9 the remembering brain Flashcards
STM
memory for information currently “in mind”; limited capacity
LTM
LTM = stored information that need not be presently accessed or even consciously accessible; has virtually unlimited capacity
Hold about facts or events
Minutes hours days years ago
Resembles hard drive - able to access if need
• All information from minutes, hours, days and years ago
is in LTM, unless it happens to be presently brought to
mind (in which case it enters STM)
what is working memory
Term WM proposed as it captures better the idea that the information currently in mind is manipulated
Working memory plays a wider role in cognition for example speech comprehension that can be accomplished only by manipulating the information
Stresses the control we exert over a shorter memory store to control and manipulate information
what does STM give the idea of
Whereas STM gives the impression of a static/passive store
(maintenance)
STM - a more passive retention of material
Stresses the memory store itself
why is WM important
we refer to WM to denote the active manipulation of information within a STM store in the service of high cognitive functions (e.g., comprehension, reading etc.)
- WM underlies the successful execution of complex behaviour, regardless of the cognitive domain or domains that are being engaged (D’Esposito and Postle, 2015)
- When working memory fails, so too does the ability to carry out many activities of daily living - every aspecxt of decision making and goal directed behaviour relies on working memory
baddeley model of working memory
Seperate STM stores - slave systems controlled by the central executive system which is responsible for manipulating and controlling information within the stores
visuospatial sketchpad
episodic buffer
phonological loop
central executivre
working memory
Working memory - responsibly for retaining an manipulating memory at the present time is shown in the model of working memory
central executive system
responsible for manipulating and controlling information within the stores
visiospatial sketchpad
Visuospatial sketchpad- stm store in which retain limited amount of information related to visuospatial details
phonological loop
Phonological loop - verbal information for limited time
episodic buffer
Later - episodic buffer - slave system to central exec - retains episodic information for limited amount of time comes form LTM episodic information already stored
Studied less systematically than other
model of working memory presents stm stores as
Separate STM stores and an executive system for manipulating &
controlling info within the stores
who argues against seperate stm and ltm stores
Other approaches argue against separate STM and LTM stores (e.g., Cowan, 2001; D’Esposito, 2007)
seperate ltm stores
visual semantic
episode ltm
language
phonological short term memory - location
the phonological store lies posteriorly
– parietal lobes
Information is passively retained in limited capacity short-term stores
Information is passively retained in limited capacity short-term stores
– phonological STM, visuospatial STM
– Capacity limit of phonological STM assessed by span tasks
(Digit span, operation span)
what is the role of the central executive
The central executive is responsible for refreshing information in the stores (rehearsal) and manipulating that information (e.g. using the list of numbers in STM to perform calculations)
– In the PFC; some suggest various division of the PFC for various working memory functions
the central executive relies on the functions of the prefrontal cortex
evidence for independence of the buffers
Verbal Task
1: Hold in mind a few words Verbal Task
2: repeat the-the-the
Reduced ability to perform Task 1
Capacity of phonological loop shared between these two task
Reduced ability
Two task share limited capacity of same buffer of phonological loop
evidence for independence of buffers
verbal task 1 - hold in mind a few words
visuospatial task 2 - retain position of moving stimulus
no effect in performing verbal task 1
Each STM store/buffer has its own capacity allocation and functions relatively independently from the other
Functional Imaging evidence for the dissociation between Verbal and Visuospatial STM
PET study
smith et al
Short-term retention of either letters (verbal STM) or
location of markers (visual STM)
Asked to remember locations or letters after a delayed period of 3 seconds
Then presented a location marker or a letter and asked whether location or letter had been in the original array
findings of smith et al
Distinct brain regions are active in the two WM tasks
- Verbal STM – Left hemisphere
- Visuospatial STM – Right hemisphere
where is verbal memory lateralised
Predominantly left lateralised - identified within inferior frontal cortex
where is spatial memory lateralised
Predominantly lateralised in the right hemisphere and included areas within the inferior frontal cortex the parietal cortex and occipital cortex
digit span test
Phonological STM
Numbers - retain and immediately report back after examiner finished reading
1 second per number
Demands increase as successfully
Stops when two sequential errors in both trials
Report numbers backwards
Increasing working memory demands
Working memory rehearses memory but manipulates by reordering
Happens in a number of seconds
operation span
Participants read and verify a simple math problem
• Then read a word after the operation (such as Truck).
• The participants recall the words
• Greater load with more intervening words and operations
Recall words
Keep track and rehearse words
Dual task procedure
Increase working memory load - mathematical questions
Target is successfully retaining verbal information
capacity of phonological STM
Miller (1956) suggests span is 7 +/- 2; he thought of these as meaningful “chunks”
what does chunking rely on
Chunking relies on LTM, and on LTM-STM interactions
what did cowan argue about chunking
limit is lower - around 4
span not strictly related to meaninfulness of chunks bc
Span length lower for polysyllabic words (skeleton,
binoculars etc.)
