Memory: Research Methods and Models

Rishita Kedia, Ganesh Ghadge, Sakshi Tambe

What is Memory?
Memory is the means by which we retain and draw on our past experiences to use
that information in the present. As a process, memory refers to the dynamic
mechanisms associated with storing, retaining, and retrieving information about
past experience. Specifically, cognitive psychologists have identified three
common operations of memory: encoding, storage, and retrieval. Each operation
represents a stage in memory processing.
● Encoding is the function by which information is coded in a form that
allows it to be stored in memory. Here, you transform sensory data into a
form of mental representation.
● Storage is the function by which information is retained in memory. Here,
you keep encoded information in memory.
● Retrieval is the function by which information is recollected as needed.
Here, you pull out or use information stored in memory.
Memory is an internal and unobservable process. At times we feel that we don’t
remember all that we had learnt earlier. In such a situation, we come face to face
with the imperfect nature of our own memories- our cognitive system for learning,
storing, and retrieving information and throwing a challenge for its assessment.
Measuring memory is linked to theories of the processes and structures involved in
memory, why something is remembered, and the goals of the study. Both the
content and the structure of memory can be measured. The credit for the first
systematic assessment of memory goes to Ebbinghaus (1900).
In 1885 Hermann Ebbinghaus carried out a study wherein provided a quantitative
measure of retention. For this purpose the designed "Nonsense syllable,” a
meaningless three letter unit consisting of vowel between two consonants such as
LEP, NOL etc. For obtaining a pure, uncontaminated by the previous learning and
association, he used these non-sense syllables. In these experiments he used
himself as a subject and as a result of these experiments, he designed many
methods for the training of memory.
Since then several studies have been reported in which various methods of
assessing (measuring) memory have evolved. The amount of information retained
in memory can be inferred from observable performance on various tasks. There
are both direct and indirect methods of measuring memory. The direct methods of
Memory: Research Methods and Models

assessing memory are: (i) recall, (ii) recognition, (iii) relearning, and (iv)
reconstruction. The indirect method focuses on the amount of transfer of previous
learning to a subsequent learning situation.

Direct Measures
Recall
The most widely used method of testing memory or measuring retention is the
method of recall. It is a passive, but direct method of measuring retention.
Reproduction of learnt materials after a time span is recall. It is actually repetition
of learned material, i.e., verbatim (word for word) recitation. For example, one
may recall a poem by reciting it even if he does not remember the circumstances
under which he had learned
Recall involves verbal reproduction or repetition of previously learnt material.
Recall becomes easier when materials are meaningful, interesting, short, and over
learned. In recall, you produce a fact, a word, or other item from memory.
Fill-in-the-blank and most essay tests require that you recall items from memory.
For example, suppose you want to measure people’s memory for famous actors.
You could ask people to name an actor. Three main types of recall tasks are used in
experiments (Lockhart, 2000): serial recall, free recall, and cued recall.
In serial recall the participant recalls the items in the exact same order in which it
is presented. For example, you could ask people to remember the following list of
actors in order: Sharaukh Khan, Amitabh Bacchan, Shahid Kapoor, Salman Khan,
Ayushmann Khurrana- and ask them to repeat the names in the list in that exact
order.
In free recall the participant can recall the items in the list in any order they please.
So if asked to recall the names from the above given list they can do so in whatever
order they remember it in.
In cued recall, the participants are first shown items in pairs, but during recall they
are cued with only one member of each pair and are asked to recall its mate. Cued
recall is also called “paired-associates recall” (Lockhart, 2000). For example, you
could ask people to learn the following pairings: Sharukh-Cabbage,
Amitabh-Carrot, Shahid-Capsicum, Salman-Potato, Ayushmann-Tomato, and then
ask them to produce the pairing for Salman (Potato).
Memory: Research Methods and Models

Ebbinghaus (1885) used recall method for studying remembering and forrgetting.
He was of the view that retention can be tested either by the method of (i)
immediate recall or, (ii) delayed recall, When recall is made immediate after
learning, it is called ‘immediate recall’; whereas, when recall is made after an
interval of time, it is called ‘delayed recall’. The failure to recall does not
necessarily mean that there has been no retention. Failing to recall despite retention
is known as recall amnesia. Sometimes due to emotional disturbances like fear,
anxiety and sudden shock, one fails to recall. Further, when the subject is
motivated to learn, she/he can recall the items easily. Thus, motive plays a major
role in recall.

Recognition
Recognition is considered as a sensitive method of measuring retention It is an
active process where identification of elements takes place. Recognition is a
common experience which refers to the fact that once the remembered event or
word is in front of us, we know that we have stored it away before is familiar to us.
Guilford (1917) viewed that recognition means knowing again Further, the
essential difference between recall and recognition is that in first case, the stimulus
is not there for one’s identification, whereas it is there in the second case. Thus,
recognition is the identification of learnt material object from a combination
(combined list) of learned and unlearned material. Increase of similarity between
original learning materials to that of the new materials, recognition becomes
difficult. There can be two kinds of error in recognitions tasks: Failure to recognise
familiar itemsand False cognition of new and unfamiliar stimulus. The primary
distinction between recall and recognition is that in recall, the stimulus is not there
for identification, whereas it is in there in recognition tasks. Recognition memory
is usually much better than recall A study by Standing and colleagues showed that
participants could recognize close to 2,000 pictures in a recognition-memory task
while best measured recall performance was typically around 80 items.
Recognition memory tasks tap into receptive knowledge. Recall memory tasks
require expressive knowledge. Informing participants of the type of future test they
will take can influence the amount of learning that occurs.

