memory Flashcards
define STM
allows a person to recall a limited string of info for a short time
define LTM
stores a range of memories + experiences that are older than 30s
define capacity
the amount of info that your memory can hold onto
capacity for STM?
A01 + A03
- (1956) Miller researched capacity in STM
- he made the term ‘magical number 7’ –> means that he proposed the ides of how we can hold 7 items , +/- 2, in our STM - the whole range = 5-9 items
– he did this using meta-analysis - he also investigated chunking - and found that this increases capacity in STM by grouping info together into longer units
(D) conflicting research (stm)
- Cowan reviewed other research & discovered the avg capacity = 4 ‘items’
- suggests Miller overestimated the amount of info that STM can hold
- conveys that Miller’s research = flawed
One limitation of miller’s research is that he may have overestimated the capacity of STM.
For example, Cowan (2001) reviewed other research and concluded that the capacity of STM was only about four chunks.
This suggests that the lower end of Miller’s estimate (five items) is more appropriate than seven times.
(D) miller doesn’t consider how other factors could affect capacity
- Jacob (1887) did a similar study - his research acknowledged that STM gradually improved w age
– his study shows how 8yrs = avg recall of 6.6 digits & 19yrs = avg recall of 8.6 digits on a digit span task
- suggests capacity is not fixed & = affected by age –> we could develop strategies to improve digit span - like chunking
what is the capacity of LTM?
it’s unlimited
what is the duration of STM?
A01 + A03
the Petersons in 1959 investigated this by conducting a lab exp
- sample = 24 psych students
- had to recall trigrams at different intervals (to prevent rehearsal), whilst the person testing them counted backwards from a specific number
- found –> longer the interval = less accurate the recall
- after 3s = 80% recall
- after 6s = 50% recall
- after 18s = >10% recall
- concluded: STM = limited duration of approx 18s-30s & w/out rehearsal, info = not passed to LTM
(D) - Peterson and Peterson’s study –> stimulus material was artificial
- Trying to memorise consonant syllables does not reflect most real-life memory activities where what we are trying to remember is meaningful. {doesn’t occur in day-to-day life} –> conveys how the study lacked external validity.
- However it can be argued that we do sometimes try to remember fairly meaningless things, such as phone numbers, so the study is not totally irrelevant.
what is the duration of LTM?
A01 + A03
in 1975, Bahrick investigated investigated this
- sample = 392 american uni grads
- procedure: had to match names + pics from their high school yearbooks –> tested after 14yrs then 47yrs
- findings:
– after 14yrs –> 90% correct recall
– after 47yrs –> 60% correct recall
- conclusion = duration = a lifetime
(A) Bahrick’s study = high ext val
- real life + meaningful memories = studied
- studies that used meaningless pic -> lower recall rates
(CP) w RL research -> confounding variables
– ptts may have looked over their yearbook more recently than others {rehearsal}
what are the coding for LTM + STM
A01 + A03
in 1966, Baddley investigated this
- in the exp - participants = given list of words that were either: acoustically (dis) / similar {sounds the same} or semantically (dis) / similar {adjectives // same meaning}
- ppts = tested on immediate recall (STM) & delayed recall (LTM)
- findings: difficulty remembering acoustically similar words in STM & difficulty in LTM with semantic words
– the STM holds onto the acoustic words - hence why it can get muddled up when being recall / retrieving the info -> same w LTM w semantic words
- conclusion –> STM = coded acoustically & LTM = coded semantically
(D) - Baddeley’s study used quite artificial stimuli rather than meaningful material.
- The word lists had no personal meaning to participants.
– suggests we should be cautious about generalising the findings to different kinds of memory task.
– eg –> when processing more meaningful information, people may use semantic coding even for STM tasks.
- This suggests that the findings from this study have limited application
AO1 for MSM?
- established by Atkinson+ Shiffrin (1968)
- describes where and how memories are stored
- depicts movement of info across the 3 permanent memory storage systems
- sensory register (SR) + STM + LTM
- starts with a change in environmental stimuli –> if this is paid attention to - it’ll briefly be stored in the SR - but if its not paid attention to, then it decays (it fades away)
- if the initial stimuli = attended to during its time in the SR - it’ll be transferred to the STM for a duration of up to 30s (max)
- if maintenance rehearsal = applied then it’ll stay in the STM (which will be coded visually / semantically) - but if not revisited, then the memory will be forgotten
- could also be forgotten if there’s too much to remember in the STM - as according to Miller, the STM’s capacity ranges from 5-9 - but it can be increased through the method of chunking {grouping info together}
- if the memory = rehearsed then it’ll move to the LTM - where it could say for a duration of a lifetime (according to Bahrick)
- this memory can also be stored alongside many others, as the capacity for LTM is unlimited
- the process to recall memories from the LTM is called ‘retrieval’ –> it’s when the memory = transferred from the LTM to the STM
evaluation for MSM?
