P-Memory

definition of memory

the process of retaining information after the original stimulus has passed through the processes of encoding storage and retrieval

definition of encoding

putting something into a code, used to store in memory, creating a memory trace

definition of storage

as a result of encoding, information is stored in memory

definition of retrieval

recovering stored information from the memory system

who made the multistore model of memory?

Atkinson and Shiffrin 1968

what are the stages of the multistore model of memory?

1. environmental stimulus

2. sensory register

   2.1 attention

   2.2 lost through trace decay

3. short term memory

   3.1 stays through maintenance rehearsal

   3.2 lost through interference displacement

4. long term memory

   4.1 elaborative rehearsal

   4.2 information retrieval

   4.3 lost through cue retrieval failure

sensory register

takes information from one of the sense organs and holds it there

sensory register encoding

sense specific: iconic(visual), echoic(auditory), gustatory(taste), haptic(tactile), olfactory(smell)

sensory register duration

fractions of a second after stimulus is unavailable

sensory register capacity

all sensory information

sensory register - key research

Crowder 1993 - iconic duration few milliseconds, echoic 2-3 seconds

Sperling 1960 - grid of letters <1s, recalled avg. 4, up to 10, duration<=2

Walsh+Thomas 1978 - iconic avg. 500 milliseconds, decrease with age

short term memory

temporary storage of information

stm encoding

Baddeley ‘66:

P- 75 ppts, acoustically similar/dissimilar list, semantically similar/dissimilar list F- number of substitution errors higher for acoustically similar, no difference for semantic

C- information encoded acoustically

stm duration

Peterson & Peterson ‘59:

A- information not rehearsed is lost rapidly

P- 24 ppt, Brown-Peterson technique, 3 consonant trigram, count back in 3s from random 3 digit number for intervals of 3 up to 18 until serial recall

F- 18s, <10% recall

C- duration 18-20s

stm capacity

LIMITED

Jacobs ‘87: P- random digits and letters, serial recall, 445F ppts; F- 5 to 9

Miller ‘56: 7±2, chunk=one space

Simon ‘74: chunking confirmed, number of chunks varies on material

long term memory

permanent store with unlimited capacity

ltm encoding

Baddeley ‘66: most errors with semantically similar, acoustic no different

conclusion: information encoded semantically

ltm duration

Bahrick ‘75:

A- duration of VLTM

P- 392 American students 17-74, uses year books

tests: free recall names, photo recognition, name recognition, match face and name

F- 90% accurate recognition, after 48yrs 80% name and 40% face; free recall 60% after 15yrs, 30% after 48yrs

C- duration potentially forever, recognition is better than recall

ltm capacity

UNLIMITED

Anokhin ‘78: potentially 1×10^{10.5km of 0s} neural connections

Types of long term memory

Cohen and Squire 1980

• procedural

• declarative

  • Tulving 1972

    • episodic

    • semantic

Procedural LTM

• Knowing how

• e.g. motor skills, cognitive skills, brushing teeth

• only consciously recalled during early stages of learning

Declarative LTM

• Knowing that

• e.g. your name, places, events

• memory for specific information

Episodic LTM

Personal information about the world and what you have experienced.

Semantic LTM

Knowledge about the world, rules, language etc.

Tulving 1972 study

• 6 ppts, given small radioactive dose for brain scan

• episodic memories activated the frontal cortex(emotional expression/personality)

• procedural memories activated the posterior regions(planned movement, spatial reasoning, attention)

• Issues: included Tulving and his wife, bias and lacks generalisability

Clive Wearing case study

• viral infection damaged hippocampus

• lost ability to transfer memories from STM to LTM

• procedural memory intact but no episodic

• supports separate memory spans and movement of information from STM to LTM

• challenges multistore model of memory and LTM as only one store

KF case study

• motorcycle injury

• STM damaged (2 digit span), LTM normal

• challenge as words better remembered if presented visually (encoded accoustically)

• supports prior research (5-9 items)

HM case study

• damaged hippocampus

• couldn’t form new LTM - thought he was 27 every day

• kept procedural but no episodic

• supports elaborative rehearsal and duration of 20s

• supports Cohen and Squire’s representation of procedural and declarative as different stores

What is a case study?