Relearning
Relearning is the number of trials it takes to learn once again items that were
learned in the past. This method, sometimes known as the ‘saving method,’ was
Memory: Research Methods and Models

developed by Ebbinghaus (1885) to assess the quantitative aspect of memory. In

this method, the subject is provided with a list of materials to perfect learning, then
after a time gap, she/he is presented with the same list to relearn. The number of
trials and time spent by the subject in the relearning condition are recorded by the
experimenter. The relearning effect was also observed in fetal rats. Assume you
studied French in high school but did not continue to study it in college. Now, for a
work, you need it to communicate effectively with consumers. When you relearn
French, you will save time in comparison to the first time you learnt it.

Reconstruction
Reconstruction entails reproducing the taught materials. Reconstruction technically
refers to rearranging the random components of an original activity. In this
procedure, the stimuli are presented in a specific order, then split up, and the
stimuli are handed over to the subject with instructions to rebuild the original order.
Reconstruction is often driven by background knowledge that suggests plausible
inferences. Suppose the parts of a plastic doll are joined to get a full figure of a
doll, then the experimenter breaks it into pieces and asks the subject to rearrange
the items to form a doll. If the subject can rearrange, then she/he gets the full credit
for the test. Such inferences may even lead us to believe we are remembering
something when we are not. A limitation to this method is that only materials that
are breakable into parts can be used.

Indirect Measures
Indirect tests are ones that require the subject to perform some type of cognitive or
physical action. The instructions are limited to the task at hand and make no
mention of previous events. The variables of interest represent a change in task
performance detected by comparing performance with and without relevant prior
experience. Indirect measures fall into four categories: Tests of factual,
conceptual, lexical, and perceptual knowledge, Tests of procedural knowledge (i.e.
skilled performance, problem solving), Measures of evaluative responseand Other
measures of behavioral change, including neurophysiological response and
conditioning measures.

Factual, conceptual and lexical tests

Memory: Research Methods and Models

Tasks in this category have traditionally been used by researchers attempting to

specify structures and processes involved in the retrieval of permanent knowledge.
In the factual and conceptual domains, subjects have been required to retrieve
items of general knowledge, generate members of a semantic category, generate
associates to stimulus words, verify category membership, and categorize or
classify stimuli. In the lexical domain, tasks include lexical decision, word naming
or pronunciation, word retrieval, word completion, fragment completion, and
spelling of auditorily presented homophone. Perceptual tasks include perceptual
identification of words, pictures, and faces, picture naming, and identifying
fragmented pictures. In these types of task the measures of interest are accuracy, or
latency of a correct response. Prior exposure to test stimuli generally increases
accuracy and/or decreases latency, a phenomenon known as direct or repetition
priming. In paradigms in which subjects are to generate any response they wish to
the test stimulus. Prior study of items increases the likelihood that those items will
be generated as responses at test. Changes in test performance are also observed
when information that is related to test stimuli is presented prior to test. The typical
example is the decrease in lexical decision latency consequent on presenting
associatively or semantically related words prior to the test stimulus, a
phenomenoknn own as associative or semantic priming. Effects of prior
presentation of stimuli that are graphemically and phonemically related to test
items have also been studied, as have effects of presenting morphologically related
words. Indirect priming is any change in performance resulting from the
presentation of information related in some way to test stimuli.

Procedural Knowledge
Procedural Knowledge investigates skill learning and problem solving, often
investigating improvements in performance as a result of task practise. Of
particular interest in the current context are studies of the development of cognitive
skills, such as proofreading and reading geometrically transformed text. Tasks used
to measure procedural knowledge typically include problem-solving tasks that
have been studied include solving jigsaw puzzles, the Tower of Hanoi puzzle,
rotary pursuit task, mirror tracing task and control tasks, such as learning rule
relating size of workforce to factory production output.

Evaluative response
Memory: Research Methods and Models

A number of researchers have investigated effects of exposure to stimuli upon

evaluative responses to those stimuli. The classic studies by Zajonc and coworkers
show that exposure to stimuli increases affective preference for those stimuli.
Cognitive judgments are also influenced by stimulus exposure; for example, prior
exposure to statements or components of statements increases the rated truth of
those statements.

Other Measures of Behavioural Change

Effects of prior exposure to stimuli can be revealed in changes in physiological
response, such as galvanic skin response and event-related potentials. Conditioning
measures, such as eye-blink conditioning, can also be included as indirect tests of
memory; measurements of interest simply reflect acquisition of a behavioral
response to an originally neutral stimulus. Similarly, Schacter & Moscovitch
(1984) have argued that habituation and novelty-preference paradigms used in
studies of infant memory can be conceptualized in terms of facilitated processing
of old (familiar) stimuli, rather than in terms of gaining access to information about
the prior occurrence of the familiar stimulus in the experimental context. Finally,
measures of savings in relearning repeated items qualify as indirect memory
measures, under the assumption that subjects are merely told to learn a list at test,
with no indication as to whether or not that list has been presented before.

The Traditional Model of Memory (Richard Atkinson & Richard

Shiffrin)
The model differentiates among structures for holding information, termed stores,
and the information stored in the structures, termed memory.
Atkinson and Shiffrin were not suggesting that the three stores are distinct
physiological structures. Rather, the stores are hypothetical constructs concepts that
are not themselves directly measurable or observable but that serve as mental
models for understanding how a psychological phenomenon works.
They proposed a model that conceptualized memory in terms of three memory
stores:
● Sensory store- capable of storing relatively limited amounts of information
for very brief periods;
Memory: Research Methods and Models

● Short-term store- capable of storing information for somewhat longer

periods but of relatively limited capacity as well; and
● Long-term store- of very large capacity, capable of storing information for
very long periods, perhaps even indefinitely.

Sensory Memory/Store
The sensory memory stores allow input from the sensory modalities (vision,
hearing, etc.) to be prolonged briefly in order for us to process relevant aspects of
that input. It is essentially a temporary sensory register, of large capacity, but which
fades rapidly. Sensory memory consists of a number of modality-specific stores:
the term iconic memory refers to the brief storage of visual stimuli; the term
echoic memory refers to auditory stimuli. Other stimulus types may also be stored,
such as haptic sensory memory for touch-related stimuli. The sensory store
prolongs sensory information so that we can attend to important parts of it; aspects
that are not attended to fade away.

