Eyewitness Testimony
Eyewitness Testimony
For example, they may be required to give a description at a trial of a robbery or a road
accident someone has seen. This includes identification of perpetrators, details of the
crime scene etc.
Juries tend to pay close attention to eyewitness testimony and generally find it a reliable
source of information. However, research into this area has found that eyewitness
testimony can be affected by many psychological factors:
Anxiety / Stress
Reconstructive Memory
Weapon Focus
Leading Question
Clifford and Scott (1978) found that people who saw a film of a violent attack
remembered fewer of the 40 items of information about the event than a control group
who saw a less stressful version. As witnessing a real crime is probably more stressful
than taking part in an experiment, memory accuracy may well be even more affected in
real life.
However, a study by Yuille and Cutshall (1986) contradicts the importance of stress in
influencing eyewitness memory.
They showed that witnesses of a real-life incident (a gun shooting outside a gun shop in
Canada) had remarkable accurate memories of a stressful event involving weapons. A
thief stole guns and money, but was shot six times and died.
What Are False Memories?
Although memories seem to be a solid, straightforward sum of who people are, strong
evidence suggests that memories are actually quite complex, subject to change, and
often unreliable. Memories can be reconstructed as people age and also as their
worldview changes. They can falsely recall childhood events, and through effective
suggestion, can even create new false memories. They can be tricked into
remembering events that never happened, or change the details of things that really did
happen. Malleable memory can have especially dire consequences in legal settings;
highlighted areas of interest are children as eyewitnesses, sexual abuse, and
misidentification.
A person’s malleable memories can entail the very mundane, such as when you second-
guess whether you really did turn off the stove. Some memories can entail the crucial, with
foggy eyewitness recollections of a crime, perhaps. Research shows that we can be given
false information and convinced to believe that an event actually occurred, even if we don't
remember any such event ever happening. Given that recovered memories may be
genuine, false, or a combination of the two, it is legitimate to question just how much of what
The theory of Flashbulb memory was first proposed by Brown & Kulik (1977). They
moment when a surprising and emotionally arousing event happened. They postulated
biological memory mechanism that, when triggered by an event exceeding critical levels
the experience. This implies that flashbulb memories have different characteristics than
"ordinary memories." They also argued that memories are resistant to forgetting.
Brown and Kulik suggested that there may be a special neural mechanism which
emotional events are better remembered than less emotional events—perhaps because
of the critical role of the amygdala. Today the most commonly accepted model of
flashbulb memory is called the importance-driven model. This model emphasizes that
know that an event is important until later, so it is unclear how flashbulb memories could
be created at the moment of the event. He suggests that the memories are so vivid
because the event itself is rehearsed and reconsidered after the event. According to
Neisser, what is called a flashbulb memory may simply be a well-rehearsed story. The
such as place (where were we?), activity (what were we doing?), informant (who told
On 28 January 1986, the seven astronauts aboard the Space Shuttle Challenger were
killed in a tragic accident. It was a shocking experience for those who watched the
students’ memory accuracy of the incident 24 hours after the accident, and then again
two and a half years later. When filling out the second questionnaire which asked
questions like - where were you when you heard the about the Challenger
disaster? Who were you with? What were you doing? The participants were also asked
The researchers found that in a collectivistic culture like China, personal importance and
intensity of emotion played less of a role in predicting FBM, compared with more
rehearsal of the triggering event compared with participants from other cultures - and
thus a lower chance of developing an FBM. However, it was found that national
Reconstructive Memory
dependent on our learned or cultural norms and values, and the way we make sense of
our world.
Many people believe that memory works something like a videotape. Storing
information is like recording and remembering is like playing back what was
recorded. With information being retrieved in much the same form as it was encoded.
However, memory does not work in this way. It is a feature of human memory that we
do not store information exactly as it is presented to us. Rather, people extract from
In other words, people store information in a way that makes the most sense to
them. We make sense of information by trying to fit it into schemas, which are a way of
organizing information.
Schemas are mental 'units' of knowledge that correspond to frequently encountered
order that we can predict what is going to happen and what we should do in any given
situation. These schemas may, in part, be determined by social values and therefore
prejudiced.
information in order to ‘fit in’ with our existing knowledge or schemas. This can,
Bartlett tested this theory using a variety of stories to illustrate that memory is an active
The aim of the research was to investigate whether the use of leading questions would
their introduction and argue that some questions are more liable than others to influence
estimates than others. Leading questions are defined by Loftus as a question which
inaccurate and that there is a tendency to overestimate the duration of a complex event.
