Memory Flashcards
Define coding
Format in which info is stored
Define capacity
Amount of info that can be held in memory store
Define duration
Amount of time info can be held in memory store
Define STM
Limited capacity memory store. Coding acoustic, capacity btw 5-9 items, duration 18-30 sec
Define LTM
Permanant memory store. Coding semantic, capacity unlimited, duration upto a lifetime
Outline study: coding STM + LTM
- Baddeley
- Acoustically similar/dissimilar words
- Semantically similar/dissimilar words
- Immediate recall: worse w/ acoustically similar, STM acoustic
- Recall after 20 mins: worse w/ semantically similar, LTM semantic
Outline study: capacity STM (digit span)
- Jacobs
- Digit span: researcher reads 4 digits + inc until unable to recall correctly
- Pps read 9.3 no. + 7.3 letters correctly
Outline study: capacity STM (chunking)
- Miller
- Made observation that things come in 7
- Span is 7 (plus/minus 2) - can be improved by chunking- grouping sets of digits/letters into meaningful units
Outline study: duration STM
- Peterson + Peterson
- 24 students given consonant syllable + 3 digit no. to count backwards for 3-18 secs
- Recalled 80% w/ 3 sec interval, after 18 sec recall fell to 3% - duration STM w/o rehearsal 18-30 sec
Outline study: duration LTM
- Bahrick
- Pp: 392 Americans aged 17 + 74
- Recognition test: 50 photos from year book
- Free recall test: listed names from graduating class
- 48 years after grad 70% in photo recogintion, less accurate in free recall
Evaluate the studies into memory
(-) Baddeley doesn’t use meaningful material - when processing meaningful info, people may use semantic even in STM
(-) Jacob’s conducted long time ago - lack control of ext variables
(-) Miller overestimated capacity STM - Cowan concluded capacity was only 4 chunks
(-) PP uses artificial stimulus - lack ext validity
(+) Bahrick high ext validity - lab study w/ meaningless pic recall rates lower. However confounding variables
Define the multi-store model
Representation of how memory works in terms of 3 stores: sensory register, STM + LTM. Also describes how memory is transferred from one memory store to another, how it’s remembered + forgotten
Describe the sensory register (MSM)
- Stimulus passes through + has 5 stores (senses)
- Capacity: high
- Duration: brief (miliseconds)
- Coding: depends on sense
- Contains sub stores: iconic + echoic
- If we pay attention to info in this store, it will transfer to STM
Describe the STM store (MSM)
- Duration: 18-30 sec unless info is rehearsed
- Capacity: 5-9 items
- Coding: acoustic
- If info is rehearsed, it will be maintained in the store (maintenance rehearsal)
Otuline how information is transferred from the STM store to the LTM store (MSM)
If we rehearse (repete)
Describe the LTM store (MSM)
- Permanent memory store
- Duration: upto a lifetime
- Capacity: unlimited
- Coding: semantic
- When we want to recall material stored in LTM it has to be transferred back to STM - retrieval
Evaluate the MSM
(+) Research showing STM + LTM are diff - Baddeley found: mix up AS (STM) + SS (LTM), shows 2 seperate stores
(-) Oversimplifies LTM - Not single unitary store: episodic, semantic + procedural
(-) Oversimplifies STM - KF study suggests 2 store: process visual + auditory, STM poor when digits read aloud but better when read himself
(-) Supporting research uses artificial material - digits, letters etc - low eco val
What are the 3 stores in LTM?
- Episodic
- Semantic
- Procedural
Outline episodic memory
- Personal events
- Time stamped
- Involve people, places, objects + behvaiours
- Have to make conscious effort to recall
Outline semantic memory
- Knowledge of world + what concepts mean
- Need conscious effort to recall
Outline procedural memory
- Knowledge of how we do things
- Recall w/o conscious effort
Evaluate LTM
(+) Case study evidence - HM difficulty recalling events from past but semantic memory unaffected
(+) Brain scan studies - Tulving: pps performed memory tasks while brain scanned w/ PET, E (r) + S (l) in prefrontal cortex
(-) Clinical evidence - lack control of variables (where damage is located) - difficult to generalise
(-) 2 types of LTM - E + S stored together (declarative memory) + P is non-declarative
Define the working memory model
Representation of STM. Suggests STM is a dynamic processor of diff types of info using sub-units coordinated by a central decision making system
Describe the central executive (WMM)
- Attentional process which moniters incoming data
- Allocates slave systems to tasks
- Limited storage capacity
Describe the phonological loop (WMM)
- Deals w/ auditory info + preserves order info arrives
- Divided into:
- Phonological store: stores words you hear
- Articulatroy process: allows maintenance rehearsal
Describe the visuo-spatial sketchpad (WMM)
- Stores visual and/or spatial info
- Divided into:
- Visual cache: stores visual data
- Inner scribe: records arrangement of objects in visual field
Describe the episodic buffer (WMM)
- Temporary store for info
- Intergrates visual, spatial + verbal info from other stores
- Maintains time seq
- Links to LTM
Evaluate WMM
(+) Case study - KF who had brain damage had poor STM ability for verbal info but process visual info normally
(+) Dual task performances - Baddeley found pps found difficulty doing 2 visual tasks than verbal + visual task
(-) Brain damaged patients - difficulty paying attention so underperform
(-) Lack of clarity over central executive - needs to be more specified than being ‘attention’ eg. seperate components
Define interference
Forgetting bc 1 memory blocks another, causing 1/both memories to be distorted/forgotten
Which part of memory does forgetting occur?
LTM - can’t access memories even though they’re available
Define proactive interference
Older memory disrupts newer one
Define retroactive interference
Newer memory disrupts an older one
Why is interference worse when memories are similar?