Iconic Memory
The iconic store is a discrete visual sensory register that holds information for very
short periods. Its name derives from the fact that information is stored in the form
of icons. These in turn are visual images that represent something.
Sperling started out with a typical memory span experiment in which participants
were presented with a visual array showing, for example, three rows of four letters.
This was presented for a brief duration of 50 milliseconds. In a ‘whole report’
condition, participants were asked to recall as many items as they could. They
typically recalled about four or five items. However, verbal reports suggested that
the participant had seen more than could be reported. Sperling introduced a ‘partial
report’ condition, in which participants were asked to recall from only part of the
array. Immediately after presentation of the array (that is, on stimulus offset), a
tone was sounded to indicate which line the participants were to report from. A
high tone signaled that they should report what they had seen within the top line of
the array. A medium tone meant they should report from the middle line of the
array. A low tone meant they should report from the bottom line of the array.
Participants had no way of knowing in advance which line would be probed. Using
the partial-report procedure, Sperling found that participants could typically recall
about three items from each line; this meant that a much larger amount of
Memory: Research Methods and Models

information was available to participants than was suggested by the data from
whole report condition.
In subsequent work, participants were shown displays of two rows of eight
randomly chosen letters for a duration of 50 milliseconds. In this investigation, a
small mark appeared just above one of the positions where a letter had appeared (or
was about to appear). Its appearance was at varying time intervals before or after
presentation of the letters. In this research, then, participants needed to report only
a single letter at a time. The procedure thus minimized output interference. These
investigators found that when the mark appeared immediately before or after the
stimulus display, participants could report accurately on about 75% of the trials.
Thus, they seemed to be holding about 12 items (75% of 16) in sensory memory.
Sperling’s estimate of the capacity of iconic memory, therefore, may have been
conservative. The evidence in this study suggests that when output interference is
greatly reduced, the estimates of the capacity of iconic memory may greatly
increase. Iconic memory may comprise as many as 12 items.
Averbach and Coriell revealed an additional important characteristic of iconic
memory: It can be erased.The investigators found that when a stimulus was
presented after a target letter in the same position that the target letter had
occupied, it could erase the visual icon. This interference is called backward visual
masking. Backward visual masking is mental erasure of a stimulus caused by the
placement of one stimulus where another one had appeared previously. If the mask
stimulus is presented in the same location as a letter and within 100 milliseconds of
the presentation of the letter, the mask is superimposed on the letter. For example,
F followed by L would be E. At longer intervals between the target and the mask,
the mask erases the original stimulus. For example, only the L would remain if F
and then L had been presented. At still longer intervals between the target and the
mask, the mask no longer interferes. This non-interference is presumably because
the target information already has been transferred to more durable memory
storage.

Echoic Memory
Echoic memory is the auditory equivalent of iconic memory; it is sensory memory
for heard information. The Darwin et al. study involved presenting auditory stimuli
independently to each ear, or to both ears, using stereo headphones, such that the
sounds would be heard from three spatial positions: from the left, right or from the
‘middle’ (i.e. in stereo). Nine letters and nine digits were used to form sequences;
three items were presented to the left channel, three to the right, and three were
Memory: Research Methods and Models

presented simultaneously in stereo. They were presented such that the first item of
each group was heard simultaneously; similarly, the second items were presented
simultaneously, and then the third. Following Sperling’s procedure, the auditory
stimuli were presented and, after a delay that varied from 0 to 4 seconds, a cue
indicated from which set the participants were to report. In this case, a visual cue
was presented to the left, middle, or right and the participants reported what they
had heard from the corresponding location. Consistent with Sperling’s findings on
iconic memory, Darwin and colleagues found that performance in a partial-report
condition suggested a large initial memory of auditory information, which decayed
rapidly. At zero delay participants could report about five of the nine items. After a
delay of 4 seconds, performance had dropped to 4.25 items on average, the same
number as would be expected in a whole-report condition. This suggested that
there was a sensory store for auditory information that was similar in some ways to
iconic memory; this became know as the echoic store. The echoic store provides an
acoustic register, allowing auditorily presented information to be prolonged so that
some aspects of the input can be retained for processing. The auditory
partial-report data are broadly consistent with findings using the shadowing
technique, in which participants must ‘shadow’ or repeat back a message presented
to one ear or the other. For example, Glucksberg and Cowan had participants
shadow a passage of prose that was presented in one ear while another prose
passage was presented to the other, ‘unattended’ ear. Participants were to ignore
the second passage, but were warned that digits would occur in that text from time
to time, and that when a light flashed, they were to report the last digit heard in the
unattended message. The duration between the presentation of the digit and the
light cue was manipulated. Performance on the task deteriorated at about 4
seconds. Similarly, Treisman found that if participants shadowed a message while a
second unattended message was presented, they only recognized that the two
messages were the same if they occurred within about 2 seconds of each other.
Sensory memory is fragile and can easily be disrupted before stimuli can be
transferred into short-term memory (STM). Backward masking procedures involve
the presentation of a ‘masking’ stimulus immediately after the target stimulus; for
example, a briefly presented visual stimulus (e.g. a letter) might be followed by a
row of hash marks (####). The participant is subsequently required to identify the
letter in a recognition test. Recognition increases as the duration between the
presentation of the target stimulus and the masking stimulus (the stimulus onset
asynchrony or SOA) increases, to about 250 milliseconds.