1969). Since previous research had shown that estimation of speed was liable to
distortion Loftus and Palmer hypothesized that people’s memory for details of a
complex event could be distorted if they were asked to estimate how fast the car was
going. Therefore they set up two experiments where participants were shown videos of
traffic accidents and after that they had to answer questions about the accident. The
participants were asked about the speed of the car in different ways. For example,
participants were asked:” About how fast were the cars going when they smashed into
each other?” or they were asked: “About how fast were the cars going when they hit
each other?” This was based on the assumption that ‘hit’ and ‘smashed’ have different
connotations and activate different cognitive schemas about the severity of the accident.
participants.
The researchers predicted that using the word ‘smashed’ would result in higher
estimation of speed than using the word ‘hit’. The independent variable was the different
words used in the critical question and the dependent variable was estimation of speed.
45 students participated in the experiment. They were divided into five groups of nine
students. Seven films of traffic accidents were shown and the length of the films ranged
from 5 to 30 seconds. These films were taken from driver’s education films.
When the participants had watched a film they were asked to give an account of the
accident they and seen and then they answered a questionnaire with different questions
on the accident with one question being the critical question where they were asked to
estimate the speed of the cars involved in the accident. There was one critical question
which was the one asking the participant to estimate the speed of the cars involved in
the accident. The participants were asked to estimate the speed of the cars. They were
asked the same question but the critical question included different words. Nine
participants were asked, “About how fast were the cars going when they hit each
other?" The critical word "hit’" was replaced by ‘collided’, ‘bumped’ or ‘smashed’ or’
contacted’ in the other conditions which each had nine participants answering the
question.
The mean estimates of speed were highest in the ‘smashed’ condition (40.8 mph) and
lowest in the ‘contacted’ group (31.8 mph). The researchers calculated a statistical test
Smashed 40.8
Collided 39.3
Bumped 38.1
Hit 34.0
Contacted 31.8
The results indicate that the critical word in the question consistently affected the
participants’ answer to the question. The researchers argued that it may be that the
different speed estimates is the result of response-bias, i.e. the participant is uncertain
about the exact speed and therefore a verb like "smashed" biases his or her response
towards a higher estimate. It may also be that the way the question is formed result in a
change in the participant’s mental representation of the accident, i.e. the verb
"smashed" activates a cognitive schema of a severe accident that may change the
reconstruction so that it is not the actual details of the accident that are remembered but
rather what is in line with a cognitive schema of a severe accident. This interpretation is
Evaluation
The experiment was conducted in a laboratory and the participants were students. Lab
experiments may be problematic in the sense that they do not necessarily reflect how
people remember in real life. There may be a problem of low ecological validity. A
support for this point could be that the films shown in the experiment were made for
teaching purposes and therefore the participants did not experience the same kind of
emotion that they would have experienced if it had been a real accident.
controlled so that it is really the effect of the independent variable that is measured. This
was the case in this experiment and Loftus and Palmer could rightfully claim that they
intensity of the critical word) and the dependent variable (estimation of speed). The fact
that the experiment used students as participants has also been criticized because
students are not representative of a general population. In addition, they were most
likely young and inexperienced drivers, so this may have influenced their ability to
estimate the speed of the cars. That being said, most people would have a problem with
The aim of the study was to determine whether leading questions would affect memory
of eyewitnesses at a real crime scene. In other words, the aim was the same as Loftus
& Palmer's (1974) study, but in this case, it would be a field study done under
naturalistic conditions.
The crime scene was in Vancouver. A thief entered a gun shop and tied up the owner
before stealing money and guns from the shop. The owner freed himself, and thinking
that the thief had escaped, went outside the shop. But the thief was still there and shot
him twice. Police had been called and there was gunfire - and the thief was eventually
killed. As the incident took place in front of the shop, there were eyewitnesses - 21 were
interviewed by the police.
The researchers chose this incident to study because there was enough witness and
there was forensic evidence available to confirm the stories of the eyewitnesses.
The researchers contacted the eyewitnesses four months after the event. 13 of the
eyewitnesses agreed to be interviewed as part of a study. They gave their account of
the incident, and then they were asked questions. Two leading questions were used.
