memory Flashcards
coding AO1
The format in which information is stored in the various memory stores, e.g. acoustic (sound), semantic (meaning)
baddeley = 4 groups - word lists to remember were acoustically similar, acoustically dissimilar, semantically similar and semantically dissimilar
words recalled immediately for coding on STM and after 20 mins for LTM
participants did worse on accoustically similar words in STM suggesting memory in STM is coded for acoustically (similar sounds conflicted each other)
did worse on semantically similar words in LTM as memory is coded for semantically and the similar meanings conflicted each other
coding evaluation
used artificial stimuli instead of meaningful material personal to participants
caution about generalising findings to all memory tasks
eg: when processing more meaningful information we may use semantic coding even for short term (despite baddeleys researching theorising the use of acoustic coding)
suggests findings have limited application to other scenarios
BUT
identified a clear difference between two memory stores
Later research affirmed that STM primarily uses acoustic coding, while LTM predominantly employs semantic coding.
capacity AO1
capacity = amount of information that can be held in the memory store
jacobs digit span test - presented numbers one at a time and only continued when the order was recited back correctly, he continued increasing the list length until ppt could no longer remember the order
miller found that our capacity is 7+-2
(same for chunks of info to increase capacity)
capacity evalutation
conducted a while ago, earlier research lacked adequate controls
eg: distractions during test affect results
confounding variables weren’t controlled so conclusions may be invalid
BUT has been replicated and similar results have been obtained despite low validity
+ lacks mundane realism as participants were tested in a lab - not personal information to ppt
duration AO1 (STM/LTM)
duration = length of time information can be held in memory
STM:
Peterson and Peterson = 24 undergraduates presented with trigams eg:TNR
Told to count backwards from 100 to prevent rehearsal
retention intervals increased by 3s (3,6,9…18) before recalling trigram
3s later recall was 80% but 18s later was 3%
STM duration = 18s without rehearsal
LTM:
bahrick et al - 392 America’s aged between 17-74
tested on recall from high school classmates year book photos
15 years later - 90% recall accuracy for photos and 60% free recall
48 years - 70% photo accuracy and free recall 30%
duration evaluation (STM/LTM)
STM: memorising consonant syllables doesnt reflect real life memory situations
lacking external validity
But not totally irrelevant as phone numbers are meaningless but we do rememeber them
forgetting may not be due to meaningless stimuli and instead due to lack of rehearsal that we do (even subconsiously) in daily life due to spontaneous decay
LTM: high external validity
By investigating meaningful memories, such as people’s names and faces, the study offered a more ecologically valid assessment of LTM duration
provided a more accurate reflection of real-world memory performance
multistore model fo memory AO1
Atkinson and shiffrin
3 stores of memory - sensory register, STM and LTM
sensory register: memory store of all 5 senses eg vision (iconic store with visual coding) or hearing (echoic store with acoustic coding)
duration of sensory register is less than half a second but have high capacity
paying attention determines whether information moves to STM store
short term memory: 7+-2 item capacity
duration = 18 seconds without rehearsal and is primarily coded for acoustically
after lots of maintenance rehearsal, information passes from STM to LTM
LTM: stored after prolonged rehearsal
unlimited capacity and years of duration
info coded semantically
retrieval transfers info from LTM to STM before recall
multistore model eval - supporting research BUT artificial
Supporting research:
baddeley proves that STM and LTM are qualitively different
Ppts mixed up words that are acoustically similar in STM but mixed up semantically similar words in LTM
clearly proves STM has acoustic coding and LTM has semantic coding
supports MSMs view that the 2 stores are independent of each other
This suggests that the MSM provides a coherent framework for understanding the organisation of memory
Empirical methods = high internal validity
BUT artificial stimuli:
Words had no meaning for baddeleys short term memory study
questions the validity of the MSM in explaining memory processes in naturalistic contexts where meaningful information dominates
this suggests that the applicability of the MSM may be limited in capturing memory operations in real-life situations
multistore model eval -evidence of more than one STM store
Evidence for more than one STM store:
MSM states STM is unitary store with one type of short term memory
but warrington analysed KF who had amnesia
Found poor retention of STM digit span test when digits were auditorily recited and improved when he visually read digits himself
should be atleast one STM store for processing visual info and auditory info (working memory model)
suggests the existence of multiple STM stores for different types of information processing