Haptic Memory
Memory: Research Methods and Models

Haptic memory is the sensory memory for stimuli sensed through touch. Support
for haptic sensory store was provided in a study by Bliss et al. who used a tactile
version of Sperling’s partial-report procedure. Their participants were trained to
associate a letter of the alphabet with three sections on each of four fingers of one
hand. Participants then placed their hand in a device that administered a puff of air
to some of these regions, and had to report which regions had been stimulated by
giving the associated letter. In the partial-report condition, a visual stimulus cued
whether participants were to report stimulation presented to the upper, middle or
lower sections of the fingers. A small advantage for the partial-report condition
was found, as long as the visual cue appeared within 800 milliseconds of
termination of the tactile stimulation.
While sensory memory provides a temporary register that is rich in sensory detail,
such memory is short-lived and cannot be manipulated.

Short-Term Memory/Store
Short-term memory (STM) holds information in consciousness; it provides
temporary storage of active information. William James’s description of short-term
memory as primary memory equated it with the psychological present, the
information that is available in consciousness. STM has a limited capacity, and
information can be lost from it relatively easily. Typically, material remains in the
short-term store for about 30 seconds, unless it is rehearsed to retain it.
We will talk about the coding and capacity of STM. Coding refers to the way in
which information is mentally represented-that is the form in which the
information is held. R. Conrad in his study presented the participants with a list of
consonants for later recall. Even though the letters were presented visually it was
seen that participants were likely to make error that were similar in sound to the
original stimuli. It was seen that participants apparently form a mental
representation of the stimuli that involved acoustic rather than visual properties.
George Miller in his classic paper tried to understand the capacity of our STM. He
presented participants with a string of units: words, letter, number and so on. It was
seen that participants could only recall upto 7 plus or minus 2 units depending on
the individual, the material and other situational factors. After coding come the
retention duration. It refers to the length of time of the memory. It can be
explained with the Brown-Peterson task in which participants were presented with
a three consecutive trigram such as BKG and when asked to recall after a set
interval it was seen that only 80% could count backward for 3 seconds and only
7% could count backward when asked to recall after 18 seconds. This implied that
Memory: Research Methods and Models

information which is not rehearsed-decays or breaks apart within 20 seconds.

Some cognitive psychologists challenged this finding by saying that it was
interference and not decay. Interference can be explained as information which can
displace some other information making it harder to retrieve the former
information. After all of this comes the retrieval of information. Sternberg lined
out various possibilities on how information might be retrieved from STM.
Sternberg hypothesised that information might be held in a parallel or a serial
manner. He gave participants a memory set of movie titles and asked them to recall
a particular movie from the list. In parallel search it was seen that no matter the
number of movies one examined everything at the same time. In serial search one
compares the recalled stimulus with each and every stimulus in the list given.
Further we need to understand if these searches are either self terminating or
exhaustive. A self-terminating search stops when a match is found so once when
you recall the movie we stop search in the movie list. Another type of search is
exhaustive search which means that even if a match is found you continue looking
through every other item in the set which means you continue to look for the movie
even when the movie is found.

Long-Term Memory
William James (1890) made the distinction between short-term memory and
long-term memory. He described long-term memory as memory where an object
which has been recollected is one which has been absent from consciousness
altogether and now revives anew. It is brought back, recalled, fished up, so to
speak, from a reservoir in which, with countless other objects, it lay buried and lost
from view. When we call something to mind, we are using short-term memory, but
all of the memories that we have, whether we are currently thinking of them or not,
are stored in LTM. There are two types of Long Term Memory.
Two types of long term memory- Declarative and non-Declarative
Declarative Memory
1) Episodic memory
2) Autobiographical memory
3) Semantic memory
Memory: Research Methods and Models

Non-Declarative memory
1) Skill learning
2) Habit learning
3) Repetition learning

Declarative Memory
Declarative(explicit) memory refers to situations that we would generally think of
as involving memory, both for specific events such as remembering facts or
information about the world. It is open to intentional retrieval, whether based on
recollecting personal events(episodic memory) or facts(semantic memory).

Episodic Memory
Memory for specific events that occurred at a particular time and place is known as
episodic memory. Episodic memory is the system
within LTM that allows us to remember our past experiences; it enables us to
consciously re-experience past events. Tulving notes three key properties of
episodic memory. First, it is associated with our subjective sense of time that
allows use to engage in ‘mental time travel’ – we can, in the mind, ‘travel back’ to
remember past events. The phrase ‘that takes me back’ reflects this property of
episodic memory and is, Tulving argues, uniquely human. Second, there is a
connection to the self. Self-reflection is a key aspect to this kind of memory as he
puts it,’mental time travel requires a traveler. Third, mental time travel is
associated with a special kind of conscious awareness called autonoetic
(self-knowing) consciousness. This type of consciousness allows us to imagine
ourselves in the future, and to plan ahead and set goals. It allows us to recall a past
incident and imagine how we might have behaved differently. Episodic memory is
not simply a record of past experiences; memories are constructed anew when they
are called to mind, and can differ from the original event, and with each recall of
the event. Bartlett in a study would present a story called The war of the ghosts.
After reading the story the participants were asked to recall the story as accurately
as they can. It was found that the remembered story was always shorter, more
coherent, and tended to fit in more closely with the participant’s own viewpoint
than the original story. One of the features of bartlett approach was the participants
Memory: Research Methods and Models

effort after meaning. It meant that rather than being a simple recipient of
information, participants were actively striving to discern the meaning of stimuli,
trying to capture the essence of the material presented. A second feature of
Bartlett’s theory was his postulation of the concept of a schema. He defined the
schema as ‘an active organization of past reactions, or of past experiences.
Schemas are organized memory structures that allow us to apply past experience to
novel situations so as to guide behaviour. Bartlett emphasized the role of social and
cultural influences on the development of schemas, which in turn determine the
way in which material is encoded, stored, and subsequently recalled. These
tendencies were especially great with a story like The War of the Ghosts, in which
several features were incompatible with european expectations. He interpreted his
findings by arguing that the systematic errors and distortions produced in the
participants’ recalls were due to the intrusion of their schematic knowledge.