• a detailed study of one individual or event

• uses questions, interviews, personality tests, experiments

Strengths of case studies

• rick in detail - relates to real person rather than an average

• allows psychologists to study unique behaviours that otherwise could not be studied

• one study is enough to contradict a theory

• complex study so many factors can be studied at once, experiments control many variables to look at one factor

Limitations of case studies

• difficult to generalise from individual cases as each have unique characteristics

• researchers get to know individuals well - loss of objectivity

• relies on retrospective data/information from individuals past may be hard to verify - especially for memory

• ethical issues (e.g. confidentiality), many cases are easily identifiable

• cannot replicate case studies

forgetting

failure to retrieve memories

• interference theory

• cue dependent forgetting

Interference theory

forgetting due to information in LTM becoming confused with or disrupted by other information during coding, leading to inaccurate recall. Likely to occur when memories are similar.

Proactive interference

• occurs when information stored previously interferes with an attempt to recall something new

• e.g. the memory of an old phone number disrupts attempts to recall a new phone number

OLD DISRUPTS NEW

Retroactive interference

• works backwards in time

• occurs when coding new information disrupts information stored previously

• e.g. a new car registration prevents you from recalling your old one

NEW DISRUPTS OLD

Interference theory → Baddeley and Hitch 1977

• ppts had many rugby union games and had to remember as many of the teams as possible

• tested by assessing how recall was affected by the number of games played, trace decay theory was tested by assessing the amount of time that had passed in between each game played

• forgetting was more due to the number of games played rather than time passed between games, supporting interference theory rather than decay theory

Interference theory - evaluation

• only really explains forgetting when two sets of information are similar (rugby team names), does not comprehensively explain forgetting in every day life

• uses lab experiments so lacks mundane realism and ecological validity

• cognitive processes are ignored

• more research to support context dependent forgetting and other, IT cannot explain all examples of forgetting

Cue-dependent forgetting

• occurs when information is still in the LTM but cannot be accessed

• sees recall as dependent on retrieval cues - like labels on a filing cabinet

• recall dependent upon accessing the information by remembering the retrieval cue under which the information is stored

CDF - context dependent failure

• occurs with external retrieval cues (space, sound, general environment)

• forgetting occurs when the external environment is different at recall from how it was at coding

context dependent failure - research

Abernethy 1940

• ppts after learning material recalled it less well when tested by an unfamiliar teacher in an unfamiliar room than opposite

Godden and Baddeley 1975

• they got divers to learn material either on dry land or while underwater

• found to be worse when it occurred in a different context to coding

• ppts in the underwater condition recalled words better when underwater than on dry land and vice versa

CDF - state dependent forgetting

• occurs with internal retrieval cues

• when individuals internal environment is dissimilar at recall to when information was encoded (e.g. emotional state, substances, injury)

State dependent forgetting - research

Overton 1972

• ppts learn material when drunk/sober, found recall was worse when ppts were in a different internal state at recall than at coding

• kind of unethical

Goodwin 1969

• found when people encoded information when they were drunk, they were more likely to recall it in the same state e.g. hidden money

Schmidt et al. 2000s

• real life, access influence of retroactive interference upon the memory of street names learned in childhood - 211ppts, 11-79yrs

• remember the street name, information taken (no. of times moved), 25% never moved, 1 moved x40

• positive association between number of times moved house and number of street names forgotten

• new street names = more old forgotten

RETROACTIVE INTERFERENCE

CDF - evaluation

• lab based, not like everyday memory tasks such as ones based on procedural memory

• G&B findings only occurred when divers had free recall, no effect on recognition and hence not a full explanation

• CDF supports idea that states the more deeply information is process when coded, the more links and associations will be created between items in LTM - decreases chance of forgetting as more cues will be available

• real life application by police as they reconstruct the crime scene to reinstate context and hence also state, and teaching

Repression as an explanation for forgetting

• memories banished into the unconscious mind due to their emotionally damaging nature

• recovered memories tend to be ‘false memory syndrome’

• Williams 1994 - repression in women who were childhood SA victims, 38% no recall,, 16% reported ‘recovered’ memories, earlier the age=less likely to remember → unknown if diagnosis of abuse were correct

