ALL STUDIES

• Coding research- Alan Baddeley (1966)

  • Gave 4 groups different lists of words to remember
  • Acoustically similar, acoustically dissimilar, semantically similar, semantically dissimilar
  • When asked to recall immediately (STM) ppts did worse with acoustically similar words
  • When asked to recall after 20 minutes (LTM) ppts did worse with semantically similar
  • Therefore, STM coded acoustically and LTM coded semantically
• Capacity research- Jacobs (1887)

  • Measured digit span
  • Researcher gives 4 digits for ppt to recall in correct order aloud
  • If correct researcher adds another digit until ppt fails to recall
  • Mean span for digits = 9.3 items
  • Mean span for letters = 7.3
• Chunking- Miller (1956)

  Capacity of STM is about 7 +- 2 (5 to 9 items)
• Duration of STM research- Peterson and Peterson (1959)

  • Tested 24 undergraduate students, 8 trials
  • Student given a trigram (e.g. YCG) to remember alongside 3 digit number
  • Student then asked to count backwards from 3 digit number to prevent maintenance rehearsal
  • Each trial, asked to stop at a different time- 3, 6, 9, 12, 15, 18s (retention interval)
  • The longer the interval, the less ppts could recall correctly. Therefore STM duration is around 18s max
• Duration of LTM research- Bahrick et al (1975)

  • 392 ppts from Ohio
  • Between 17-74 to test photo recognition and free recall
  • PR- 50 yearbook photos from highschool
  • Within 15 years of graduation = 90% accurate
  • After 48 years, 70% accurate
  • FR- ppts recalled all the names of their graduating class
  • After 15 years, 60% accurate
  • After 48 years, 30% accurate
  • LTM duration has no limit, long time
• Against chunking- Cowan (2001)

  • Reviewed other research and suggested Miller overestimated
  • STM capacity is around 4 chunks
  • Miller's lower end of the estimate may be more appropriate (5 instead of 7 items)
• MSM- Atkinson and Shriffin (1968/1971)

  • Memory is made up of 3 stores
  • Sensory register, contains iconic and echoic memory
  • STM (requires maintenance rehearsal to pass info to LTM)
  • LTM (info transferred back to be recalled from STM through retrieval)
• Case study- HM

  • Tissue removal from the hippocampus to relieve epilepsy
  • He suffered from anterograde amnesia after the procedure meaning he could not form any new memories
  • He also had retrograde amnesia, losing the ability to recall events a few years before his surgery
  • Despite this he performed well on tests of immediate memory span (STM)
• Case study- KF

  • Studied by Shallice and Warrington to suggest there is more than one type of STM
  • KF suffered from amnesia
  • His STM for digits was poor when read aloud to but recall was better when reading the digits to himself
  • This may suggest there could be another short-term store for non-verbal sounds
  • Therefore the MSM is a limitation, the STM may have separate stores to process auditory vs visual info
• More than one type of rehearsal- Craik and Watkins (1973)

  • There are 2 types of rehearsal, maintenance and elaborative
  • Elaborative is needed for LTM storage as it links information gained to existing knowledge you have previously stored
• Case study- Clive Wearing

  • Suffered from a damaged hippocampus (from herpes encephalitis)
  • Led to anterograde amnesia and retrograde amnesia
  • Clive is able to remember some memories such as he has children, he recognises his second wife etc.
  • He is still able to play piano and read music, however does not remember learning
  • This suggests and supports the idea that LTM has separate stores which are stored in different areas of the brain
• Working Memory model

  Baddeley and Hitch (1974)
• Dual task performance support- Baddeley et al

  • Ppts struggled with performing two visual tasks (tracking a light and describing the letter F)
  • Rather than a visual and verbal task at the same time
  • Increased difficulty due to the visual tasks competing for the slave system (visuo-spatial sketchpad)
  • Therefore visual stimuli have a separate slave system to be processed by (supports WMM)
• Effects of similarity- McGeoch and McDonald (1931)

  • Studied retroactive interference (RHINO new info interferes with old)
  • 6 groups tested on: synonyms, antonyms, words unrelated, consonant syllables, 3 digit numbers and no new list (retested)
  • Changed the amount of similarity between 2 sets of materials
  • Ppts learned a list of 10 words until they could recall with 100% accuracy
  • They then learned a new list
• McGeogh and McDonald conclusion