Autobiographical Memory
Autobiographical memory refers to the memories that we hold regarding ourselves
and our interactions with the world around us that help to define who we were at
different times in the past, who we are currently, and who we hope to be in the
future. It involves personal experience and it is closely associated with the self. It
includes not only episodic memories that form part of our life stories, but also
semantic autobiographical memory that includes historical facts, traits, and
knowledge states that are not unique to any particular place or time.
Autobiographical memory is clearly important to each of us, as it helps to define
our sense of ourselves as people.
Conway proposes four functions of autobiographical memory. These include
directive functions, for example what happened the last time you tried to change a
car tire, and a more social function; sharing autobiographical memories can be a
very pleasant and socially supportive activity. These memories can also play an
important role in creating and maintaining our self-representation function by
using personal memories to create and maintain a coherent self-identity over time.
Autobiographical recollection can be used for emotion regulation, as when we
need to cope with adversity, or build confidence. Recalling positive personal
experiences can be used to maintain desirable moods or alter undesirable moods.
These highly personal memories are not free from bias either.
Memory is a reconstructive process, and when we recall life events we reconstruct
or interpret the memory ‘record’ rather than play it back passively. An
Memory: Research Methods and Models

experimental demonstration of such false memories was provided by loftus Her

procedure was designed to increase the probability with which participants would
report a fictional event as having occurred, in essence to plant a false memory that
was ‘at least mildly traumatic, had the experience actually happened’. Loftus had
her participants recall childhood events that had been provided by close family
members, who acted as confederates in the experiment. Three of the events had
actually happened, one had not, but was a reasonably plausible childhood event – a
shopping trip when the participant was five years old during which he/she had
become separated from family members. The ‘lost-in-the-mall’ scenario contained
elements that were plausible: the child was separated from family and lost for an
extended period of time; he or she was upset and crying; an older woman helped
the child; and he or she was subsequently reunited with the family. Participants
were invited to provide as much detail as they could about the four events.
Immediately after presentation of the scenarios, participants recalled, on average,
68 per cent of the true events. But close to a third of participants reported
remembering the false event, and a quarter of participants continued to report
recollection of the false event during two follow-up interviews.
Research has also shown that imagining false events increases the likelihood that
they will be ‘recalled’.Mazzoni and Mamon conducted an experiment using a false
event which, while plausible, was not possible. They had their British participants
consider two events; one was relatively common (having a dentist extract a tooth);
the other described a medical test that is not conducted in the United Kingdom
(having a nurse remove a skin sample from the little finger). They found that
imagining the false event increased the number of reported memories and belief
that the event had taken place, an effect referred to as imagination inflation.
Imagination inflation refers to strengthening of a false memory through repeated
retrieval.
The data on disputed memory seems to be consistent with another unusual memory
phenomenon called deja vu. Déjà vu is a type of illusion of autobiographical
memory; it might be described as the ‘knowledge that a situation could not have
been experienced, combined with the feeling that it has. Brown and Marsh propose
three possible mechanisms for déjà vu. The first is split perception; we get a brief
glimpse of a visual scene before becoming fully aware of the scene. A second
mechanism is implicit memory; we have already experienced the scene or part
thereof but it has been stored such that only a feeling of familiarity is elicited when
we re-encounter it. The third mechanism involves the notion of gestalt familiarity,
that overall configuration of the present scene closely resembles a scene that we
have encountered in the past, though the specifics are different. We therefore
experience a sense of familiarity without being able to put our finger on why.
Memory: Research Methods and Models

Semantic Memory
Semantic memory is our store of general knowledge about the world, the people in
it, as well as facts about ourselves. It includes our knowledge of facts, language,
and concepts. Binder and Desai provided a detailed definition of semantic
memory: “It is an individual’s store of knowledge about the world. The content of
semantic memory is abstracted from actual experience and is therefore said to be
conceptual, that is, generalized and without reference to any specific experience.
We know from studies of amnesia that much of semantic memory remains intact
and available even after a brain injury affecting memory. Our knowledge about
language and concepts about the world tends to be formed early on in life and to be
used throughout life. Bahrick examined memory for Spanish learned at school in a
large sample with up to a 50-year retention period. Participants were tested on
measures of recall, recognition and comprehension. The data showed an initial
sharp decline in retention over a six-year period, following which the remaining
memories became stable and there was no further loss of knowledge until up to 25
years later. Some further forgetting occurred beyond the 30-year period. Once
knowledge has stabilized in semantic memory, it remains resistant to forgetting
over potentially a very long period; this long-lasting store of knowledge was
referred to by Bahrick as the permastore. Permastore involves the long term
retention of content that has been acquired and relearned over a period of time,
even if rarely used thereafter. Further we explored various theories given by
different theorist explaining semantic memory.

Hierarchical Network Theory(Collins and Quillian)

The first systematic theory of semantic memory was put forward by Collins and
Quillian. Their key assumption was that semantic memory is organized into a
series of hierarchical networks. The major concepts (e.g., animal; bird; canary) are
represented as nodes, and properties or features (e.g., has wings; is yellow) are
associated with each concept. The underlying principle is that property information
is stored as high up the hierarchy as possible to minimize the amount of
information needing to be stored in semantic memory. According to the model, we
can decide very rapidly that the sentence, “A canary is yellow,” is true because the
concept (i.e., canary) and the property (i.e., is yellow) are stored at the same level
of the hierarchy. In contrast, the sentence “A canary can fly,” should take longer
because the concept and property are separated by one level in the hierarchy. The
Memory: Research Methods and Models

