Memory Flashcards
Coding
Format in which information is stored in the various memory stores
Research on coding
“Baddeley (1966) Word lists: acoustically/semantically similar or dissimilar, Asked to recall in correct order
Immediate recall: worse with acoustically similar words, After 20 minutes: worse with semantically similar words. STM acoustic, LTM semantic”
Acoustically similar
Sound similar
Semantically similar
Similar meaning
Research on coding - artificial stimuli (Evaluation)
- Words had no personal meaning. 2. Cautious about generalising. 3. Meaningful tasks: may use semantic even for STM. 4. Limited application
Capacity
Amount of information that can be held in a memory store
Research on capacity - digit span
“Jacobs (1887) Researched reads 4 digits and increases until ppt cannot recall order correctly
Mean span: 9.3 numbers, 7.3 letters”
Research on capacity - digit span (Jacob’s, lacks validity)
- Conducted a long time ago 2. Early research lacked adequate control of extraneous variables. 3. Ppts may have been distracted. 4. Results may not be valid - confounding variables not controlled
Results confirmed in other research, supporting validity (miller) Research on capacity - span of memory and chunking.
- Things often come in 7s (weekdays, deadly sins, musical notes). 2. Suggests capacity of STM is 7+/-2 items
People can recall 5 words as well as they can 5 letters
Chunking: grouping sets of digits/letters into meaningful chunks
Research on capacity - span of memory and chunking (overestimated capacity)
Cowan (2001) reviewed other research. 1. Capacity of STM only around 4 chunks. 2.Lower end of Miller’s estimate (5 items) more appropriate than 7
Duration
Length of time information can be held in memory
Research on duration - STM
Peterson & Peterson (1959). 24 students given consonant syllable (e.g. YCG) to remember and 3 digit number to count backwards from. Counting backwards prevented rehearsal. Retention interval: 3, 6, 9, 12, 15, 18 seconds. 3s: recalled 80% syllables correctly. 18s: only 3%. STM without rehearsal: 18-30s
Research on duration - STM (Evaluation, artificial stimuli)
Does not resemble real-life memory activities (meaningful) Lacked external validity. Do sometimes try to remember meaningless things e.g. phone numbers, not totally irrelevant
Research on duration - LTM
Bahrick et al (1975) 392 Americans 17-74 yrs. Recognition test: 50 photos from high school yearbook. Free recall: names of graduating class. Free recall: 15 yrs: 60% 48 yrs: 30%. Photo recognition: 15 yrs after graduation: 90% accurate 48 yrs: 80% accurate. LTM lasts a very long time
Research on duration - LTM (Research support + Evaluations)
Real-life meaningful memories. Lab studies with meaningless pictures: recall much lower (Shepard 1967). Confounding variables not controlled (looked at yearbooks and rehearsed memories over the years)
Multi-store model of memory (MSM)
Atkinson & Shiffrin (1968)- How info flows through memory system. 3 stores linked by processing
MSM - 1: sensory register
Stimulus from environment passes into SR along with lots of other sights/sounds e.t.c. 5 stores (each sense), main iconic (visual) and echoic (auditory). Duration: less than half a second. Capacity: high (over 100 million cells in each eye storing data). Coding: depends on sense. Needs attention to transfer to STM
MSM - 2: STM store
Limited capacity/duration Coding: acoustic. Capacity: 7+/-2 items (5-9). Duration: 18-30 seconds without rehearsal. Maintenance rehearsal: repeat to ourselves. Keep in STM with rehearsal. Rehearse long enough: passes into LTM
MSM - 3: LTM store
Permanent memory store. Coding: semantic. Capacity: unlimited. Duration: up to a lifetime. Recall: must be transferred back to STM by retrieval
MSM - supporting evidence
Baddeley: mix up acoustically similar words in STM and semantically similar in LTM Coding: STM acoustic, LTM semantic. Supports MSM view that stores are separate/independent
MSM - more than one type of STM
Shallice & Warrington: study of KF (amnesia patient). Poor STM recall for digits when read out loud. Recall much better when read digits himself. MSM states only one STM store. KF: one STM to process visual info and another for auditory. Working memory model better explains (includes separate stores)
MSM - more than one type of rehearsal
Craik & Watkins: two types of rehearsal - maintenance and elaborate. Maintenance described in MSM. Elaborate needed for long term storage. Occurs when you link info to existing knowledge/process it. Type of rehearsal matters, not amount. Research findings cannot be explained by model
