Social Effects on Memory Flashcards
Social context: memories have a personal and social significance
Remembering is a social process - cultural transmission of memories involves repeated retelling
Memories can be constructed and reconstructed both by individuals and by larger groups
Subject to SAME SOCIAL PRESSURE as other behaviours, beliefs or decisions
Terminology for social effects on memory
Memory conformity
Social contagion of memory
Effects of co-witness information
Story telling tradition
Share stories with each other telling and retelling of a story - we introduce changes and distortions
Process of construction and reconstruction at level of individual can be generalised to larger groups
“CLASSIC ASCH EXPERIMENT” 1951 - HIGH LEVEL OF CONFORMITY
1 line and 3 lines (A, B, C) - Experiment introduced it as a “vision test”
Experiment in “groups” made up of 1 ppt and 3 confederates
Some confeds gave correct answer, some gave wrong answer
If all 4 confeds give wrong answer, ppts were seen to conform and also give wrong answer - PUBLIC CONFORMITY
The conformity effect (example 1)
Study - view scenes with high expectancy items and low expectancy items (e.g. kitchen)
Two conditions; 15 seconds and 60 seconds
Collaborative recall: in conformity studies, give out intermediate test with confederate - take turns recalling information for some scenes, confed starts introducing CONTAGION ITEMS (never studied items)
Final recall - do test individually - no-one to bias own answers
see conformity here - ppt has accepted answers that were given as contagion items by confederate
More contagion in which items?
HIGH expectancy items especially in FAST (15 second) trials
Why is this?
Participants are going to trust memories of someone else when they have had a shorter amount of time to look
Conformity depends on
Our own beliefs of the strength of the memory trace that . we have
Example 2 (same outcome as 1)
If you are going to tell participants to pay attention as you might be hearing incorrect information presented to you at some point during the experiment
Warning participants of recall errors potentially made by confederates - effect on social contagion
Warnings reduce but DON’T ELIMINATE social contagion
Example 3 (same procedure with exposure to confederate’s written responses)
Test: initial recall test
Contagion phase: participants asked to compare their recall to that of other participants
Final test
Experiment written protocols had 1-4 of protocols containing contagion items
Outcome of example 3
MORE FALSE CONTAGION
MORE EXPOSURE
STRONGER EFFECT
What is a remember judgement?
When participants can explicitly remember the exact time/moment where they acquired to knowledge (encoded item in study phase)
What is a know judgement?
Lower judgement - think you studied it but can’t recall exact moment - aren’t sure how they got this information but have a sense it was studied –> POWERFUL EFFECT
Adding/removing detail to memory
Is it possible to have participants ‘forget items’ they studied as well as falsely remember items that were never studied? YES
NO - easier to implant false memory than make them forget real memory
Study about adding/removing detail - take it in turns - different types of stimuli
Participants get a chance to follow different types of stimuli
words/pictures of cars/pictures of faces
Ex 1: study words and tested in presence of another person (or without person)
If you had a confed with you, they had a chance to recall a word and your turn to say whether a word had been . studied or not
Confed always went first - they had to say whether the given word had been studied previously or not
Results for this study adding/removing detail to memory
13% difference in recall if confederate said that the old word was new - if confederate said this was a new word, participant was likely to say it was a new word
27.57% difference in recall of new words when confeds said the new words had in fact been studied (old)
Conclusion of this study
EASIER to ADD than REMOVE memories
Easier to report new false memory than to omit a true memory (say a new word was an old word - pretend they’d seen it before)
Especially if there is a remember judgement - can’t dissuade you of the accuracy of the memory
Why do we look at these conformity errors?
BLAME CONFORMITY - in the real world, when people have to recall real events for eyewitness testimonies
Blame conformity phenomenon
Your opinions of who is to blame can be influenced by opinions of another person - a co-witness
Attributions of blame are malleable - study - 2 men bumping into each other
Watch a video of the two men bumping into each other
Listen to another EYE-WITNESS describing event - at the end, they blame one man or neither
Post study question
Who was at fault?
Control condition: only you making a judgement
Experimental condition: 1/3 ppts tended to blame man eyewitness indicated was to blame (and these results were checked to see if variable was ppts poor memory of event - it was not)
We trust judgement of another person - influenced by opinion of another eyewitness
Who do we tend to believe?
Are all sources equal?