sentence, “A canary has skin,” should take even longer because there are two
levels separating the concept and property. There are some limitations of this
theory too. First Conrad discovered people are slow to verify sentences such as,
“A canary has skin,” because it is very unfamiliar rather than because of the large
hierarchical distance between the concept and its property. Second, the statements,
“A canary is a bird,” and “A penguin is a bird.” According to the theory, both
statements should take the same length of time to verify, because they both involve
moving one level in the hierarchy. In fact, the latter statement takes longer because
canaries are much more typical or representative of the bird category than
penguins. Rips, Shoben, and Smith found verification times were faster for more
typical or representative category members than atypical ones. This is the
typicality effect. Third Collins and Quillian mistakenly assumed the concepts we
use belong to rigidly defined categories. The evidence found was contrary.
McCloskey and Glucksberg gave 30 people tricky questions such as, “Is a stroke a
disease?” and “Is a pumpkin a fruit?” They found 16 said a stroke is a disease, but
14 said it was not. A pumpkin was regarded as a fruit by 16 participants but not as
a fruit by the remainder. When McCloskey and Glucksberg tested the same
participants a month later, 11 of them had changed their minds about “stroke”
being a disease, and eight had altered their opinion about “pumpkin” being a fruit!
Verheyen and Storms (2013) identified two reasons for individual differences in
deciding which items belong to a given category. First, there is
ambiguity—individuals may use different criteria for categorization (e.g., is
strenuous activity a necessary criterion for something to be regarded as a sport?).
Second, there is vagueness—individuals may use different cut-offs to separate
members from nonmembers. For example, two individuals may agree that being
strenuous is a criterion for an activity being a sport but may disagree about how
strenuous it must be.

Situated Simulation Theory (Barsalou)

Barsalou rejected all the theoretical assumptions made by the traditional approach.
He argued we rarely process concepts in isolation. Instead, we process them in
various settings with that processing being influenced by the current context or
setting. More generally, our concept processing is influenced by our current goals
and the major features of the situation.
Baslow explained his theory with the help of the concept of bicycle. According to
Barsalou, those aspects of the bicycle concept activated depend on your current
goals. For example, information about the tires will be activated if you need to
Memory: Research Methods and Models

repair your bicycle, whereas the height of the saddle will be activated if you want
to ride it. In sum, Barsalou’s situated simulation theory makes various predictions.
Of particular importance, it predicts that conceptual processing involves extensive
use of the perceptual system and the motor or action system. Concept processing
can have a perceptual or imaginal quality about it. Object qualities not visible if
you were actually looking at the object itself are harder to think of than those that
would be visible. Although it rejects most of the traditional assumptions for
semantic memory it has it own limitations. First, he exaggerates the extent to
which concept processing varies across time and across situations or contexts. The
traditional view that concepts possess a stable, abstract core has not been disproved
by Barsalou. Second, much of our concept knowledge does not consist simply of
perceptual and motor features. Borghesani and Piazza give the following example:
“Tomatoes are native to South and Central America.” Third, we can recognize the
similarities between concepts not sharing perceptual or motor features. For
example, we categorize watermelon and blackberry as fruit even though they are
very different visually and we do not eat them using the same (or similar) motor
actions. Fourth, the finding that concept processing often includes perceptual
and/or motor features does not mean it is generally necessary to use perceptual
and/or motor processes to understand concepts. The finding that some patients with
damage to their motor system can nevertheless understand action-related words.

Non-Declarative Memory
Nondeclarative memory (sometimes called implicit memory) is a component of
long-term memory devoted to “knowing how” to do something and is expressed
through performance and does not afford access to any conscious memory content.
Such memories are essentially the same as motor skills that have been practiced
severally to the point that they can be performed automatically without conscious
thought. Playing the piano, knowing how to produce a clear tone on a wind
instrument, knowing how to ride a bicycle, knowing how to jump rope, and the
likes are examples of nondeclarative memory. These memories are acquired
through experience and learning just like declarative memories. Thus one may
know how to produce a clear tone on an instrument, but not be able to tell anyone
else how to do it. For example, bicycle riders remember how to balance a bicycle
even though they cannot declaratively retrieve that information. Because
nondeclarative memory operates subconsciously, its functions do not demand
attentional resources and working memory capacity. The groups of neurons that
process these memories are probably not connected to the groups of neurons that
process language; nondeclarative and declarative memory are thought to reside in
Memory: Research Methods and Models

different parts of the brain. Although nondeclarative memory cannot be tested by

asking examiners to recall information, the existence of nondeclarative memories
can be documented by observing performance or a change in behavior. Further we
will talk about some of the types of non-declarative memory.

Skill Learning
Procedural memory refers to the ability to perform learned actions or skills. It is
“knowing how” to do something. It includes habits and skills that accumulate
slowly through practice and repetition. We know how to cut our food with a knife
and fork, how to walk, how to read and write, how to do long division, how to peel
a banana, how to ride a bicycle. And we can do these procedures without conscious
effort. Although conscious processes are involved in the early stages of most
procedural learning, and the training can generally be recalled, procedural learning
can happen without the ability to recall the training. Even though individuals can
access the steps involved in a learned procedure or skill, conscious retrieval is not
necessary for performance of the skill. Procedural learning includes cognitive,
perceptual, motor, and other types of learning. For example, it involves learning to
blend phonemes into smoothly pronounced words or learning to dribble and shoot
a basketball. Procedural memory accrues through experience and is evidenced by
changes in behavior. Conscious awareness of procedural memory is not necessary
for accurate performance of a skill. For instance, the automated and reflexive
actions involved in driving an automobile depend on procedural memory.
Procedural memory accumulates slowly through practice and repetition; and once
it is established, it usually lasts a lifetime. For example, people never forget how to
ride a bicycle, and even individuals with advanced Alzheimer’s retain their
procedural knowledge.