Who Came up with Types of LTM
Tulving (1985) MSM view of LTM too simplistic. 3 LTM stores, different kinds of info: Episodic, semantic, procedural memory
Types of LTM - episodic
Stores events from our lives. Complex memories: time-stamped (when they happened) and contain several elements (people, places, objects, behaviours). Conscious effort to recall
Types of LTM - semantic
Knowledge of the world. Includes facts/what words & concepts mean. Not time-stamped. Less personal - knowledge that we all share. Also usually need to be deliberately recalled
Types of LTM - procedural
Stores memories for how we do things (actions/skills). Hard to explain to others - recall without conscious awareness/effort
Types of LTM - episodic supported by case study evidence
Clinical studies of amnesia (HM, Clive Wearing): both had difficulty recalling events from past. Semantic memories unaffected (HM did not remember stroking a dog but did not need concept explained). Procedural memories also intact. Support for multiple LTM stores. One store can be damaged but others left unaffected
Types of LTM - brain scan studies
Tulving et al: ppts perform memory tasks during PET scan. Prefrontal cortex: semantic (L) episodic (R). Physical reality of different types. Supports validity
Types of LTM - problems with clinical evidence
Based on what happens when memory is damaged. Lack of control of variables (precise location of damage/personality). Difficult to generalise to determine exact nature of LTM
Working memory model (WMM)
Baddeley & Hitch (1974) How one aspect of STM is organised/functions. Concerned with part of mind active when temporarily storing/manipulating information e.g. playing chess
WMM - 1: central executive
Attentional process. Monitors incoming data/makes decisions/allocates slave systems to tasks. Co-ordinates activities/allocates processing resources to 3 subsystems. Coding: flexible. Capacity: very limited
WMM - 2: phonological loop
Deals with auditory information. Preserves order in which info arrives. Subdivided into phonological store and articulatory process
WMM - phonological loop (phonological store)
Stores words you hear
WMM - phonological loop (articulatory process)
Allows maintenance rehearsal (repeating to keep in WMM while needed). Capacity: 2s worth of what you can say
WMM - 4: visuo-spatial sketchpad
Stores visual/spatial information. Coding: visual. Capacity: limited, 3-4 objects (Baddeley). Subdivided into visual cache & inner scribe
WMM - visuo-spatial sketchpad (visual cache)
Stores visual data
WMM - visuo-spatial sketchpad (inner scribe)
Records arrangement of objects in visual field
WMM - 5: episodic buffer
Added by Baddeley (2000) - Temporary store. Integrates visual/spatial/verbal info from other stores. Maintains sense of time sequencing (recording events that are happening). Coding: flexible. Capacity: about 4 chunks. Links to LTM/other cognitive processes e.g. perception
WMM - KF supports separate STM stores
Shallice & Warrington: case study of KF (brain damage) Poor STM ability for verbal info, could process visual info normally. Difficulty with sounds, could recall letters & digits. PL (Parietal Lobe) damaged, other areas intact. Separate visual/auditory stores
Evidence from brain damaged patients may be unreliable and not generalised
Concerns unique cases/traumatic experiences
WMM - dual task performance
Baddeley et al: ppts more difficulty doing two visual tasks (tracking light, describing letter F) than visual & verbal at same time. Greater difficulty: both compete for same limited resources. No competition with verbal/visual. Evidence for existence of VSS (MSM can’t explain)
WMM - lack of clarity over CE
Cognitive psychologists: unsatisfactory, doesn’t explain anything. Baddeley: most important but least understood element. Should be more clearly specified than just ‘attention’. Some believe may consist of separate components. WMM hasn’t been fully explained
Explanations for forgetting - interference
Two pieces of info conflict with each other. Resulting in forgetting of one/both or distortion of memory. LTM forgetting: can’t get access to memories even though they are available
Explanations for forgetting - interference (1 - proactive)
Older memory interferes with a newer one. E.g. teacher learned so many names in past, have trouble remembering current class
Explanations for forgetting - interference (2 - retroactive)
Newer memory interferes with older one. E.g. teacher learns so many new names this year, can’t remember last year’s students
Explanations for forgetting - interference (3 - worse when similar)
