y+C call into question the
basis of this theory. As Y+C
AO3 - CF explanations occurred in real world situations
Yerkes and Dobson argue that it has higher ecological validity
stress/arousal and performance compared to the artificial aspect
have a curvilinear relationship in of J+S’s experiment, which
an inverted U diagram some psychologists claim does
Deffenbacher applied this to not even measure anxiety and
memory suggesting that when EWs rather surprise. After conducting
experience anxiety greater than an a replication which involved
optimal point memory is drastically bananas, raw chicken, a
impacted. handgun etc, Pickles concluded
This could explain why the findings that EWT were more influenced
are so contradictory. by the unusual nature of the
item. Thus the phenomena of
weapon focus actually tells us
little about the effects of anxiety
on memory.
AO3 Compare – anxious
susceptible to leading questions Pickle experiment
and post event discussion Loftus

EWT: Cognitive Interviews

AO1 –
Fisher and Geiselman found issue with the short leading questions
asked by police officers resulting in inaccurate information which
would eventually lead to inaccurate EWT. Claimed that the EWT
could be improved by rooting it in psychological insights into
memory. Thus developed the cognitive interview.

The concept of schema is an important explanation of the

organisation of the human memory which relies on this concept,
simplified generalised representations of things based on our
experience. Memory capacity limitations prevents us from
remembering precise details about our every day lives, but
schemas help us overcome these limitations as we can summarise
the regularities in out lives.

But this means that we can mistakenly recall events that never
really happened as they make sense within our schema

Cognitive interviews work by overcoming this challenge through

these 4 key principles that they believed would improve recall:

1. 1) CR – context reinstatement: based on the principles of RF,

witnesses are encouraged to recall the context of the event
(e.g. the weather, their outfit, their emotions) as a way to
decrease the impacts of CDF and SDF on their recall. These
cues may act as a trigger to other memories
2. 2) RE – report everything: witnesses are encouraged to talk
freely and openly about the event, without the influence of
leading questions or time constraints which may pressure
them or alter the memories. Small details may also act as a
trigger for other memories

3. 3) CP – changing perspective: encourages the individual to

think differently about the same situation (for example, could
be asked to do this from the offender’s perspective). Aim is to
reduce the influence of schemas
4. 4) RO – reverse order: this avoids people using their schemas
so interviewers often ask them to recall in non-chronological
order as this way people are less likely to attempt to fill in the
gaps in their memory using schemas

Enhanced CI – Fischer (1987) developed further elements to CI

which emphasised the social dynamics (e.g. when to establish and
relinquish eye contact)
AO3 – Geiselman and findings AO3 – Practical implications
Geiselman (1986) used a sample of When interviewed about the
89 students who watched a practical applications of CI,
simulated crime and were police officers said that they
interviewed 2 days later using either often did not use them as
a standard police interview or a they found them time
cognitive interview. Found that there consuming and require
was an increase in overall amount of expensive and intensive
information recalled as well as training (highlighted by
amount of accurate information
recalled. (41.5 / 29.3). Highlights the
practical values that can be gained
from using the CI in investigations.
Kokhen).
But there was also an increase in the
As they are harder to conduct,
amount of inaccurate information
more expensive and time
recalled. This calls CI’s effectiveness
consuming, whilst yielding
into question further as it is
large amounts of inaccurate
supported by research such as
data dependent on skill of
Kohken’s meta-analysis. This found
interviewer, the applications
that while CI increase accurate
are limited and thus the CI
information up to 81%, inaccurate
may not always be
information is also detected in about
appropriate.
61% of cases. A correlation was
found between the level of accuracy
and the skill of the interviewer.
AO3 – Stigmatisation
AO3 – importance of CR and RE Milne CI have been found to have
and Bull found that CR and RE yield the greates effect on the
as accurate results as completing all elderly – indicating that
four steps. Indicates that Geiselman individual differences change
may have overstated the importance the accuracy of EWT, A study
of cognitive schemas in memory and found that between a 72-
year-old and a 22-year-old the
former had the greatest
improvement in recall
between a standard police
suggests that the cognitive interview interview and CI. Concluded
should be re-evaluated. This could that this was due to the
reduce timing and training (thus removal of negative stigma
cost) that it takes to conduct around the elderly being
cognitive interviews without associated with impaired
sacrificing the accuracy and results memory.
gained. This combats some negatives So, not only does it provide a
outlined by police officers, making safe, more accurate and
the cognitive approach more reliable location to remember
effective but also a more emotionally
safe environment. Improve
accuracy of EWT due to
environment that is created