(A) - has supporting research from Glanzer + Cuntiz (1966)
- in their study they showed ppts a list of 20 words and it was presented 1 at a time - they were then asked to recall the words
- ‘when asking ppl to remember a list if words > than the STM capacity - they have a tendency to remember words from the beginning + end of the list’
- this supports the MSM as:
– the remembered first words of the list can be recalled due to internal rehearsal of them - meaning these words are stored in the LTM
– the last words = remembered as they’re paid attention to & are fresh in their memory - meaning they’re briefly stored in the STM
- this clearly illustrates how there are different sectors that make up memory - 2 being STM + LTM - which is reflected in the MSM
(A) - the case study of HM supports the MSM
- HM suffered from epilepsy & was left brain damaged after an operation to remove his hippocampus
– discovered that he couldn’t form new long term memories & he remembered little about his personal events (death of parents) or public events –> but his STM remain intact
- this supports the MSM as this clearly illustrates how the STM + LTM are seen as 2 separate entities : despite the damage of his LTM he can still form STMs
- however - because this case study only focuses on 1 person - there could be other factors that caused HM’s state - which make this case study lose credibility as support for the MSM
(D) - it doesn’t consider how info = processed to transfer to LTM
- Craik + Lockhart (1972) suggest that memory = by-product of the depth of processing info
– also how there’s no clear distinction between STM + LTM
- we remember things due to how deeply they’ve been processed - instead of maintenance rehearsal (like astkinson + shiffrin suggest)
- therefore it contradicts MSM as it suggest that processing = > important than rehearsal & how there’s no clear difference in STM + LTM
A01 for working memory model
- Baddley & hitch proposed the WMM in 1974 based on Baddley’s prev research on memory → Baddley realised that memory (STM) = more complicated that the MSM shows and is only a unitary system and it needed to be rectified
- the WMM shows how info = stored + coded as memories
- it is made up of the central executive + visuospatial sketchpad + phonological loop + the episodic buffer
- the CE allocates the info to various components of the slave systems
↳ Hence why the name can be considered a metaphor to a company boss → they make decisions about what ‘issues’ deserve attention and what should be ignored - it has a limited capacity (info can be stored but it won’t be held for long) and it’s coding is modality free (can be visual // acoustic // semantic)
- the phonological loop = 1st slave system
- deals w acoustically coded memories
- it has 2 subdivisions: the phonological store & the articulatory control system
↳ PS = remembers everything heard & has limited capacity
↳ ACS = remembers words that are seen + heard which are internally repeated (inner voice) - a form of maintenance rehearsal → hence why this subdivision has a better capacity + duration - the VSS = 2nd slave system
- stores visually coded memories & has 2 subdivisions: visuo cache + inner scribe
↳ VC = stores everything seen
↳ IS = has the spatial relations of everything seen - has a limited capacity of 3-4 ‘objects’
- the episodic buffer = temp store for info
- it integrates information from all other stores
- it’s another store targeted for more generalised info / memories & is used as an extra storage system
- has limited capacity (can hold up to 4 ‘chunks’
- uses modality free coding
A03 for WMM?
(A) - supporting evidence from baddley + hitch
- they conducted an exp where ppts performed 2 separate tasks simultaneously (dual task performance)
- using a pointer - ppts had to track a point of light & mentally move around the sides of an imaginary letter F
- - can complete this separately w/out any difficulty –> performance = hindered when the tasks were done at the same time
- supports the WMM as it conveys how doing tasks using the same stores / components hinders performance; both of the tasks = uses visuospatial sketchpad (there was interference)
(CP) however - there are several issues w/ this research - weakening its use as support for the WMM
- it has low eco val as it’s a lab exp –> it;s artificial & the environment / variables = high controlled (not natural)
- the tasks completed wouldn’t really take place under real-life circumstances
- also - baddley + hitch were the ones to propose the idea of the WMM –> thus so the interpreted results of the exp could be bias to ensure their model has supporting evidence - therefore weakening the credibility of the findings
(D) - CE too vague + functions simlar to attention in the MSM
what are the types of long term memory
A01 + A03
- LTM = split into non-declaritive & declaritive [facts / data / events]
-
procedural: it’s implicit & non-declaritive [consious effort = not needed to retrieve memories]
– eg -> msucle memory -> knowing ‘how to do’ tasks - declaritive: explicit [conscious effort = needed to retrieve memories]
– semantic = [facts + definitions + knowledge] knowing ‘what’ a task is (general knowledge)
– episodic = [personal events *experienced *-> what + where + when] knowing ‘what’ a task is (from personal experience) -> complex (event + context + time)
overlap:
- procedural LTM = implicit + non-declaritive + knowing ‘how to’ (overlap w timing) {no emotional attachment}
- episodic LTM = complex -> time + event + context (overlap w timing)
- semantic LTM {no emotional attachment -> not personal memories}
eval:
(A) supporting evidence providede by brain scans
- diff areas of the brain = active when diff LTMs = active
- procedural -> cerebellum & basal ganglia & limbic system {can be see in subconsious act of fight ot flight respose}
- semantic -> temporal lobe
- episodic -> hippocampus & parts of the temproal lobe & frontal lobe
(A) supporting research from the case study of HM
- HM suffered from epilepsy & was left brain damaged after an operation to remove his hippocampus
- discovered that he couldn’t form new long term memories & he remembered little about his personal events (death of parents) or public events –> but his STM remain intact
- post-surgery, HW = form new procedural memories but not episodic + semantic (declraitive + explicit memories)
-> showing the diff types of LTM (as it wasn’t effected as a whole)
(CP) however - because this case study only focuses on 1 person - there could be other factors that caused HM’s state => which make this case study lose credibility as support
(A) supporting research from case study of Clive Wearing
- damage to episodic memory
- suffered from complete amnesia -> can’t encode new LTMs
- has semantic + procedural but no episodic -> can play piano but doesn’t remember musical education
- found: damage to hippocampus = cause for memory damage
(D) episodic + semantic memory may not be dinstict types of memory
- Squire + Zola inevstigated ppl w LTM loss due to brain damage {temporal lobe -> should impact declaritive}
- TL = impacts semantic + episodic -> hard to dinstinguish the 2, hard to seperate the 2 (could be the same?)
