Multistore model
Coding, Capacity, Duration
Working Memory Model
Long Term Memory
Interference
Retrieval Failure
Anxiety
Leading questions and PED
Cog Interview
Multi-Store Model
A01 Atkinson and Shiffron suggested the multi store model as 3 permanent stores. The sensory register is where
the senses are stored with a duration of half a second and a coding modality specific. Environmental stimuli enter
here.
A01 The sensory store moved to the STM for temporary storage - coded visually or acoustically. The capacity is
5-9 items but it is increased with chunking as proposed by Miller. There is a duration of 30 seconds. Rehearsing
information is done within the rehearsal look which retains information in the STM. LTM is coded semantically
with unlimited duration and capacity.
A03 Glanzer and Cunitz gave participants a list of 20 words one at a time. They were shown to remember the
words from the start (primacy effect) and at the end (recency effect) but not in the middle (serial position effect).
This supports the existence of separate stores.
A03 HM had his hippocampus removed for epilepsy. He developed amnesia and couldn't form LTM memories but
he could recall LTM from before. The hippocampus acts as a memory ‘gateway’ for new memories to pass to the
LTM. So if it is broken, this cannot form. The LTM is seen as a separate store.
A03 It is shown to be oversimplified and single and uniform. It is more complicated though. The WMM by
Baddeley and Hitch shows STM is not unitary with more components, thus not linear.
A03 Oversimplified LTM - Endul Tulving proposed a more complex version. Episodic buffer (events, time stamped
and conscious), semantic memory (common knowledge, rare distortion) and procedural (skills/actions,
automatically learnt through practice, not conscious).
Cognitive interview
Fisher et al - examined real Florida interviews for 4 months - issues with methodology
Brief, direct and close ended questions - witnesses interrupted
Fisher and Geilsman - recall better with retrieval cues
Context reinstatement - returning to scene through environment and emotions
Report everything - state every detail
Recall from change in perspective - other POV disrupts expectation and schema effects
Recall in reverse order - avoid expectations, preventing dishonesty
A03 Limitation - More time is needed to establish rapport with witnesses so they can relax. CI requires special
training - many can’t provide more than a few hours (Kebbel and Waestaff). The proper CI is rarely used so
successful implications aren't widespread. Shortening it for more practicality may be useful - used for serious
crimes.
A03 Limitation - Mello and Fisher compared older and younger adults in their responses to a filmed simulated
crime. CI produces more information than the SI - greater for older than younger. Geilselman reviewed several
studies concluding children under 6 reported less accurate events in CI. Individual differences - CI is useful for
adults, not kids. Police should be cautious with kids.
A03 Support - Kohnken et al - a meta-analysis of 50 studies showed CI provided more correct information than
the SI. CI - special and practical benefits to the police. Greater change of catching criminals.
A03 Support - RWA for retrieval failure as it uses context-related cues. This increases ecological validity
Anxiety Affecting Recall
Yuille and Cutshall - shopkeeper shot thief in Canada
, 21 witnesses (13 took part) - interviewed at time of incident (stress rated on 7 point scale) and 4-5 months later
(emotional problems since event)
Results compared - accuracy = number of details
Little changes e.g age, height, weight etc
High anxiety = higher recall
Johnson and Scott - lab study in waiting room
1) Low anxiety condition - casual conversation next room, man holding greasy pen
2) High anxiety condition - heated argument next room, broken glass sound, man holding bloody knife
PPTs given 50 pictures - 1) 49% accuracy and 2) 33% accuracy
Weapon focus effect due to anxiety
‘Yerkes Dodson Law’ - Inverted U - recall improves with arousal up to a point (the ‘optimum’) but higher arousal
then reduces performance. curvilinear relationship.
Pickel - showed ppts hairdressing salon video - someone holding either scissors, handgun, wallet or raw chicken.
EWT poorer in high unusualness - surprise may explain weapon focus rather than anxiety
Ethical issues - harm and anxiety, unethical, violating protection. Lab useful for comparing to less controlled field
Lab can’t create the same anxiety levels as controlled vs real life studies. Lacks ecological validity - ppts
anticipated something going to happen, affecting their accuracy of judgements. Real life cases have more
extreme anxiety.