Habit Learning
Habit learning refers to memory acquired over time through repeated associations
between stimuli and responses. This type of memory has been studied extensively
in experimental animals but it remains poorly understood in humans because of the
difficulty in eliminating the influence of conscious (declarative) memory on the
learning situation. One set of tasks that has been used to investigate habit learning
in the absence of input from declarative memory involves probabilistic
Memory: Research Methods and Models

classification learning. In such tasks, participants are required to learn a set of

associations. The associations are not obvious, and they cannot be readily
memorized, because of the probabilistic nature of the associations between
stimulus and response. The participant has to use information gleaned across many
trials to complete the task successfully. In one such study, Squire and Zola had
participants complete a weather prediction task. On each trial, participants had to
predict a weather outcome (rain or sunshine) based on one, two, or three cues (out
of a total of four possible cues) that were presented. Cues consisted of cards
featuring squares, triangles, circles and diamonds. Each cue was associated with a
weather outcome with a fixed probability, with associations of 75 per cent, 57 per
cent, 43 per cent or 25 per cent. When more than one cue was presented on a trial,
outcome was predicted by conjoint probabilities. Participants responded by
pressing a key to select a weather outcome, and they were given immediate
feedback as to whether their response was correct or incorrect. Participants could
not rely on memory for previous trials to learn the task, because the same
configuration of cues had the potential to generate different outcomes, and so this
ensured that declarative memory was not involved in the task. Typically on this
task, participants’ performance improves from guessing (50 per cent) to about 70
per cent correct. Patients with amnesia typically learn the task at about the same
rate as controls, with similar response accuracy (about 65 per cent) after training
over a 50 trial block. Patients with amnesia, while able to perform almost as well
on controls on the task, remain unable to report factual details about the training
episode, however

Repetition Priming
The term priming refers to the facilitatory effect of previous exposure to a stimulus
on the subsequent processing of that stimulus or a related stimulus. Priming can be
conceptual or perceptual, depending on whether it is the stimulus form or the
stimulus meaning that is salient. Examples of perceptual priming tasks include
word fragment completion, word stem completion, and identification of degraded
stimuli or stimuli presented at the visible threshold. Conceptual priming tasks
include category exemplar tasks and word association. An example of a priming
task is provided by Tulving et al. Their participants learned a list of low frequency
words (e.g. toboggan, theorem, pendulum). The words were presented singly on a
screen at a rate of one every 5 seconds. Participants were instructed to look at each
word and they were told to ‘Do the best you can to learn each of the words as they
appear, as you will be tested later for your memory of them.’ Participants were
tested after one hour and again after one week. There were two types of test, a yes-
Memory: Research Methods and Models

no recognition test (declarative memory) and a fragment completion test

(non-declarative memory). In the fragment completion test, participants were given
incomplete word fragments, with dashes indicating the position of missing letters
(e.g. _ob_gg_ _ ) and they were asked to complete the word by replacing the
dashes with letters. Half were words from the target list, and half had not been seen
during the learning phase. The results showed that performance was facilitated for
fragments of words that had been presented during the learning phase. Tulving et
al. also compared performance on the fragment completion task with that on the
recognition task, thereby comparing implicit and explicit memory. They noted that
recognition was not better for fragments that had been successfully identified
compared to those that had not been identified, supporting the notion that fragment
completion requires non-declarative, but not declarative, memory.

Levels Of Processing Model

The deeper the level of processing, the higher is the The Levels of Processing
model, created by Fergus I.M. Craik and Robert S. Lockhart. Level of processing
focuses on the depth of processing involved in memory and predicts that, the
deeper the information is processed, the longer a memory trace will last. It
postulates that memory does not comprise three or even any specific number of
separate stores, but rather varies along a continuous dimension in terms of depth of
encoding.It is the probability that an item may be retrieved. There are three
different levels of processing – physical, phonological and semantic.
Physical:- Words are learnt by remembering their physical features.For.eg flower.
Phonological:- Words are learnt by their sound. For.eg, rhyming
Semantic:- Words are encoded by their meaning.
The benefits of level of processing can be seen for a variety of populations,
including people with schizophrenia. People suffering from schizophrenia often
suffer from memory impairments because they do not process words semantically.
Deeper processing helps them in memory. The self-schema is an organized system
of internal cues regarding our attributes, our personal experiences, and ourselves.
One criticism of the level-of-processing theory is that it is difficult to distinguish
the depth or benefits of one level from another. It does not explain how the deeper
processing results in better memories. Deeper processing takes more effort than
shallow processing and it could be rather that the depth of processing that makes it
more likely people will remember something. The concept of depth is vague and
Memory: Research Methods and Models

cannot be observed. Eysenck claimed that the levels of processing theory describes
rather than explains.The ideas of 'depth' and 'elaboration' are vague and ill defined.
As a result, they are difficult to measure. Indeed, there is no independent way of
measuring the depth of processing. However, recent studies have clarified this
point; it appears that deeper coding produces better retention because it is more
elaborate.

Working Memory
Working memory holds only the most recently activated, or conscious portion of
the long-term memory. It usually combines the temporary storage and also
manipulates the information for cognition . Alan Baddeley has suggested an
integrative model of memory. It synthesizes the working-memory model with the
LOP framework. However it is said that the LOP framework is an extension,
rather than as a replacement for, the working-memory model. He suggested that the
working memory comprises of five elements such as, visuospatial sketchpad, the
phonological loop, the central executive, subsidiary slave systems, and the episodic
buffer.

Visuospatial Sketchpad:-
Visuospatial sketchpad is a component of working memory model which stores
and processes information in a visual or spatial form. It is an important element of
the working memory. The visuospatial sketchpad also allows us to recreate images
either based on something we’re seeing in real time or something we’ve seen in the
past. The whole system is controlled by the central executive.