No - MORE credible sources produce a stronger misinformation effect
Factors influencing credibility and misinformation
Role in the event - innocent bystanders vs. guilty party
Identity/status/competence of co-rememberer
- Partner > stranger
- Psychologist > child
- Careers enhance credibility
- Young > old witness
- People who have MORE time studying event /accident - better memory trace - more believable
- Confidence levels
Confidence
CONCERNING - Human memory is unreliable and social context of recall information influences info itself and confidence attached
Belief is very sensitive to perceived confidence
Confidence bias
We tend to believe believe people who sound more confident - if they are CONVINCED they saw something
It is not a good measure/poor predictor of accuracy
Studies have shown POOR correlation between people’s accuracy of recall and level of confidence
Confidence is malleable
We can boost our own confidence by repeating stuff to ourselves over and over again –> with no change in accuracy –> but FEEL more confident - manipulate confidence in answers
How can we see evidence of confidence malleability in mock police line up
by manipulating what other co-witnesses say
Confidence inflation
If co-witnesses identifies the same suspect as you - or has improbable/unlikely different response
–> your confidence in own recall/response increases
Confidence deflation
If co-witness identifies different suspect/no suspect
You question your own recall
Study (Wells & Bradfield, 1998)
Demonstrated ease at which you can manipulate confidence via FEEDBACK GIVEN
Conditions of study (manipulating confidence)
- No feedback
- Confirming feedback (inflation)
- Disconfirming feedback (deflation)
Study instructions (manipulating confidence)
Participants watched a video showing store robbery
Blurry picture (by design)
Hard to identify subject
The robber wasn’t actually in police line up of mugshots
Ppts had to identify who it was
Correct answer: none of them
Also had to rate their confidence in their answer
Results of study (manipulating confidence)
If they are told to pick out a suspect and given no feedback/confirming/disconfirming feedback 1-7 confidence rating scale: NO feedback = 4 DISCONFIRMING = 3.5 CONFIRMING = 5.4 Shows that confidence on response is influenced by - CO-WITNESS - EXPERIMENTER - LAW ENFORCEMENT OFFICERS
Outcome of Wells & Bradfield
Feedback with disconfirming statement makes ppt less confident on response and thus on trial maybe less believable (lack social contagion) than a more confident (but not necessarily more accurate co-witness)
Collaborative memory (second phenomena after social contagion)
2 papers
Shared pool of memories, knowledge and information of a social group that is significantly associated with the group’s identity
Collaborative inhibition
If we try to remember information with other people
- Presence of those does not benefit recall of more information accurately
We have a belief that doing things in groups/sharing opinions/brainstorming
AIDS memory/productivity - but not necessarily
General method for studying collaborative inhibition
Study separately
Complete memory test together - collaborative groups
Pool individual task answers together - nominal groups (potential performance)
Typical result of studies in collaborative inhibition
Normally find COLLABORATIVE INHIBITION or SUBOPTIMAL GROUP RECALL
On average, people show poorer memory in groups when tested individually
Study 1 (collab inhibition)
Studying and recalling The War of the Ghosts individually and in groups
Results of study 1 (collab inhibition) - quiz on facts of book (bit like a pub quiz)
- Each individual remembers fewer items than all individuals do together
- Collaborative recall is lower than nominal groups
Why does collaborative inhibition happen?
- SOCIAL LOAFING - diffusion of responsibility, Steiner’s model of group productivity, less effort/motivation
- Cognitive explanation - RETRIEVAL DISRUPTION HYPOTHESIS
What is the Retrieval Disruption Hypothesis?
Each individual when you try to remember something, store memory and try retrieve it
Different people rely on DIFFERENT STRATEGIES/retrieval cues
Different people rely on different strategies and retrieval cues
Retrieval cues are not all compatible - retrieval strategies brought up by other people are going to INTERFERE with your own
–> RESULT: RECALL LESS THAN ACTUALLY REMEMBER
Collaborate inhibition - influencing factors
- Organisation of stimuli - if we give you something to study that lends itself to MULTIPLE retrieval strategies - see MORE collaborative inhibition
- Test type - if test RELIES on retrieval strategies e.g. plain recall - then MORE inhibition
Recognition tasks rely on fewer retrieval strategies and thus less inhibition on performance - Group size - more inhibition with larger group
Consolidation and false memory - do false memory rates increase or decrease in groups?
MORE memory distortion in groups
- Study DRM lists (words created from a particular concept/critical word) separately
- Complete perceived group recall (PGR) test
- Complete individual recall tests
Results: PGR with critical word mentioned (confeds): 79% intrusion rate in individual tests
without critical word mentioned: 39% intrusion rate (due to similarity)
without a PGR test at all: 30% intrusion rate
–> FALSE MEMORIES CAN BE TRANSMITTED
If the critical word (which DRM list is based off) is mentioned by confederates in PGR test
Highly likely to intrude word into own recall
Socially-shared retrieval induced forgetting (RIF) study
Two people (a "speaker" and "listener" study new information and practice selected parts of that information The speaker engages in OVERT (open) recall Is the listener a "passive" addressee? Or do they engage in COVERT/discrete recall?
Socially shared RIF results
Expected result; see a RIF effect in speaker - does it have effect on listener?
Speaker recall: RIF effect present
- better memory for items practiced
- poorer memory for items related to the ones they practiced
Listener recall: also had RIF effect - socially shared
Study 2 - speaker and listener encode information from :
Ex 1 - word pairs
Ex 2 - story
Ex 3 - same story recalled jointly - flowing conversation
Ex 1 retrieval practice
Speakers complete cue-target pairs while listener monitors speaker’s responses for
- ACCURACY (errors/correct answer etc)
- FLUENCY (superficial listening condition) - not judging semantics
Ex 1 test (socially shared RIF)
Compare recall of
- Practiced items (Rp+)
- Unpracticed related items (Rp-)
- Unpracticed unrelated items (NRp)
Inhibition/RIF effect - difference between NRp - Rp-
NRp being a control/baseline condition
The listener monitoring for accuracy
showed RIF effect
Listener monitoring for fluency
Speaker has RIF
Listener has NO RIF!! - no inhibition/no difference in recall of unpractised related/unrelated items
Due to not engaging with what (content/semantics) speaker was saying but how they said it
Stories and conversations (and listener there to check speaker accuracy)
Saw listener RIF effect
Conversational setting - difference in results compared to story/cue-word pairs
Engaged with task MORE - more interesting
Stronger learning effects (more practiced items recalled)
Stronger inhibition effects (less unpracticed related items recalled)
Conclusion of socially-shared RIF
People remember what the speaker remembers and tends to forget what the speaker does not mention