Fazey & Hardy (1988) Catastrophe Theory - when physiological arousal increases beyond the optimum level, there is a
dramatic decline (not a gradual one as with the inverted U hypothesis). This is because of increased mental anxiety (worry)
whereas the inverted U only describes physiological anxiety.
Misleading Information and post event discussion
Loftus and Palmer - wording of questions affects eyewitness accuracy
PPTs given film clip of car accident - filled in self report form
‘About how fast were the cars going when they hit each other’
5 groups of 9 - given different verb (‘Hit’, ‘Contacted’, ‘Bumped’, ‘Collided’ and ‘Smashed’)
Contacted - 31.8 mph and Smashed - 40.5 mph
‘Did you see any broken glass?’ - None but those who said faster, more likely to assume
Gabbert - studied ppts in pairs or individually. Watched a video of the same crime (girl stealing money from
wallet) but different POVs. Discussed what they saw then completed the recall test individually. 71% ppts
mistakenly recalled post discussion information. Control group - 0% mistakes. Memory conformity (NSI or ISI).
Actual memory is unchanged
Bodner et al - effects of PED reduced if warned of effects. Recall more accurately when warned not to recall
anything not on film. Police officers know the danger of leading questions in a jury situation.
Yuille and Cutshall - accurate reports of crime 4 months after event, even though given 2 leading questions.
Misleading information has no effect? Consequences as real life study unlike L and P in lab, not generalisable
Low external and ecological validity - artificial tasks not recall car crash, high internal as replicable
Loftus and Pickrell - presented ppts with stories from age 4 and 6. 3 true and 1 false (lost in shopping mall) -
asked to write all events. After a week, 5/24 falsely recalled.
Retrieval Failure
A01 Endul Tulving reviewed research into retrieval failure finding a consistent pattern - called the encoding
specificity principle. A cue has to be present in encoding and present at retrieval and if they are different,
forgetting is more likely to occur. Some cues are encoded in a meaningful way e.g cue ‘STM’ may trigger loads of
Coding, Capacity, Duration
Working Memory Model
Long Term Memory
Interference
Retrieval Failure
Anxiety
Leading questions and PED
Cog Interview
Multi-Store Model
A01 Atkinson and Shiffron suggested the multi store model as 3 permanent stores. The sensory register is where
the senses are stored with a duration of half a second and a coding modality specific. Environmental stimuli enter
here.
A01 The sensory store moved to the STM for temporary storage - coded visually or acoustically. The capacity is
5-9 items but it is increased with chunking as proposed by Miller. There is a duration of 30 seconds. Rehearsing
information is done within the rehearsal look which retains information in the STM. LTM is coded semantically
with unlimited duration and capacity.
A03 Glanzer and Cunitz gave participants a list of 20 words one at a time. They were shown to remember the
words from the start (primacy effect) and at the end (recency effect) but not in the middle (serial position effect).
This supports the existence of separate stores.
A03 HM had his hippocampus removed for epilepsy. He developed amnesia and couldn't form LTM memories but
he could recall LTM from before. The hippocampus acts as a memory ‘gateway’ for new memories to pass to the
LTM. So if it is broken, this cannot form. The LTM is seen as a separate store.
A03 It is shown to be oversimplified and single and uniform. It is more complicated though. The WMM by
Baddeley and Hitch shows STM is not unitary with more components, thus not linear.
A03 Oversimplified LTM - Endul Tulving proposed a more complex version. Episodic buffer (events, time stamped
and conscious), semantic memory (common knowledge, rare distortion) and procedural (skills/actions,
automatically learnt through practice, not conscious).
Cognitive interview
Fisher et al - examined real Florida interviews for 4 months - issues with methodology
Brief, direct and close ended questions - witnesses interrupted
Fisher and Geilsman - recall better with retrieval cues
Context reinstatement - returning to scene through environment and emotions
Report everything - state every detail
Recall from change in perspective - other POV disrupts expectation and schema effects
Recall in reverse order - avoid expectations, preventing dishonesty
A03 Limitation - More time is needed to establish rapport with witnesses so they can relax. CI requires special
training - many can’t provide more than a few hours (Kebbel and Waestaff). The proper CI is rarely used so
successful implications aren't widespread. Shortening it for more practicality may be useful - used for serious
crimes.