Phonological Loop:-
The phonological loop is a component of working memory which deals with
written and spoken materials. That means it deals with the inner speech for verbal
comprehension and for acoustic rehearsal. The phonological loop is subdivided
into two parts. One is the phonological storage, which holds information in the
memory, and the other is the articulatory control process, which processes speed
production and also rehearses and stores verbal information from the phonological
loop. Articulatory suppression is more pronounced when the information is
presented visually versus aurally.
Memory: Research Methods and Models

Central Executive:-
Central executive is the most important component of the working memory
model.It drives the whole system and allocates the data to the sub system. However
it is responsible for monitoring and coordinating the operation of the slave
systems(the phonological loop and the visuospatial sketchpad). Its main mode of
operation was proposed by Norman and Shallice (1986). Central executive
involves higher-order reasoning and comprehension and is central to human
intelligence.

Subsidiary Slave System:-

Subsidiary slave system performs other cognitive or perceptual tasks. The slave
system are short-term storage systems. It is considered to have a limited capacity
system that provides temporary shortage of information by conjoining information
from the subsidiary system. The central executive which acts as a supervisory
system and controls the flow of information from and to its slave systems: the
phonological loop and the visuo-spatial sketchpad. Baddeley & Hitch's argument
for the distinction of two domain-specific slave systems in the older model was
derived from experimental findings with dual - task paradigms. The dual-task
paradigm is a procedure in which subjects are asked to perform two behavioral
tasks concurrently, each of one involves a distinct goal with a unique
stimulus-response association.

Episodic Buffer:-
The original model of the working memory was updated by Baddeley after the
model failed to explain the result of various experiments. So an additional
component was added known as the episodic buffer.The episodic buffer is the third
member of the slave system. An episodic buffer has a limited capacity storage
system which is responsible for integrating information from several sources to
create a unified memory, sometimes referred to as a single 'episode'. It is mostly
said that the episodic buffer is situated in the left anterior hippocampus. The
function of the episodic buffer is to learn, store and retrieve information about
unique personal experiences that occur in daily life.
Neuroscience And Working Memory.
Memory: Research Methods and Models

Neuropsychological studies have shown abundant evidence of a brief memory

buffer. The buffer is used for remembering information temporarily. Neurological
methods, especially neuro imaging can be very helpful in understanding the nature
of memory. However the field has been dominated by neuroimaging studies based
on healthy human participants.
Neuroimaging Working Memory
The initial studies used Positron emission tomography (PET) techniques, which
involves introducing a radioactive substance into the bloodstream and using it to
monitor the amount of activity occurring in different brain regions. It has found
evidence for distinct brain areas involved in the different aspects of working
memory. They identified two separate regions, one in the area between the parietal
and temporal lobes of the left hemisphere, which appeared to be responsible for
phonological storage. The second one is the more frontally based region known as
Broca’s area, which is involved in speech production, that appears to be linked to
subvocal rehearsal. Jon Jonides and Edward Smith at the University of Michigan,
further extending the use of neuroimaging to investigate working memory. They
gave the first direct comparison on visual and verbal working memory. Further
studies observed a distinction between spatial working memory and memory for an
object or shape. Spatial memory activates in the upper region of the brain whereas
object memory tends to be more concentrated on lower or ventral area. The central
executive function appears to involve activation mostly in the frontal lobes. A very
different but crucial role of the central executive and of the frontal lobes is to
energize and drive behavior. This capacity was studied by Alexander, Stuss,
Shallice, Picton, and Gillingham.

Intelligence And Working Memory.

Working memory, in particular, is strongly correlated with intelligence in children
and adults. Recent studies show that intelligence is a critical part of the working
memory. The ability to store and manipulate information in working memory may
be an important aspect of intelligence. Information processing, problem solving
and learning are often related or connected to intelligence. Fluid intelligence
involves our current ability to reason and deal with complex information around
us. It involves comprehension, reasoning and problem solving. Other researchers
have demonstrated that there is a significant relationship between working memory
and intelligence.
Memory: Research Methods and Models

Multiple Memory System.

The multiple memory system states that the brain stores information based on the
independent and parallel activity of a number of modules, each with distinct
properties, dynamics, and neural basis. The multiple memory system theory is
based on evidence that different kinds of information are processed and stored in
different parts of the brain. This system suggests that there are at least two separate
explicit memory systems. One would be for organizing and storing information
with a distinctive time referent. The second system would be for information that
has no particular time referent. Elvan Tulving gives a distinction between general
knowledge (semantic memory) and memory for events (episodic memory).
Semantic memory: It is a long-term memory that involves the recollection of ideas,
concepts and facts commonly regarded as general knowledge. Semantic is the
study of how meaning is stored in the mind. It is our memory for facts that are not
unique to us and that are not recalled in any particular temporal context. Episodic
memory: It refers to the conscious recollection of a personal experience that
contains information on what has happened and also where and when it happened.
Basically it is the memory of everyday event that can be explicitly stated or
conjured. According to him, we use episodic memory when we learn lists of words
or when we need to recall something that occurred to us at a particular time or in a
particular context.

Conclusions:
Memory is the means by which we draw on our knowledge of the past to use the
knowledge in the present. It is a basic cognitive process used in almost every
cognitive activity. It involves encoding information, storing it and later retrieving it
from that storage. Cognitive psychologists consider memory as an active,
constructive process. This means that information does not “sit still” in a
storehouse, waiting to be retrieved, but instead it is elaborated and sometimes
distorted or constructed.

References:
Kathleen M. Galotti(2014). Cognitive Psychology In and Out of the Laboratory
(5e)
ISBN: 978-93-515-0277-7
Memory: Research Methods and Models

Ken Gilhooly, Fiona Lyddy and Frank Pollick(2014). Cognitive Psychology

ISBN: 100 077122666
Alan Baddeley, Michael W. Eysenck & Michael C. Anderson. Memory (3e)
ISBN: 978-1-138-32607-1
Robert J. Sternberg, Karian Sternberg (2009) Cognitive Psychology (6e)
ISBN: 978-1-111-34476-4