memory

• coding

  Baddeley: set of acoustically/ semantically dis/similar words - found STM is stored acoustically and LTM semantically

  • S: STM + LTM separate stores - supported by multi-store-memory model and other studies
  • W: application to everyday life - artificial stimuli
• capacity

  Jacob’s digit span test found capacity is 9-3 digits, 7-3 letters

  Miller: everyday observations showed things come in 7s (chunking= 7 +/-2)

  • S: Bopp + Verhaegan: replicated findings
  • W: Jacobs filled with confounding variables + old study (inadequate controls)
  • W: Miller overestimated chunks - Cowan claims its only 4/-2
• duration

  STM:

  Peterson + Peterson: 24 participants, 8 trials - 3 seconds = 80% accuracy, 18 seconds = 3% without rehearsal

  • W: external validity: artificial stimuli’s HOWEVER sometimes we remember meaningless information (phone numbers)

  LTM:

  Bahrick et al: 392 participants 17-74

  face recognition of 50 photos: 15yrs post graduation = 90%, 48yrs = 60%

  free recall: 15yrs = 70%, 48yrs = 30%

  • S: Sheperd: meaningless stimuli = decrease in recall - high external validity as it was meaningful material
  • W: confounding variables
• types of LTM - Tulving

  episodic: events in life (diary), time stamped, several elements to 1 memory, conscious effort to recall

  procedural memroy: automatic action/ skill, hard to explain

  semantic memory: general knowledge of the world, less vulnerable to distortion/ forgetting
• types of LTM evaluations

  • S: clinical evidence: HM + Wearing, HM damaged only episodic memory (didnt remeber stroking a dog but doesn’t need ‘dog’ explaining) HOWEVER lack of control variables: no knowledge of memory prior to injury
  • S: application: Belleville et al - old age memory loss improved by intervention to target episodic memory
  • W: conflicting neuroimaging: Buckner + Peterson show episodic = right of prefrontal cortex, semantic = left + Tulving et al show left prefrontal cortex = episodic retrieval
• multi-store memory model

  sensory register - (attention) -> STM -(prolonged rehearsal, maintenance rehearsal back to STM) -> LTM

  STM <-(retrieval)- LTM

  — coding - capacity - duration

  SR. - modality specific - high - 0.5s

  STM - acoustic - 7+/-2. - 18-30s

  LTM - semantic - infinite - infinite
• multi-store-memory model evaluations

  • S: Baddeley: STM + LTM stored separately - other duration + capacity studies HOWEVER artificial stimuli: mundane realism
  • W: Craik + Watkins: type of rehearsal (elaborate the best) > amount
  • case study: Shallice + Warrington: KF showed oversimplification of STM + LTM types - STM for digits poor when read to him bit better when he read them
• working memory model - Baddeley + Hitch

  central executive: allocates data, limited capacity

  GOES INTO ANY OF THREE:

  phonological loop: phonological store (inner ear) or articulately process (inner voice, 2 second capacity)

  episodic buffer: 4 chunks, ‘diary’ of verbal, spatial, visual info

  visuo-spatial sketchpad: visual cache, inner scribe, inner eye, 3-4 objects (Baddeley)

  ALL LEAD TO LTM
• working memory model evaluations

  • S: Baddeley: dual-task - 1 visual + 1 verbal task easy, 2 visual tasks difficult HOWEVER Baddley: central executive most important but least understood part
  • S: Shallice + Warrington: KF damaged phonological loop not visual-spatial-sketchpad HOWEVER may have had other cognitive impairments
  • W: most supporting studies are artificial
• interference

  proactive interference: old information disrupts new

  retroactive: new information disrupts old

  explanation for (mainly) LTM

  McGeoh + McDonald - similarity: 6 groups learn 2 lists (synonyms, antonyms, unrelated, consonant syllables, 3- digit numbers, control) - synonyms worst recall, control best
• interference evaluations

  • S: Baddeley + Hitch: more rugby games played = worse recall HOWEVER interference is rare - memories have to be fairly similiar
  • S: Coenen + Luijtelaar: diazepam tested on list recall (retrograde facilitation) - Wixted: drug prevents new information reaching precessing memories parts of the brain (can’t interference retroactively)
  • W: Tulving + Pstoka: interference only temporary + over come by cues = recall rose back to 70% with recall
  • W: external validity: most supporting studies lab based - artificial + unrealistic
• retrieval failure

  encoding specificity principle: Tulving - cues at encoding + retrieval to be effective, can be meaningful ( abrieviations) or meaningless (context/ state), used in mnemonic techniques

  contetxt-dependent forgetting: Godden + Baddeley - deep sea divers, learn/ recall on/off land, 40% lower in non-matching conditions

  state-dependent forgetting: Carter + Cassady - learn/ recall on/off antihistamine, high recall in matching conditions
• retrieval failure evaluations

  • S: Eysenck + Keane: wide range of support states it’s the most common reason for forgetting HOWEVER Baddeley: only occurs in extreme condition (underwater)
  • S: real world application: Baddley - cues used in real life (e.g. returning to a room to remember something)
  • W: dependence on the type of memory tested: Godden + Baddeley: no context effects testing recognition - repeated underwater test
  • W: no way to measure how good cue is at encoding
• EWT - anxiety

  positive effects: Yuille + Cutshal - gun shop shooting, 13 witnesses, 7 point scale of stress - high anxiety high recall (physiological arousal triggers fight or flight)

  negative effects: Johnson + Scott - weapon focus pen= 49% vs knife = 33% remember attacker, tunnel theory of memory (physiological arousal prevents recall)

  Yerkes + Dodson law: meta-analysis of 21 studies shows graph going up and then down as anxiety increases
• misleading information

  leading questions: Loftus + Palmer - 45 participants, 5 groups, watch video of car crash, asked how fast they were travelling, verb change smashed (40.5 mph) vs contacted (31.8mph) - response bias or substitution (2nd experiment: some gave reports of broken glass others didn’t)

  post-event discussion: Gabbert et al - pairs watch video from different perspectives, then co-witness discuss - 71% give information they didn’t see - memory conformity (NSI or ISI) or contamination (alters memory)
• misleading information evaluations

  S: application: Loftus - officers need to be careful + psychologists asked to assess validity of EWT HOWEVER issues with research: lab experiments (demand characteristics) + Foster: EWT more real life consequences

  S: Skagerberg + Wright: post- event discussion on hair colour - participants gave a mix of the answers (light brown + dark brown = medium brown)

  W: Sutherland + Hayne: central details unaffected by misleading information
• cognitive interview - Fisher + Geiselman

  1. report everything
  2. reinstate context - context-dependent forgetting
  3. reverse order - expectations
  4. change perspective - expectations/ schema

  enhanced CI - Fisher et al: social dynamics, eye contact, open questions, no distractions

  • Köhen et al: meta-analysed 55 studies - 41% more accurate information than standard HOWEVER increases inaccurate information - even more for ECI
  • W: Milne + Bull: ‘pick and mix’ - flexible, but harder to compare
  • W: Kebbell + Wagstoff: time consuming + unrealistic
• EWT evaluations - anxiety

  • S - support for positive: Christianson + Hubenette - 58 robbery witnesses, high anxiety, high recall HOWEVER interviews 4-15 months after even - confounding variables
  • S - support for negative: Valentine + Mesout measure heart rate to determine anxiety, high anxiety low recalll in london dungeons
  • W - Pickel - weapon focus is from unusualness not anxiety - chicken, sciccors in hair salon
  • W - Yerkes + Dodson law ignored cogntive, behavioural, emotional effects