LTM - unitary but we might have one LTM store for memories of world (semantic memory) and one for actions (procedural memory) etc
MSM oversimplifies the complexity of STM functioning by assuming a unitary store
multistore model eval - more than one type of rehearsal
More than one type of rehearsal:
MSM focuses on the amount of times you rehearse something increasing likelihood of transfer of info to LTM
Craik and Watkins found type of rehearsal effects more
MSM uses maintenance rehearsal but doesnt transfer info from STM to LTM
(stays in STM)
but elaborative rehearsal is needed for LTM storage (occurs via linking of info to existing knowledge creating meaning)
Craik and Watkins indicates that elaborative rehearsal, rather than sheer repetition, facilitates the transfer of information to LTM
occurs when information is linked to existing knowledge or when you think about what it means
this suggests that the MSM does not fully capture the mechanisms involved in the transfer of information from STM to LTM
types of long term memory Ao1
Tulving challenged the oversimplified model of long-term memory (LTM) by proposing three distinct LTM stores: episodic, semantic, and procedural memory
episodic memory - LTM store for personal events
includes time stamped memories of when it happened and includes several elements such as details of people, places, and actions
memories from this store must be retrieved consciously and with effort
semantic memory - stores shared knowledge of the world, such as facts, concepts, and meanings
lacks time stamps and relates to impersonal, factual information, continually expanding
require deliberate recall to retrieve info
Less vulnerable to distortion/forgetting than episodic memory
procedural memory - LTM store for knowledge on how to do things including memories of learned skills
encodes actions and skills, such as driving a car, which become automatic through practice
recalled without making a conscious or deliberate effort + challenging to articulate
types of long term memory eval -
types of long term memory eval -practical applications
Practical applications:
belleville et al demonstrated that episodic memories could be improved for older people with mild cognitive impairment via training
This is especially the case for memories of events that happened recently, as past episodic memories remain intact
trained participants performed better in episodic memory test than control group
episodic memory most often affected by mild cognitive impairment highlighting benefit of distinguishing between types of LTM for treatment
Can lead to creation of rehabilitation programs for those recovering from strokes or suffering from dementia
Insights into human memory systems inform the development of AI algorithms mimicking memory processes for retrieval systems
types of long term memory eval - neuroimaging evidence
Neuroimaging evidence:
Tulving had ppts perform many memory tasks whilst under a PET scanner
episodic and semantic memories recalled from prefrontal cortex (one on each side of hemisphere)
supports how memory stores are located in different parts of brain and are separate entities
further research studies proved this - increasing validity of findings
BUT other research suggests episodic memory is coded in the left side of the prefrontal cortex and retrieval is in the right (rather than semantic on left and episodic on right)
This inconsistency challenges the neurophysiological evidence supporting LTM localization in the brain.
working memory model AO1
Baddeley and hitch: dynamic processor which is a representation of STM consisting of a central executive, phonological loop. visuo-spatial sketchpad, and episodic buffer
central executive - oversees incoming data, allocating attention and tasks among subsystems
lacks storage capacity and primarily supervises operations
phonological loop - processes auditory information
utilises a phonological store for word storage and an articulatory process for maintenance rehearsal
Capacity is limited to approximately two seconds worth of speech
visuospatial sketchpad - processes visual or spatial info called ‘inner eye’
consists of visual cache storing visual data and inner scribe records arrangement of objects
Capacity = 3-4 objects
episodic buffer - links working memory to LTM
integrates visual, spacial, and verbal info processed by other stores
Limited to around four chunks = storage component of the CE
working memory model eval - clinical evidence
Shallice and Warrington’s case study of KF who suffered from brain damage
KF had poor short term memory ability to recall verbal info but could process visual info easily in STM
could remember digits over sounds etc
proves that only the phonological loop was damaged but other areas of memory were intact (visuo-spacial sketchpad)
supports existence of separate visual and acoustic stores
however results not generalisable as KF is a unique case after trauma
KFs injuries occured due to trauma of motorcyle accident raising the possibility of trauma-related cognitive effects beyond the identified phonological loop damage
This uncertainty challenges the reliability of clinical studies relying solely on brain injury cases to support memory models
this highlights the need for caution in interpreting findings from such studies due to potential confounding factors.
