Transcript Yuille and Cutshall (1986)

You’re the psychologist


You can pick either L&P, Pickel or
Yarmey. You must be then and
answer questions from the rest of
the group for 3 minutes.
Rest of group = Field questions to
the psychologist about their study.
Think of all aspects (describe and
evaluate).
Ethical Consideration?

What are the 3 main ethical
considerations when planning a
laboratory or field experiment?



Informed consent and deceit
Protection of participants
Right to withdraw
Yuille and Cutshall (1986)
A case study of eyewitness
memory of a crime.

Can the findings of laboratory based
EWT studies be generalised to more
realistic situations?
Problems With EWT Lab Based
Studies



Slide sequences and filmed events are
not relevant to real work events
Lack of seriousness or consequences
of actual events
There are ethical issues if participants
are deceived into thinking an event is
real.
Problems With EWT Lab Based
Studies



Problems with the participants.
41 research articles about EWT from
1974 – 1982 (38 of them, 92%) tested
students only (according to Brain,
2009).
Therefore, field observation are
required to help generalise lab study
findings.
Background to
Yuille and Cutshall, 1986)
One spring afternoon in Vancouver, Canada, a thief
entered a gun shop, tied up the owner and stole
some money and guns. The owner freed himself
and picked up a revolver. He went outside to take
the car registration number but the thief had not got
into the car and he fired two shots at the store
owner from a distance of about 6 feet. The owner,
after a short pause, fired all 6 shots from his
revolver. The thief was killed but the store owner
recovered from serious injury. Witnesses viewed
the scene from different locations – passing cars,
buildings or in the street.
This case was chosen because…





There were enough witnesses to compare accounts.
The thief was killed and the weapons and money recovered,
so there was a lot of forensic evidence to verify the witness
accounts.
The death of the thief closed the file, so research would not
interfere with a police case.
There were many visible elements to the scene (car, gun
boxes) so eye witness statements could be checked and
compared.
Witnesses could be asked about elements the police would
not have focused on, so previous police questioning would not
interfere with or affect the study results.
AIMS



To record and evaluate witness accounts.
To examine issues raised by laboratory research
To look at witness verbatim accounts – their accuracy and the
kind of errors made.
Loss and distortion of memory takes place over time. The
idea was to look at eyewitness interviews immediately after
the event, which were conducted by a police officer, and to
compare these with interviews carried out by research staff.
Misleading questions were incorporated into the research
interviews to see how an eyewitness might be affected by
distortion.
Participants


21 witnesses interviewed by police.
13 agreed to take part in the research
interviews (two had moved from the
area, five declined and the other was
the victim, who did not want the relive
the trauma).
Interview Procedure

Police had interviewed the witnesses
and recorded the interviews by hand.
Each witness had been asked to
describe the event and the officer then
asked a series of questions. The
reports were verbatim (word for word).
4 – 5 months later………….



1.
2.
13 witnesses interviewed by the researchers –
audio taped and transcribed.
Research interviews used the same format as
police interviews (own account then questions).
But TWO misleading questions:
“a busted headlight” Vs “the busted headlight”.
(There was no broken headlight)
“the yellow quarter panel” Vs “a yellow quarter
panel”. (Quarter panel was blue)
Interview Procedure

Researchers also asked about the
degree of stress each witness
experienced at the time of the
accident. This was a 7 point likert
scale, with 1 being perfectly calm and
7 being extremely anxious. They were
asked about their emotional state
before and after the incident.
Scoring Procedure



Research interviews were compared with
police interviews by using a scoring
technique.
Details were divided into ACTION details
and DESCRIPTION details of either
OBJECTS or PEOPLE.
Various reconstructions were set up and
evidence was carefully researched so actual
details revealed
Results
Action Details
Person
Descriptions
Object
Descriptions
TOTAL
Police
Interview
392
180
Research
Interview
77.5
238
649.50
1056.5
551.5
267
Total reported classifiable details from 13 witnesses
Results


Of the 13 participants, 7 were central
witnesses and 6 peripheral witnesses.
However, both were equally accurate.
In the police interviews 84.56% of the
central witnesses were accurate, compared
with 70.31% of the peripheral group. The
accuracy remained similar and high for most
of the witnesses even after 4 – 5 months
and errors were relatively rare.
Results


Misleading information had little effect
(10 said there was no broken headlight
or no yellow quarter panel, or said they
had not noticed the detail).
Stress did not affect memory
negatively. Researchers found that
witnesses experienced adrenalin more
than stress. The stress came later.
Conclusions


Eyewitnesses are not inaccurate in
their accounts.
Y&C suggest they may have
investigated flashbulb memories. The
fact that those directly involved in the
event remembered more might support
this and explain the difference in
findings from lab studies.
Conclusions

Scoring procedures may have
undermined the accuracy of the
accounts.
For example, “he looked like he was
in his early 20s’ was scored as
incorrect (even though he did look like
he was in his early 20s’) because he
was 35!
Conclusions

The study also shows that just
because some details may be wrong
(such as the colour of a blanket), it
does not necessarily make other
details wrong and that the witness
testimony should not then be rejected.
Strengths


Field study looking at a real incident
with real witnesses. Has validity
lacking in lab experiments.
Researchers took great care with
counting details from real incident to
make sure that witness testimonies did
not alter what ‘really’ happened. This
makes finding ‘reliable’.
Weaknesses


Problems in generalising from a unique and
specific incident. Could be a flashbulb
memory, so may be unfair to use these
findings to criticise lab experiments.
Problems with the scoring (previously
discussed). However, as the accounts were
found to be largely accurate, emphasising
inaccuracies would not have affected the
findings in this case. Turning qualitative
into quantitative data can always lead to
bias and inaccuracy.
YUILLE & CUTSHALL (1986)
FIELD STUDY
1.
Where did the study take place?
2.
The thief fired X shots and the
owner fired X shots.
3.
There were originally X
witnesses. X agreed to the
research interview.
4.
After X months witnesses were
interviewed by the researchers
5.
There were two misleading
questions. One regarded a
‘broken headlight’ the other
one was …
6.
Did stress affect recall?
7.
In conclusion, eye witness
accounts are more
____________ than laboratory
research would suggest.
8.
Why is this study difficult to
generalise?
9.
This study is valid because…
Answers
1.
2.
3.
4.
5.
6.
7.
8.
9.
Vancouver, Canada
2 (thief), 6 (owner)
21 (witnesses), 13 (research interview)
4/5 months
a/the yellow panel
No – stress did not affect recall negatively
Accurate (or something similar will do!)
Difficult to generalise from a unique incident and may have
been a ‘flashbulb’ memory so difficult to compare to lab
exp findings
Valid as it was a field study investigating real life incident