– Span length lower for phonologically similar words (map, cat, cap etc.)
what does capacity of phonological STM depend on
Limit depends on opportunity for rehearsal (e.g. articulatory suppression)
Rehearsal - important for phonological STM
Lower when speech production task
Disrupts ability to rehearse the verbal material
length of words has an efect on capacity of STM
Length of words has an effect on capacity of STM
Significant reduction in short term capacity for longer words vs shorter words
The span of verbal stm does not rely on the meaning of the word and chunks - if this was the case wed be able to store longer words at the same level as short words
Instead the capacity seems to rely on the phonological characteristics of the items themselves
Span length - lower for phonologically similar words
Capacity of STM - organised by phonological characteristics of items and not so much my meaningfulness
what does the phonological loop contain - 2 components
According to the Baddeley model the Phonological loop contains a
phonological store component (i.e., verbal STM) and a rehearsal mechanism
what is the evidence for seperation of verbal stm and rehersal mechanism
Paulesu et al. (1993) PET study while
participants performed tasks
engaging:
• a) Short-term memory for letters (both store and rehearsal components) - sequences of letters
• b) rhyming judgments of letter (rehearsal system only) - does letter rhyme
where is the phonological store
left supramarginal gyrus
posterior and parietal areas
where is the rehersal system
Rehearsal system à Brodmann’s area 44 (Broca's area).
prefrontal cortex
inferior frontal cortex
visuospatial short term memory - slot model
• Slot model: A small number of memory ‘slots’, each capable of storing a single visual object with fixed precision.
what are the two approaches to measuring visiospatial short term memory
slot model
resource model
resource model - visuospatial stm
Resource model: No upper limit on the number of items stored;
instead, the more items that are held in memory, the less precisely
each can be recalled.
evidence for slot model - luck and vogel 1997
Luck and Vogel (1997) demonstrated evidence supporting the slot model.
• They used brief displays of arrays (including coloured squares and oriented coloured lines)
- Retain information about only four colours or orientations in visual working memory at one time
- Retain both the colour and the orientation of four objects
• Visual short-term memory capacity understood in terms of integrated objects rather than
individual features
evidence for resource model
location and colour tested after delay
The more items that are held in memory, the less precisely each could be recalled
Visual working memory consists of a common resource distributed dynamically across the visual scene, with no need to invoke an upper limit on the number of objects represented
Ranganath et al., (2004) explored visual STM/WM maintenance (and
long-term associative retrieval)
coded unique and pairs of houses or faces
DMS = Delayed matching-to-sample - probed visuospatial working memory
DPA = Delayed paired associate - probed visuospatial associative recall
findings of Ranganath et al., (2004) Visual WM Results
Activity within category-selective regions of inferior temporal cortex reflected the type of information that was actively maintained during both the associative memory and working memory tasks.
- Maintaining single object in STM involves activating ventral stream representations
- These regions are functionally connected to frontal and parietal regions during the delay period
area responsible for face infoo
fusiform face ares
area responsible for place info
parrahippocampal place area
what does the delayed response task measure
working memory in monkeys
what happens in delayed response task
The animal must continue to
retain the location of the
unseen food during the delay
period (working memory)
what affects performance in delayed response task
prefrontal lesions
associative memory and delayed response task
food is paired with a visual cue (plus sign)
The task measures the animal’s ability to retain long-term rules
- No need for the animal to retain visuospatial information during the delay period (as in the working memory task)
- PFC damage disrupts a but not b
Prefrontal neurons differentially respond to different stages of the experiment
(cue – delay – response)
true
what do neurons do in delay
Neurons active during the delay period provide a neural correlate for keeping a
representation active after a triggering stimulus is no longer active
They remain active only if they animal needs to use the information for a
forthcoming action
• If the task conditions change, the same neurons become responsive to a new set
of stimuli
PFC in working memory for faces
Response sensitive to demands of working memory
Sustained response in delay period was greater when the pps had to remember 3 or 4 intact faces
As compared to one or 2
lpfc compared to ffa
Response stronger in FFA than LPFC
Greater activation in FFA when faces on screen
Delay - LPFC sustained activation
Substantial drop in FFA the response did not drop to baseline
Although sustained response was more in the FFA it was higher than what would have been expected with non facial stimuli
Consistent - pfc is critical for working memory bc it sustains a representation of the task goal to remember faces in this case
Works with inferior temporal cortex to sustain information that is relevant for achieving that goal across the delay period
what does the PFC represent
Prefrontal cortex activation
reflects a representation of task
goal
what does working memory rely on
Working memory relies on the interaction between PFC and other parts of the brain that contain perceptual and long- term knowledge relevant to a goal
what happens when we want to achieve a goal
Every time we need to achieve a goal these goals remain within our working memory and directs activation within LT representations in other parts of the cortex
These activated representations remain active locally within the posterior parts of the brain but also within the prefrontal cortex as well