  • When ppts recalled the original list of words, performance depended on the nature of the second list. The most similar material (synonyms) produced the worst recall. Interference is therefore strongest when memories are similar
• Real-life study to support interference- Baddeley and Hitch (1977)

  • Asked rugby players to remember team names they played week by week
  • Some players had missed games so the last team they played may have been 2/3 weeks ago
  • Results showed accurate recall did not depend on how long ago the matches occurred but the number of games played in the meantime
  • A player's recall of a team from 3 weeks ago was better if they had played no matches since then. Therefore supports interference theory in real life
• Context-dependant forgetting- Godden and Baddeley

  • Studied deep sea divers
  • Divers learned a list of words on land or underwater and were asked to recall the list in one of 4 conditions
  • (e.g. learn on land, recall on land, learn on land, recall underwater etc.)
  • In non-matching conditions, recall was 40% lower
  • Since the external cues were different when recalling, ppts experienced retrieval failure
• State-dependant forgetting- Carter and Cassaday (1998)

  • Gave ppts anti-histamine drugs for a mild sedative effect
  • Ppts were given words and passages of prose to memorise and recall
  • 4 conditions (e.g. learn on drug, recall not on it)
  • Non-matching conditions had significantly worse recall resulting in retrieval failure from absent cues
• Leading questions- Loftus and Palmer (car crash study)

  • Students watched film clips of car accidents and had to respond to questions on it
  • The critical question (a leading question) asked ppts 'About how fast were the cars going when they hit each other?'
  • 5 groups, each had a different verb
  • Hit, contacted, bumped, collided, smashed
  • Contacted -> mean speed of 31.8 mph
  • Smashed -> 40.5 mph
• Post-event discussion- Fiona Gabbert et al.

  • Studied ppts in pairs
  • Each ppt watched a video of the same crime but filmed from different POVs
  • Some ppts could see elements/details that others could not (e.g. seeing a book title a woman was carrying)
  • Both ppts then discussed what they had seen, then individually completed a test of recall
  • 71% of ppts mistakenly recalled details they did not see in the video but had picked up from discussion
  • Control group (no discussion) 0%
• Individual differences in EWT- Anastasi and Rhodes(eval point)

  • People between 18-25 and 35-45 were more accurate than people in 55-78 years
  • Suggests older people are less accurate than younger when giving eyewitness reports
  • However, all age groups were more accurate when identifying people of their own age group (own age bias)
• Effects of anxiety (negative) Johnson and Scott

  • Led ppts to believe they were taking part in a lab study
  • Ppts were sat in a waiting room and heard an argument in the next room
  • Low-anxiety condition: man walks out of room with a pen and grease covered hands
  • High-anxiety condition: man walks out after breaking glass, paper knife covered in blood
  • Ppts were then asked to pick out the man from a set of 50 photos
• Johnson and Scott findings

  • Low anxiety: 49% successfully identified
  • High anxiety: 33%
  • Supports tunnel theory- argues that a witnesses' attention narrows to focus on a weapon as it is a source of anxiety
• Effects of anxiety (positive)- Yuille and Cutshall (1986)

  • Real life shooting in a gun shop, Vancouver, Canada
  • 21 witnesses, 13 agreed to take part in study
  • Interviewed 4-5 months after incident
  • Interviews were compared to the original police interviews after the shooting occurred
  • Witnesses were asked to rate stress levels during the incident using a 7-point scale
  • Also asked if they had emotional problems since event, e.g. sleeplessness
• Yuille and Cutshall findings

  • Witnesses were very accurate in their accounts, little change in amount/accuracy after 5 months
  • Some details were less accurate in the recollection of item colour, age/height/weight estimates
  • Ppts who reported highest levels of stress were most accurate (88%)
  • Less stressed, 75%
• Weapon focus challenge- Pickel (1998) EVAL POINT

  • Conducted an experiment using scissors, handgun, wallet, raw chicken as hand held items in a salon
  • Low anxiety condition = low unusualness (e.g. scissors/wallet)
  • EWT accuracy was significantly poorer in the high unusualness conditions
  • Weapon focus is due to unusualness rather than anxiety
• Support for ECI effectivity Kohnken et al. (EVAL POINT)

  • Meta analysis
  • Combined data from 50 studies
  • The enhance CI consistently provided more correct information than the standard interview used by police
  • ECI could therefore be beneficial to society to help with crime/EWT
• Cognitive interview


  • Fisher and Geiselman (1992)
  • Report everything, reinstate the context, reverse the order, changing perspective