• Karon and Widener 1997, ww2 vets repressed battlefield trauma and recovered later

• Bradley and Baddeley 1990 - anxiety and aroused depress STM but enhance LTM, initially repressed but disappears overtime

Working model of memory

Baddeley and Hitch 1974

• Central executive

• Phonological loop

• Visuospatial sketchpad

• The episodic buffer

• KEY IDEA: dual tasking

Central executive

• attention system

• attends/inhibits selectively

• process from any sense

• involved in higher order mental processes

Central executive

• Mikaye 2000

  • support role/function of the CE

  • suggests that the CE is not just one central system as it has three separate functions(inhibition, shifting, updating) hence adding another dimension to understanding

• Wilson et al.

  • found damage to frontal lobe can lead to BADs, involves problems with planning and organising

phonological loop

• Inner ear

  • holds speech based information for 1-2 seconds

• Inner voice

  • rehearses verbal information before decay in 1.5-2 seconds, can be maintained by process

Phonological loop - research

• Larsen, Baddeley and Andrade (2000s) gave ppts a list of words, one w similar sounds and one w dissimilar, found recall of similar was 25% worse, suggesting speech based rehearsal processes for visually presented lists

• KF because he could remember words presented visually but not auditorily; suggests damage

Visuospatial sketchpad

• Visual cache

  • visual material about form or colour

• Inner scribe

  • spatial relations and rehearses/transfers information to the visual cache/central executive

Visuospatial sketchpad - research

• Smith and Jonides

  • one spatial stim → right hemisphere

  • one visual stim → left hemisphere

• Logie et al. 1989

  • complex video game(space fortress), involved manoeuvring a space ship

  • found performance was much worse when they had to perform another visuospatial task

• Montello et al. 1999

  • found males outperformed females in spatial processing but not visual (processed more spatial whilst walking around)

Episodic buffer

• added in 2000s

• Integration and brief storage of information from other components before transfer to the LTM

• Buffer

  • hold temporarily (modality free)

• Episodic

  • binds things together into chunks/episodes

Episodic buffer - research

Alkhalifa 2009

• report on patient with severely impaired LTM who demonstrated STM capacity, 25 prose items hence far exceeding the PL and VSS, supporting existence of an EB which holds items until recalled

Dual tasking

• if two tasks use the same component, they cannot be performed successfully

• if two tasks use different components, they can be performed successfully

Eyewitness testimony

legal term for evidence given in a court or police investigation by someone who has witnessed a crime or accident and recounts the details from memory

Reliability of eyewitness testimony

• innocent project claims EWT is the greatest factor in wrongful convictions as it is the most persuasive to jurors (least to everyone else)

• DNA testing led to 334 post-conviction exonerations in the US

Psychological factors of EWT

1. reconstructive memory and errors

2. anxiety and stress (weapon focus and personality)

3. post information blending (misleading information and leading questions)

Three stages of EWT

• Encode: may be partial

• Storage: memories may be modified whilst in storage

• Retrieval: reconstruction of memories may be influenced

Reconstructive memories and errors

• Information is not stored exactly as presented, we attempt to make sense of information and fit it into schemas

Reconstructive memories and errors: schemas

• organised pack of information about the world, events, or people stored in the LTM (cognitive shortcuts

• effective as they remove the need to store similar information more than once but affects the reliability of EWT

  • witnesses aren’t only recalling facts as they happened, reconstruction biased by schemas active at the time of recall

Barlett 1932

• memory is an active process in which what we remember depends on the information we were exposed to at the time of learning and our relevant schematic knowledge

• participant recall showed detailed changed to fir the norms of British culture, story became shorter with each recall as ‘unimportant information’ was omitted, changed order of story and added details/emotions

• overall remembered main themes but changed unfamiliar elements

• concluded that memory is: suggestable, fragile, can be reconstructed

leading question

a questioned phrased in such a way as to prompt a particular kind of answer

misleading information

information that suggests a desired response

post event discussion

information added to a memory after the event has occurred

What did Loftus suggest about new information affecting EWT

suggested any new information about the crime (media, witness statements, leading questions) had the potential to distort their recall of events

Loftus and Palmer 1974 Experiment AIM

investigate how information provided to a witness after an event influences their memory