interference theory
A01 + A03
[when diff sources of info = confused in, due to similarity of memories]
retroactive inteference -> old info = forgotten {new info learned replaces / intereferes w old memories}
– eg -> learning new phone number & forgets old one
proactive inteference -> new info = forgotten {old info inteferes}
– eg -> using new married name instead of ‘maiden’ name
McGeoh + McDonald did a test of similarity of test materials
- ptts = given list of 10 word, then given list of either synonyms, nonsense syllables or numbers
- results:
– synonyms = 12% inteference (most inteference)
– nonsense syllables = 26%
– numbers = 37% (least inteference)
-> the more similar the subject is => the more inteference
(A) supporting evidence from Underwood (1957)
- reviwed results from studies that looked into memory
- ptts = asked to learn word lists & recall them 24hrs later
– some studies: only learn 1 lists & others had to learn many
- Underwood tested the recall of the last word list learnt {to see how learning the other word lists would intefere when recalling the last WL}
- learning one WL: 24hrs later recall = 80% correct
- learning multiple WLs: 2hrs later recall of last WL = 20% accurate
-> hence supports procative inteference {new info = hard to recall due to info learnt before}
(CP) lacks ecological validity -> rare that we need to remember WLs IRL & usually words remembered in everyday life have more meaning
(A) Underwood + Potsman (1960)
- ptts = given list of word pairs
2 grps:
– ctrl grp = had to learn the list
– experimental grp = had to learn another list as well (where the first words = same as the other list}
- grp = asked to recall first list
- found: ctrl grp = better recall > experimental grp due to their lack of retroactive inteference {the words from the new word list intefered w remembering the first WL}
(CP) lacks ecological validity -> rare that we need to remember WLs IRL & usually words remembered in everyday life have more meaning
(D) inteferences assume forgetting = due to confusion of similar memories
- info can be forgotten w/out similar info to intefere
forgetting / retrieval failure
A01 + A03
- when new memories = encoded, other info = also encoded (like environment + emotional setting)
- cues help w remembering
- if one can’t access the cues to the memory -> cannot retrieve memory (forgetting)
- encoding specific principle (Thlving): the more similar the coding event and retrieval event are => the higher the likelihood of recalling the original memory
types of forgetting:
state dependent forgetting [not in the same emotional / internal state as when the info was encoded / learnt]
– there’s no internal cues
context dependent forgetting [not in the same environment as when the info was coded]
– no external cues
cue-dependent forgetting = cannot access a retieval cue for a memory
cue overloading = one retrieval cue = associated w many memories
cue-dependent forgetting theory = more likely to confuse similar memories, as they = associated w same cues
(A) research support from Godden + Baddley (1975)
- evidence for context dependent forgetting:
– scuba divers learnt a list of words on land + in water -> then recalled them in water + land
– results: W+W = 11.4, W+D = 8.5, D+W = 8.6, D+D = 13.5
- shows that the best recall = when intitial enviro macthes recall enviro
- evidence for state dependent forgetting:
– male volunteers - asked to remember wors drunk + sober -> they had to recall after 24hrs
– found; recall = better if ptts = same states
(CP) lacks ecological validity, as the activities = not used in everyday + normal life
- there’s no guarentee that this behaviour is a standrad / normal reaction -> possibly lacks generalsiability
- the limites reliability & credibility of this evidence is weakened as support
(D) study supports dor CDF = lab exps -> lacks ecological validity {may impact their behaviour by being in a new environment + no gurantee it’s generalisable}
(D) CDF explains how episodic (state) + semantic (conext) memories = forgotten
- ext + int cues doesn’t effect procedural memories -> limiting reliability?
(D) doesnt offer explanation of how inteference causes ‘retieval failure’
- retro + pro active forgetting only state what happens, not how or why -> limits retrieval failure as exp for forgettng -> other factors (like capacity) that impact the retrival of memory = ignored ??????????