location
parietal
colour
temporo-occipital
shape
inferior-temporal
petrrides theory of working memory - two processes
Petrides’ model assumes division of PFC into at least two separate
processes – maintenance and manipulation
DLPFC
Maintained activations about representations passed onto othe DLPFC
Responsible for manipulating and monitoring information
Monitoring accuracy of information in
order to perform action
role of DLPFC
manipulation and monitoring
role of VLPFC
maintain and retrieve information
role of posterior cortex
storage site of information
VLPFC
Assumes vlpfc
Sub serves the maintenance of information
Responsible to find the right representations in posterior brain regions for example
Find right representations to support the current goal
Maintain this activity within the vlpc
are Patients with PFC damage impaired at self-ordered pointing task
yes
what does PET show about short term retention of spatial information
PET study showed that short-term retention of spatial information =
ventrolateral PFC
location
what does PET show about retention and updating new locationsw
retention + update new locations = dorsolateral PFC
what does the modal model propose about ltm and stm stores
The Modal Model (Atkinson and
Shiffrin) proposes different stores for
STM and LTM
what model did baddeleys wm model expand
Baddeley’s WM model extended the
model and made it more flexible but
retained the idea of separate STM
and LTM stores
whats the strongest evidence for seperate stores
Neuropsychological (patients) and
behavioural (primacy and recency
effects)
patient HM what happened
Inability to make new memories (i.e., cannot transfer new information into LTM)
• But intact short-term memory
what happened in patient KF
KF (Shallice & Warrington, 1970)
• Left parieto-occipital damage
• Normal LTM, good long-term learning (word lists, paired associates) Healthy
• Small STM span (low digit span)
what does the unitary model of wm/stm say
Working memory is just the temporary activation of long-term memories
More items active in WM = More interference there is
what is cowans WM model
Cowan’s WM model (Cowan, 2001) is similar to Baddeley’s formulation but does not necessitate separate stores for STM
Instead central executive is responsible for selecting and activating LTM representations to bring them into “Focus of Attention”
unitary models of wm/stm
Cognitive neuroscience evidence consistent with the idea that WM entails temporary activity in parts of the brain involved in long-term storage
• Baddeley: studies of this kind support the idea that working memory and
long-term memory are not completely distinct
what does WM involve
Working memory involves the manipulation of information held within STM and is linked to the functions of the prefrontal cortex
is STM distinct from LTM
STM can be considered as distinct from LTM, although an alternative view suggests that STM is the temporary activation of LTM
what does working memory rely on
Working memory relies on the interaction between PFC and other parts of the brain that contain perceptual and long-term knowledge relevant to a goal
what does the PFC do
PFC helps to select information that is relevant for the current goal and points to the information in posterior cortex that needs to be activated
what did petrides do
Petrides discriminated the functional role of DLPFC (manipulation and monitoring) from VLPFC (maintenance of activated representations)
what are PFC functions typically associated with
The PFC is very diverse and its functions typically
associated with higher cognitive functions (e.g.,
executive functions)
in working memory what is PFC identified as
In WM PFC is identified as the site of the central
executive in Baddeley’s model
what is PFC responsible for
Responsible for maintenance and active control of information represented in LTM systems
• PFC functions are also prevalent in purely LTM tasks (encoding and retrieval)
PFC and memory encoding
Lateralised responses in PFC at encoding depend on the type of materials
• Encoding of words or semantic materials (e.g., objects that can be verbalised) involve the left PFC
• Encoding of spatial information or faces (as in Kelley et al., 1998 study) involve the right PFC
what does PFC aid in
PFC regions aid in the organisation, selection, monitoring, and
evaluation of processing that occurs at retrieval
PFC in recall
Fletcher and Henson (2001): Evaluation of what has been retrieved from LTM = monitoring – in DLPFC
- Activity in this region increases with increased demands (or increased need for monitoring retrieval)
- E.g., free recall, recall (vs recognition), low confidence judgments increased DLPFC activity
what happens in PFC damage
• PFC damage results in more severe impairment during free recall as
compared to recognition (e.g., Wheeler et al., 1995; MacPherson et
al., 2016)
free recall
minimal cues at test to aid memory performance
PFC and free recall
Greater strategic search, organisation, selection and evaluation of retrieved information takes place in free recall than in recognition or cued recall
maximal activation in pfc
source monitoring and PFC
Related to recall and recollection that stresses the ability to attribute retrieved memories to their original context
The PFC is involved at placing an event in context as this requires
active evaluation before we are able to access the origin of the
memory
Did I hear something about COVID-19 pandemic on BBC
what happens when damage to PFC
Damage in PFC
Difficulty in putting memories in their correct spatial and temporal context
Prefrontal cortex has a critical role in placing an event in each context
This is bc doing this requires an active evaluation before we are able to access the origin of a memory
what does the PFC do
Putting memories in their spatial and temporal context
- Subjectively may experience “remembering” (e.g., they can perform a recognition task) but they fail to retrieve the correct source
- They are more likely to confabulate: report narratives that include false memories (fabricated events)