A03 Limitation - Mello and Fisher compared older and younger adults in their responses to a filmed simulated
crime. CI produces more information than the SI - greater for older than younger. Geilselman reviewed several
studies concluding children under 6 reported less accurate events in CI. Individual differences - CI is useful for
adults, not kids. Police should be cautious with kids.
A03 Support - Kohnken et al - a meta-analysis of 50 studies showed CI provided more correct information than
the SI. CI - special and practical benefits to the police. Greater change of catching criminals.
A03 Support - RWA for retrieval failure as it uses context-related cues. This increases ecological validity
Anxiety Affecting Recall
Yuille and Cutshall - shopkeeper shot thief in Canada
, 21 witnesses (13 took part) - interviewed at time of incident (stress rated on 7 point scale) and 4-5 months later
(emotional problems since event)
Results compared - accuracy = number of details
Little changes e.g age, height, weight etc
High anxiety = higher recall
Johnson and Scott - lab study in waiting room
1) Low anxiety condition - casual conversation next room, man holding greasy pen
2) High anxiety condition - heated argument next room, broken glass sound, man holding bloody knife
PPTs given 50 pictures - 1) 49% accuracy and 2) 33% accuracy
Weapon focus effect due to anxiety
‘Yerkes Dodson Law’ - Inverted U - recall improves with arousal up to a point (the ‘optimum’) but higher arousal
then reduces performance. curvilinear relationship.
Pickel - showed ppts hairdressing salon video - someone holding either scissors, handgun, wallet or raw chicken.
EWT poorer in high unusualness - surprise may explain weapon focus rather than anxiety
Ethical issues - harm and anxiety, unethical, violating protection. Lab useful for comparing to less controlled field
Lab can’t create the same anxiety levels as controlled vs real life studies. Lacks ecological validity - ppts
anticipated something going to happen, affecting their accuracy of judgements. Real life cases have more
extreme anxiety.
Fazey & Hardy (1988) Catastrophe Theory - when physiological arousal increases beyond the optimum level, there is a
dramatic decline (not a gradual one as with the inverted U hypothesis). This is because of increased mental anxiety (worry)
whereas the inverted U only describes physiological anxiety.
Misleading Information and post event discussion
Loftus and Palmer - wording of questions affects eyewitness accuracy
PPTs given film clip of car accident - filled in self report form
‘About how fast were the cars going when they hit each other’
5 groups of 9 - given different verb (‘Hit’, ‘Contacted’, ‘Bumped’, ‘Collided’ and ‘Smashed’)
Contacted - 31.8 mph and Smashed - 40.5 mph
‘Did you see any broken glass?’ - None but those who said faster, more likely to assume
Gabbert - studied ppts in pairs or individually. Watched a video of the same crime (girl stealing money from
wallet) but different POVs. Discussed what they saw then completed the recall test individually. 71% ppts
mistakenly recalled post discussion information. Control group - 0% mistakes. Memory conformity (NSI or ISI).
Actual memory is unchanged
Bodner et al - effects of PED reduced if warned of effects. Recall more accurately when warned not to recall
anything not on film. Police officers know the danger of leading questions in a jury situation.
Yuille and Cutshall - accurate reports of crime 4 months after event, even though given 2 leading questions.
Misleading information has no effect? Consequences as real life study unlike L and P in lab, not generalisable
Low external and ecological validity - artificial tasks not recall car crash, high internal as replicable
Loftus and Pickrell - presented ppts with stories from age 4 and 6. 3 true and 1 false (lost in shopping mall) -
asked to write all events. After a week, 5/24 falsely recalled.
Retrieval Failure
A01 Endul Tulving reviewed research into retrieval failure finding a consistent pattern - called the encoding
specificity principle. A cue has to be present in encoding and present at retrieval and if they are different,
forgetting is more likely to occur. Some cues are encoded in a meaningful way e.g cue ‘STM’ may trigger loads of