PI: previously stored info makes new info more difficult to store. RI: new info overwrites older memories which are similar
Explanations for forgetting - interference (4 - effects of similarity)
“McGeoch & McDonald (1931) - Ppts asked to learn word list to 100% accuracy. Then given new list to learn, varied in similarity. Synonyms, antonyms, unrelated, consonant syllables, 3 digit numbers, no new list (rest). Performance depended on second list
Synonyms: worst recall. Very different material: mean number of items recalled increased. Interference strongest when memories similar
Synonyms: blocked access to original/new list became confused with old”
Explanations for forgetting - interference (lab study evidence)
Many lab experiments carried out on interference (e.g. McGeoch & McDonald). Most show both types very likely causes of LTM forgetting. Control effects of extraneous variables - confidence that interference is valid explanation
Explanations for forgetting - interference (artificial materials)
Word lists often used. More realistic than consonant syllables, still differs from things we remember in everyday life. Everyday life: faces/birthdays/ingredients. Interference much more likely in lab - may not be likely cause of everyday forgetting
Explanations for forgetting - interference (real life studies)
Baddeley & Hitch: rugby players recall team names played so far in season, week by week. Accuracy did not depend on how long ago match was, but number of games played in meantime. Interference can apply to at least some everyday situations
Explanations for forgetting - retrieval failure
Forgetting due to insufficient cues. Info initially placed in memory: associated cues stored at same time. Not available at recall: may not be able to access memories that are actually there
Explanations for forgetting - retrieval failure (1 - encoding specificity principle)
Tulving (1983) - States cues help retrieval if same cue present at encoding and retrieval. Forgetting if cues different/absent at retrieval. Closer cue is to original, better it works
Explanations for forgetting - retrieval failure (2 - some cues have meaningful link)
Some cues linked to material to be remembered in meaningful way E.g. STM may lead to recall of all sorts of info about short term memory
Explanations for forgetting - retrieval failure (3 - some cues have no meaningful link)
Encoded at time of learning but not in meaningful way. Context-dependent: retrieval dependent on external/environmental cue. State-dependent: retrieval dependent on internal cue/state of mind
Explanations for forgetting - retrieval failure (4 - context-dependent forgetting)
“Godden & Baddeley (1975): deep sea divers - Cues: contexts where learning took place (land(L)/underwater(U))
Learned word lists. Learn - recall. L - L, L - U, U - L, U - U. Environments did not match: accurate recall 40% lower than those that did. Different external cues led to retrieval failure due to lack of cues. Demonstrates context-dependent forgetting (info not accessible when contexts at encoding/retrieval did not match). Explanations for forgetting - retrieval failure (5 - state-dependent forgetting)”
Explanations for forgetting - retrieval failure (5 - state-dependent forgetting)
“Carter & Cassiday (1998) - Gave anti-histamine drugs to ppts (mild sedation effect - slightly drowsy). Creates different internal physiological state. Learn word lists/passages of prose. Learn - recall, on - on, on - off, off - on, off - off. Mismatch of internal state: performance significantly worse
Internal cues absent = more forgetting”
Explanations for forgetting - retrieval failure (supporting evidence)
Impressive range of research evidence e.g. Godden & Baddeley, Carter & Cassiday. Eysenck (2010): argues retrieval failure perhaps main reason for LTM forgetting. Increases validity of explanation (studies conducted in real-life conditions as well as highly controlled lab conditions)
Explanations for forgetting - retrieval failure (context effects not as strong in real life)
Baddeley: contexts have to be very different before effect is seen (e.g. on land/underwater). Learning in one room and recalling in another unlikely to result in much forgetting (not different enough). May not explain much real-life forgetting
Explanations for forgetting - retrieval failure (recall vs recognition)
Effect may be related to kind of memory being tested. Godden and Baddeley: replicated study, recognition test rather than recall. No context-dependent effect, performance same in all conditions. Limits explanation for forgetting: presence/absence of cues only affects memory when testing recall