Loftus and Palmer 1974 Experiment 1 PROCEDURE/FINDINGS

P: 45 student participants shown short video clips, 5 groups of 9 all asked:

• ‘about how fast were the cars going when they [smashed/collided/bumped/hit/contacted] each other’

• DV: estimation of speed, IV: verb used

F: Smashed=40.8mph, collided=39.3mph, bumped=38.1mph, hit=34.0mph, contacted=31.8mph

Loftus and Palmer 1974 Experiment 1 DISCUSSION

suggest two explanations:

1. response bias - different speed estimates occurred because the critical word influences/biases a persons response

2. memory altered - critical word changes memory so they actually ‘see’ the accident differently (more or less severe)

Loftus and Palmer 1974 Experiment 2 PROCEDURE/FINDINGS

P: 150 student participants shown short film that showed multi-vehicle car accident and were then asked questions about it, 3 groups of 50

• two asked ‘how fast were the cars going when they [hit/smashed into] each other’

• one not asked about the speed

• one week later all asked ‘did you see any broken glass?’ (there was none)

F: ‘Yes’ - smashed=32%, hit=14%, control=12%

Loftus and Palmer 1974 Experiment 2 DISCUSSION

• not due to response bias as all participants were asked, therefore suggesting that the leading question altered the participants memory of the event

• suggest two kinds of information goes into a persons memory for an event: person’s own perception and information supplied after (reconstructive hypothesis)

• support: Loftus and Zanni 1975, ‘did you see [a/the] broken headlight’ = 7%/17%

Loftus and Palmer 1974 - evaluation - ecological validity

• low as lab study, participants knew they were taking part

• lacks mundane realism as there would be element of surprise/lack of focus in real life

• real life would cause an increase in emotions: fear, shock, etc. that isn’t experienced watching a video, there may be a victim

• might be asked questions about a real crash until some time later

Loftus and Palmer 1974 - evaluation - participants

were all psychology students hence not representative of general population (age, driving experience, attention and testing)

Loftus and Palmer 1974 - evaluation - usefulness

• police and legal system, how witnesses are questioned

• teachers asking/ setting questions

Loftus and Palmer 1974 - evaluation - other issues

• how easy it is to estimate speed, may be easier for some (police or taxi drivers)

• driver of car isn’t mentioned (elderly or young)

• type of car e.g. Porsche/smart car

• ethical - if participant recently involved in a car accident

Post event discussion

Gabbert et al. 2003 studied ppts in pairs, each watched videos of a crime from different POVs, found 71% of ppts mistakenly recalled aspects they couldn’t see but picked up on in discussion

How could anxiety effect EWT

• state-dependent forgetting

• repression as a defence mechanism

• overwhelm - too much input leaves too many gaps for distortion

• focus - fill in later/don’t remember

  • info that is attended/not attended to, but people have different normal levels of anxiety

Anxiety - situational and dispositional factors

Situational: context induced anxiety

Dispositional: state of the person prior to event

Loftus: Weapon focus

eyewitnesses exposed to aspects of the scene that pose a direct threat

Loftus: office situation PROCEDURE

• Condition 1: ppts overheard a hostile argument followed by person emerging w a knife/letter opener in blood

• Condition 2: heard harmless conversation followed by person emerging holding a pen w greasy hands

Loftus: office situation FINDINGS

• 33% in weapon condition could identify culprit vs 49% in the other

  • believed anxiety elicited by the weapon narrowed the focus of attention to the weapon

Loftus: office situation EVALUATION

• high ecological validity as it mirrors real life (bad comparison though)

• ethical issues: weren’t aware before, may have thought it was real, past trauma

Loftus: corner shop PROCEDURE/FINDINGS/CONCLUSIONS

• asked to watch one of two sequences:

  • a person pointing a gun at a cashier and receiving some cash

  • a person passing a cheque to the cashier and receiving some cash

• witness looked more at the gun than the cheque; as a result memory details unrelated to the gun/cheque were poorer in the weapon condition

• heightened anxiety and stress has a negative impact on eyewitness identification accuracy