- PI - previously stored info makes new info difficult to store
- RI - new info overwrites previous memories which are similar
Outline the study into interference
- McGeogh + McDonald
- PP learnt list of words to 100% accuracy
- Given new list to learn: synonyms, antonyms, unrelated, consonant syllables, 3 digit no. + no new list
- Most similar material produced worst recall
- 3 digit, mean no. of items recalled inc
- Interference is strongest when memories are similar
Evaluate interference as an explanation to forgetting
(+) Research support - McGeogh + McDonald - lab study therefore high control over variables
(+) Real life studies support - Baddeley + Hitch: rugby players recalled teams they played week by week, accurate recall didn’t depend on how long ago it was but the amount played between
(-) Artificial task - list of words
(-) Time btw learning - time btw learning + recall are short in lab studies - doesn’t reflect real life
Define retrieval failure
A form of forgetting. It occurs when we don’t have the necessary cues to access memory. Memory is available but not accessible unless suitable cue is provided
Outline encoding specificity principle
- Cues help retrieval if same cues are present at encoding + retrieval
- Closer the retrieval cue to OG cue, better the cue works
Define cues
Triggers of info that allows us to access a memory. Can be meaningful (mnemonics) or maybe indirectly linked by being coded at the same time as learning (ext - env / int - mood)
Outline a study into context-dependent forgetting
- Godden + Baddeley
- Divers learnt words (4 conditions)
- Learn land - recall land / learn UW - recall UW
- Learn land - recall UW / Learn UW - recall land
- Recall more accurate when context of learning + recall matched
- Ext cues available at learning diff at recall so led to retrieval failure
Outline a study into state-dependent forgetting
- Carter + Cassaday
- Given antihistimines (drowsy) to change internal state - Learn on drugs - recall on it etc.
- Recall more accurate when context of learning + recall matched
- Ext cues available at learning diff at recall so led to retrieval failure
Evaluate retrieval failure as an explanation to forgetting
(+) Research evidence - Godden + Baddeley
(+) Context-related has real life application - cognitive interview
(-) Context effects only occurs when memory tested in certain way - replicated underwater study w/ recognition, recall = same
(-)Context effect not v strong in real life - context has to be v different before effect is seen eg. learning in 1 room + recall in another produces similar results
Define misleading information
Incorrect information given to the eyewitness usually after the event. It can take many forms, such as leading questions and post-event discussion
Define leading question
Question that suggests a certain answer
Why do leading questions affect EWT?
- Response bias exp - influences kind of ans given
- Substitution exp - interferes w/ OG memory, distorting it
Outline research into leading questions
- Loftus + Palmer
- 45 pp watched clips of car crash + ans Q about speed
- Q: how fast were the cars going when they hit each other?
- Hit, contacted, bumped, collided + smashed
- Contacted: 32 mph
- Smashed: 41 mph
- Leading Q biased recall
Define post-event discussion
Occurs when there is more than one witness to an event. Witnesses may discuss what they have seen- influencing the accuracy of their recall
Define memory contamination (post-event discussion)
When co-witnesses disscus crime, they mix info from other witnesses w/ own memories
Define memory conformity (post-event discussion)
Witnesses go along w/ each other to win social approval bc believe other witness is correct
Outline research into post-event discussion
- Gabbert et al
- Paired pp watched diff video of same event
- PP discussed what they saw before recall test
- 71% mistakenly recalled aspects of even they didn’t see on their video
- Control: no discussions = no error
Evaluate misleading information
(+) Real life application - police need to be careful when phrasing Q is eyewitness interviews
(-) Artificial material - experiencing + watching crime is diff - low eco val
(-) Demand characteristics - pps want to help researcher so ans how they would be expected to - low internal val
(-) Individual diff - older people less accurate than younger people when giving report + own age bias
Outline research into anxiety having a negative effect
- Johnson + Scott
- PP waiting room believing they’re taking part in lab study
- Each pp heard argument: LA: man carrying pen w/ greasy hands HA: sound of breaking glass + paper knife covered in blood, pps asked to pick man from photos
- 49% LA were able to identify him, 33% HA
- Tunnel theory of memory - witness’ attention of weapon bc source of danger + anxiety
Outline research into anxiety having a positive effect
- Yuille + Cutshall
- Real life crime, gun shop owner shot theft, 21 witnesses + 13 agreed to pp
- Pp interviewed 4-5 months after + compared to police accounts at the time - also rated stress levels at the time
- V accurate + little change
- PPs who reported highest levels of stress were more accurate (88% compared to 75%)
Explain the contradictory findings of anxiety
- Inverted U theory: relationship btw performance + arousal is curvilinear
- Recall inc w/ anxiety upto an optimal point
Evaluate the effects of anxiety on EWT
(+) Real life applications - understanding effects of anxiety on accuracy of EWT helps determine how likely witness saw what they believe they saw
(-) J+S test surprise not anxiety - scissors, handgun, wallet + raw chicken in salon, recall poorer for high surprise than high threat
(-) Field lacks control - post event discussion
(-) Ethical issues - creating anxiety causes psychological harm
Who came up with cognitive interview?
Fisher + Geiselman
What are the components to a cognitive interview?
- Report everything - even if irrelevant, might trigger
- Mental reinstatement of context - encouraged to mentally recreate physical + psychological env of incident
- Reverse order - asked to consider alternative timeline
- Change perspective - recall from another perspective
Evaluate the effects of cogntive interview on EWT
(+) Research support - Kohnken et al meta analysis 53 studies + found inc 34% amount of accurate info generated in CI
(-) Takes time - requires training
(-) Diff in old + young people - Mello + Fisher found CI produced more info from older pps - more effective for older
(-) Inc quanity not quality - Kohnken found 81% inc in accurate info but also 61% inc in incorrect