Eye witness testimony (EWT)
Ability of people to remember events (e.g. accidents/crimes) that they have observed. Affected by misleading information/anxiety
Misleading information
Incorrect information given by the eyewitness after an event. Includes leading questions/post event discussion
Leading questions
Question which suggests a certain answer due to they way it is worded. Incorrect information given by the eyewitness after an event. Includes leading questions/post event discussion
Post-event discussion (PED)
Occurs when there is more than one eyewitness. May discuss what they have seen with co-witnesses or other people. Can influence the accuracy of each witness’s recall of the event
Factors effecting EWT: misleading information - leading questions (1 - key study)
Loftus & Palmer (1974) - 45 students watched film clip of car accidents. Answered questions about speed. Critical question: about how fast were they going when they ___ each other? 5 groups, 5 verbs: hit, contacted, bumped, collided, smashed. Mean estimates: Collided: 31.8mph, Smashed: 40.5mph. Leading question biased recall. Smashed’ suggested faster speed
Factors effecting EWT: misleading information - leading questions (2 - response bias explanation)
Wording has no enduring effect on memory. Influences kind of answer given
Factors effecting EWT: misleading information - leading questions (3 - substitution explanation)
Wording does effect memory. Interferes with original memory, distorting accuracy. Loftus & Palmer: 2nd experiment. Ppts who heard ‘smashed’ more likely to recall seeing broken glass (there was none) than those who heard ‘hit’
Factors effecting EWT: misleading information - PED (1 - memory contamination)
Co-witnesses discuss crime: mix (mis)information from others with own memories
Factors effecting EWT: misleading information - PED (2 - memory conformity)
Go along with others to win social approval. Believe other witnesses are right
Factors effecting EWT: misleading information - PED (3 - key study)
Gabbert et al (2003) - Paired ppts watched videos of same crime. Filmed so each could see elements other could not. Both discussed video before individually completing recall test. 71% mistakingly recalled aspects they did not see (picked up in PED). Control group: no discussion, no errors
Factors effecting EWT: misleading information (real life applications) (Leading Questions)
Research led to important practical uses for police/investigators (consequences of EWT can be very serious). Loftus: leading questions can have such distorting influence police need to be careful about phrasing interview questions. Important difference to lives of real people (improving legal system)
Factors effecting EWT: misleading information (Loftus & Palmer - artificial materials) (Leading Questions)
Watched film clips, very different from witnessing real accidents (less stressful). Yullie & Cutshall: witnesses of real armed robbery had very accurate recall after 4 months. Tells us little about how leading questions affect EWT in real crimes
Factors effecting EWT: misleading information (EWT studies lack external validity)
Foster et al (1994): what you remember as eyewitness has important real world consequences, not true in research studies. Search memory with more effort: testimony may lead to a successful conviction (or wrongful if inaccurate) - not true in research studies. EWT accuracy may be greater in real world (seriousness of role)
Anxiety
State of emotional/physical arousal, Worried thoughts/feelings of tension
Factors effecting EWT: anxiety (1 - negative effect on recall)
“Creates state of physiological arousal. Prevents us paying attention - recall is worse
Study effects of weapons on accuracy of EWT recall (creates anxiety). Johnson & Scott (1976) - Ppts sat in waiting room believing they were taking part in lab study. Heard argument in next room. Low-anxiety condition: man walked out holding pen/greasy hands. High-anxiety condition: breaking glass, man walked out holding paper knife covered in blood
Asked to pick man from set of 50 photos. Low-anxiety: 49% correctly identified. High-anxiety: 33%. Tunnel theory of memory: attention focused on weapon because it is a source of danger/anxiety”
Factors effecting EWT: anxiety (2 - positive effect on recall)