Weapon focus - EVALUATION

• person may use something as a weapon that does not cause anxiety

• could have more to do with state-dependent forgetting

• weapon versus check is not equal comparison as cheque is not expected

• individual differences - experience/stress/job

• Pickel found no evidence of weapon focus when a weapon is expected

• Valentine studied 300 line ups and found that presence of a weapon had no effect on probability EW identifies suspect

Stressful situations - Deffenbacher 2004

Meta-analysis of studies on the effects of anxiety and stress on EW memory

1. effects on face ID: 54% correct for low stress conditions and 42% correct for high stress conditions

2. effects on recall of culprit/crime scene details and actions of main characters: 64% correct for low and 52% correct for high

Stressful situations - Yerkes-Dodson Law

Inverted U theory of arousal

Research: Bothwell studied IDs between groups of low/high neuroticism combined w low/high stress situations

• Low neuroticism: low50%, high75%

• high neuroticism: low68%, high 32%

Evaluation of lab experiments for EWT

• focus on one aspect

• watching videos - detached

• no consequences

• not the victim

• demand characteristics

• large group, experimenter/guidance present

Evaluation of lab experiments for EWT : if real

• many things going on, lack of focus

• watching in real time - emotional involvement

• may be the victim

• responsible for self

• act naturally without demand characteristics

• Ihlebaek → one field, one video → recalled more from video

• Yuille and Cutshall → witnesses to shooting (one dead one injured), reinterviewed months later and found same accuracy as prior witness statement

Cognitive interviews: Devlin Report

• EW accounts to police interrupted repeatedly by interviewer

• questions designed to find truth caused lapses in concentration, caused memory blending

• informed home office guidelines:

  • Free recall → open questions → specific questions

Cognitive interviews: Geiselman 1985 - IDEA and AIM

• argued interview techniques should account for basic characteristics of memory:

  • effectiveness of retrieval depends on the extent which the information contained is similar to information stored in memory trace

  • retrieval cues may permit access to any given memory

• AIM: improve effectiveness of interviewers and apply results of psychological research which show that memory is not like a video camera but an active process

Cognitive interviews: Geiselman 1985 - FOUR STAGES

1. Context Reinstatement:

• recall the scene, the weather, thoughts and feelings

• details act as cues for recall

2. Report Everything:

• every detail you can. even if it seems irrelevant or trivial

• may not realise relevance, might prompt significant information

3. Recall from Changed POV:

• describe event as it would have been seem from a different POV

• encourages many retrieval paths

4. Recall in reverse order:

• describe event in reverse order

• when forward, reconstructed with scheme; when reversed more accurate

Cognitive interviews: Geiselman 1985 - EXPERIMENT

• testing effectiveness through comparison with standard police interviews

• P: showed police training videos of violent crimes to group of 89 students, interviewed by police 48 hours later

• CI - 45 minutes, SPI - 15 minutes

• F: SPI avg.29.9 correct vs CI avg.41.4 correct

Enhanced cognitive interview: Fisher et al. - METHOD

• minimise distractions → quiet and contained

• Encourage slow speech → helps to think

• Allow pauses for recollection → lack of pressure means less falling back on schemas

• Tailor language to suit witness → comfort

• Follow up with interpretive comment → ‘is this what you meant?’

• Reduce anxiety → e.g. sofa or tea

• Avoid judgemental/personal comments → social desirability, leads witness away from information that may be important

• Review statements → confirm content and presentation

Enhanced cognitive interview: Fisher et al. - FINDINGS

Lab experiment findings:

• ECI more effective

• ECI avg.57.5 correct vs CI avg.39.6 correct

• BUT more incorrect statements (as there were more overall)

Training Miami police findings:

• Number of pieces of information increased by 46%

• where possible, information was proven to be 90% accurate

CI and ECI evaluation

• Holliday 2003 produced modified cognitive interview → stresses building trusting relationships, removed ‘change of POV’ as children too young to effectively empathise

  • 4-5/9-10 year olds, MPI >accurate than SPI

• Milne and Bull 2002, ‘report everything’ and ‘context reinstatement’ are key techniques, suggests modification for focus

• proven to be successful in getting more pieces of info

• ECI>CI

• increase in incorrect information concerning

• less effective at enhancing recall when used after longer period of time

• several components → unsure if all successfully contribute

• doesn’t reduce negative effects of misleading information

• expensive to train police

• time consuming