Reliability of memory introduction

Download Report

Transcript Reliability of memory introduction

Write them down
Write them down
Did you note down ‘sweet’ and
‘angry’?
Learning outcome
• With reference to relevant research studies, to
what extent is one cognitive process reliable?
• Our cognitive process will be Memoryspecifically reconstructive memory.
Reconstructive memory
• Schemas
– Knowledge structures that relate to commonly
encountered objects, situations or people
– Enable us to predict events, make sense of
unfamiliar circumstances, organise our own
behaviour
– Act as filters to perception & recall
Computer Information Processing
BANG!
www.psychlotron.org.uk
Can you
wreck a
nice
beach?
Schema Driven Processing
Yes. I can
recognise
speech.
www.psychlotron.org.uk
Can you
wreck a
nice
beach?
Bartlett (1932)
Input
Schema
Output
‘Turf cutter’
Bartlett (1932)
www.psychlotron.org.uk
‘Pickaxe’
Bartlett (1932)
• War of the Ghosts
– Shorter
– Less detailed
– Some details changed (e.g. seal hunting
changed to fishing)
– More ‘Western’ structure
www.psychlotron.org.uk
• When recalled by UK ppt’s:
Allport and Postman (1947)
• Showed ppt’s a picture of a black man and a
white man in an argument.
• White ppt’s incorrectly remembered the black
man as the the one who was holding the
cutthroat razor.
Reasons for inaccuracies in memory
• Reason 1
• Memories are reconstructive (Bartlett 1932
and Allport and Postman 1947) because
information processing is schema driven.
Why does this matter?
• The reliability of memory has interested
psychologists due to its repercussions within
the legal system.
• Court cases are especially dependent on the
reliability of eyewitness testimony (EWT)
Eye-witness testimony (EWT)
• Eye witness testimony is evidence
supplied by people who witness a specific
event or crime, relying only on their
memory.
• Statements often include descriptions of
the criminal (facial appearance and other
identifiable characteristics)
• Subsequent identification
• Details of the crime scene (eg the
sequence of events, time of day, and if
others witnessed the event etc).
How accurate is eyewitness testimony?
• In 1984, American college student Jennifer Thompson was raped at
knifepoint by a man who burst into her flat.
• During the ordeal she concentrated on every detail of her attacker so that
she could later accurately recall him.
• Later that day she worked with police to compose a sketch of her
assailant.
• A few days later she identified Ronald Cotton as the rapist and picked him
out of an identity parade.
• On the strength of her EWT Ronald Cotton was imprisoned. Jennifer was
so sure of his guilt and wanted him electrocuted.
• In 1995, after serving 11 years in prison, DNA evidence proved that Ronald
Cotton was innocent and he was released.
How accurate is eyewitness
testimony?
• In the USA, in more than 100 convictions in
which people have later been shown to be
innocent by DNA testing, 75% of them were
convicted on the evidence of an eyewitness.
• There is the serious prospect that innocent
people have been executed in the USA on the
basis of EWT.
EWT: Schema Driven Errors
• Witnesses to crimes filter information
during acquisition & recall
– Distortions may occur without the witness
realising
www.psychlotron.org.uk
– Their schematic understanding may influence
how info is both stored & retrieved
EWT: Schema Driven Errors
• Past experiences
• Stereotypes & beliefs
about crime &
criminals
www.psychlotron.org.uk
• Assumptions about
what usually happens
Anxiety/Shock
Age
Leading
questions
Individual
differences
Consequences
Time delay
Factors that affect EWT
Leading questions:
This is a question that contains hints about what
the right/desired answer should be.
• Key study: Loftus and Palmer 1974
Aims
• Loftus and Palmer’s experiment was
actually two experiments.
• They wanted to investigate in
general how accurate or inaccurate
memory was. Specifically they
wanted to see the effect of leading
questions upon estimates of speed.
Aims
• Experiment 1
– See if the speed estimates would be
influenced by the wording of the
question asked.
– Hit vs Smashed
Aims
• Experiment 2
– To see if the leading questions just
changed the responses given to the
questions, or whether the participant’s
memories had actually altered as a
result of the leading questions.
Procedures
• Type of study: Laboratory
experiments
• See hand out of Loftus and Palmer 1974
Task
• Find a clip of an incident (download so it can
be played).
• Formulate four questions about the event,
one of these must be a leading question.
Other factors that affect EWT
Interested in Loftus?
• Find out more:
http://faculty.washington.edu/eloftus/
Practical applications
• In the UK, the Devlin Report (1976)
recommended that that no jury should convict
on the evidence of a single eyewitness testimony
alone. Barristers are also not allowed to aske
leading questions for fear of generating false
testimonies.
• In the USA (1990’s) the “Innocence project” has
helped to overturn the wrongful convictions
through faulty EWT of 220 men (by 2008).