Anxiety triggers fight or flight response (increases alertness/improves memory due to more awareness of cues in situation). Yullie & Cutshall (1986) - Gun shop owner shot thief dead. 21 witnesses, 13 participated in study. Interviewed 4-5 months after event. Accounts compared to police interview. Witnesses rated stress levels at time of incident. Very accurate: little change after 5 months. Some details less accurate e.g. colours/age/weight/height. Highest levels of stress most accurate (88% compared to 75% less-stressed)
Factors effecting EWT: anxiety (3 - explaining contradictory findings)
Yerkes & Dodson (1908): relationship between emotional arousal/performance is inverted U shape. Deffenbacher applied law to EWT. Lower anxiety levels = lower recall accuracy. Accuracy increases with anxiety to optimal point, then drastically declines
Factors effecting EWT: anxiety (weapon focused not due to anxiety)
Johnson & Scott may test surprise rather than anxiety - Ppts may focus on weapon because they are surprised rather than scared. Pickel (1998): scissors/handgun/wallet/raw chicken as hand-held item in hairdressing salon. Scissors = low anxiety, low unusualness. EWT accuracy poorer for high unusualness (chicken/handgun). Weapon focus due to unusualness rather than anxiety/threat. Tells nothing specifically about effects of anxiety on EWT
Factors effecting EWT: anxiety (field studies lack control of variables)
Real-life witnesses interviewed sometime after event. Things happen in meantime that researchers cannot control. Discuss event with others, read/view media accounts (PED). Extraneous variables may be responsible for (in)accuracy of recall, not anxiety. Effect of anxiety may be overwhelmed/difficult to asses by time of interview
Factors effecting EWT: anxiety (ethical issues)
Creating anxiety potentially unethical. May lead to psychological harm purely for research purposes. Real life studies beneficial - interview those already witnessed event, no need to create it. Don’t challenge findings but raise questions of conducting research. Do benefits outweigh costs
Improving the accuracy of EWT: cognitive interview
Fisher & Gieselman (1992) - EWT improved if police use techniques based on psychological insights into how memory works. Cognitive interview = foundation in cognitive psychology. Rapport established with interviewer
Improving the accuracy of EWT: cognitive interview (1 - report everything)
Include every detail (even if seems irrelevant/not confident). Trivial details could be important/trigger other memories
Improving the accuracy of EWT: cognitive interview (2 - reinstate the context)
Return to crime scene in mind. Imagines environment e.g. weather/view/emotions. Based on concept of context dependent forgetting (cues from context may trigger recall)
Improving the accuracy of EWT: cognitive interview (3 - reverse the order)
Recall in different chronological order. Prevents use of expectations rather than actual event. Also prevents dishonesty (harder to produce untruthful account if reversed)
Improving the accuracy of EWT: cognitive interview (4 - change perspective)
Recall from other people’s perspective (another witness/perpetrator). Prevents use of expectations/schema. Schema: packages of information developed through experience, develop framework for incoming data. Create expectations for what would have happened rather than what actually happened
Improving the accuracy of EWT: cognitive interview (enhanced cognitive interview)
Fisher et al: additional elements - Focus on social dynamics (when to establish/relinquish eye contact). Reducing anxiety, minimising distractions, getting witness to speak slowly, asking open ended questions
Improving the accuracy of EWT: cognitive interview (time consuming)
Police reluctant to use - takes much longer than standard interview. More time needed to establish rapport/allow witness to relax. Also requires special training, most forces not been able to provide more than a few hours. Unlikely that proper version is actually used. May explained why police not that impressed by it
Improving the accuracy of EWT: cognitive interview (some elements more valuable than others)
Milne & Bull: each element equally valuable - Also found combination of ‘report everything’ and ‘context reinstatement’ produced better recall than any of the other conditions. At least 2 elements should be used to improve police interviews (even if full CI isn’t used). Increases credibility among police officers
Improving the accuracy of EWT: cognitive interview (support for ECI effectiveness)
Kohnken et al meta analysis: data from 50 studies - ECI consistently provided more correct info than standard police interview. Real practical benefits of using enhanced version of CI. Greater chance of catching/charging criminals = better for society as a whole