working memory model eval - dual task performance
Dual task performance:
baddeley showed ppts had greater difficulty doing 2 visual tasks than doing both a visual and verbal task at the same time
increased difficulty occurs as both visual tasks compete for the same slave system
(visuo-spacial sketchpad) compared to no competition between phonological loop and VSS
proves theres a different slave system
processing visual info
Visuospatial sketchpad = visual
Phonological loop = verbal
working memory model eval - support from neuroscientific studies
Support from neuroscientific studies:
Studies using FMRI or PET scans found correlational evidence for the different components of WMM being localised within the brain
Phonological loop linked to activity in left hemisphere (particularly brocas area)
VSS linked to right hemisphere
cognitive model reflect biological systems BUT correlation doesnt equal causation
working memory model eval - little is known about the CE
Little is known about the central executive:
CE vague + simplistic so not understood fully
the study of EVR (by eslinger and damasio) who had a cerebral tumour removed
performed well on reasoning tasks but difficulty with decision making
suggests CE was intact but doesnt fully work at same time
so CE more complex than WMM believes
Lack of specificity raises questions about how well the WMM can accurately explain cognitive processes as it doesnt operationalise the mechanisms of which the central executive manages information
other researchers propose the CE may consist of separate subcomponents
interference theory AO1
interferance is forgetting because one memory blocks another - one or both memories distorted/forgotten
proactive interferance - older memory interferes with new one
retroactive interferance - newer memory interferes with old one
research on effects of similarity: mcgeoch and mcdonald
participants had to learn 10 word wordlist until they remember 100% and then learned a new list consisting of 6 groups (synonyms,antonyms,words unrelated to original ones, consonant syllable, no list (rest) etc)
synonyms produced worst recall of original list - interferance strongest when memories are similar
The impact of similarity on recall suggests two potential mechanisms: Proactive , where existing information prevents storage of similar new information, or retroactive where new information overwrites previous memories due to similarity
interference theory eval - real world interference
Real world interference:
baddeley and hitch asked rugby players to recall names of teams they played against
number of intervening games varied between players due to eg: injury etc
players who played most games = greater interference = poorer recall
increases validity of theory - real life
High external validity - interference affects scenarios outside of controlled lab conditions
BUT
mcgeoch/mcdonalds research proves only fairly similar memories can be interfered with so doesnt occur in real life as much
Lab conditions means the stimulus can be controlled to be very similar to each other - unlike real life
forgetting could be due to lack of cues instead of interference (retrieval failure)
Therefore limited applicability of interference theory to most scenarios
BUT laboratory study of lists dont reflect real life memory
Therapy for those suffering from cognitive impairments allows practitioners to tailor intervention strategies by reducing likelihood of interference
interference theory eval - temporary but can be overcome by cues
Interference is temporary and can be overcome by cues:
Tulving et al gave ppts lists of words
70% recall for first list but got progressively worse for each additional list learnt
argued words werent lost from LTM - just needed cues for recall
at the end of test ppts were given a cued recall test (given category name) and recall increased to 70% again
interference temporarily loses access to LTM but not permanently gone (against interference theory)
interference theory eval - practical applications
Practical applications:
Can be used in educational settings by understanding how interference affects memory to enhance teaching strategies
Eg: spread out content of similar nature to prevent interference
Or students can alternate between multiple subjects to prevent retroactive interference
interference theory eval - drug study support
drug study support:
Ppts learning a word list under influence of diazepam had poor recall later on
But ppts who learnt a list then took diazepam had greater recall a week later as diazepam prevents new memories from being processed preventing retroactive interference
So original word list was better recalled
learning list before drug = better recall than placebo
could be because diazepam prevents new memories after taking drug from being processed by brain so retroactive interference cant occur
BUT no control over confounding variables that could have caused interferance within week
Emotional or physiological states also affect memory retrieval that interference theory ignores