Half the group was asked if they saw a broken headlight on the getaway car. The other
half was asked if they saw a yellow panel on the car (the panel was actually blue). They
were also asked to rate their stress on the day of the event on a seven-point scale.
It was found that eyewitnesses were actually very reliable. They recalled a large amount
of accurate detail that could be confirmed by the original police reports. They also did
not make errors as a result of the leading questions. 10 out of 13 of them said there was
no broken headlight or yellow quarter panel, or that they had not noticed those particular
details.
The researchers found that the accuracy of the witnesses compared to the original
police reports was between 79% and 84%. It appears that this research contradicts the
study by Loftus & Palmer (1974). It could be that the lack of emotional response to the
video that was shown in their study played a key role in the influence of the leading
questions. The witnesses reported that they didn't remember feeling afraid during the
incident, but they did report having an "adrenaline rush."
Evaluation:
The study was a field study and thus has very strong ecological validity. Note, it is not
a natural experiment. In a natural experiment, the IV (the leading question) would have
to be manipulated by environmental forces beyond the control of the researcher. In this
study, the researchers manipulated the variable. The study does use a purposive
sample - that is, the sample is chosen because they were eyewitnesses to the
crime. Only people that meet that requirement are eligible to be participants in this
study.
There was archival evidence (police records of the original testimonies) to confirm the
accuracy of the memories.
The study is not replicable and also not generalizable since it was a one-off incident.
There was no control of variables, so it is difficult to know the level of rehearsal that was
used by the different eyewitnesses. It could be that those who agreed to be in the study
had spent the most time thinking and reading about the case.
Because the eyewitnesses' safety was threatened, it could be that this is a case of
flashbulb memory, which would mean that it cannot be directly compared to Loftus's
original research.
There was an attempt at deceiving the participants. As consent was given by all
participants, the idea that undue stress or harm would be caused by being asked to
recall the incident is unfounded.
The quantification of the qualitative responses from the participants is problematic and
may be open to researcher bias.
The aim of the study was to determine whether flashbulb memories are susceptible to
distortion.
On the morning after the Challenger disaster – less than 24 hours after the event - 106
questionnaire at the end of the class. They were asked to write a description of how
they heard the news. On the back of the questionnaire was a set of questions:
How did the person who told you seem to feel about it?*
2 ½ years later they were given the questionnaire again. 44 of the original students - 30
women and 14 men - were now seniors at the university. They were not told the
purpose of the study until they arrived. They were given the original questionnaire to fill
in again. This time they were also asked for each response to rate how confident they
were of the accuracy of their memory on a scale from 1 (just guessing) to 5 (absolutely
certain).
They were also asked if they had filled out a questionnaire on this subject before.
Seeing that there were discrepancies, semi-structured interviews were carried out a few
months later in order to determine if the participants would repeat what they had written
a few months earlier or revert to the original memory. The interviews were taped and
transcribed. The interviewer presented a prepared retrieval cue with the hope of
prompting the original memories. Participants whose 1988 recall had been far off the
mark were given a cue based on their original records; for example, the interviewer
might ask “Is it possible that you already knew about the explosion before seeing it on
television?”
At the end of the interview, the participants were shown their original 1986 reports in
The researchers were surprised to see the extent of the discrepancies between the
original questionnaire and the follow-up 2 ½ years later. Here is a typical example:
24 hours after the accident: I was in my religion class and some people walked
in and started talking about it. I didn’t know any details except that it had
exploded and the schoolteacher’s students had all been watching which I thought
was so sad. Then after class, I went to my room and watched the TV program
2.5 years later: When I first heard about the explosion I was sitting I my
freshman dorm room with my roommate and we were watching TV. It came on a
news flash and we were both totally shocked. I was really upset and I went
In order to come up with a “score,” the researchers looked at the seven “content”
questions – that is, not the two that are about emotion (see asterisks above) – and gave
a point if they matched the original response. The maximum total response was then
seven.
The mean score was 2.95/7.0. Eleven participants scored 0. Twenty-two of them scored
2 or less. Only three participants scored the maximum score of 7. What is interesting is
that in spite of the lack of accuracy, the participants demonstrated a high level of
confidence. The average level of confidence for the questions was 4.17.
For the most part, participants told the same story in the spring as in the fall, when they
were interviewed. Additional cues had little effect on accuracy. When presented with the
original questionnaire, participants were surprised and could not account for the
discrepancies.
Evaluation
The study was a case study. The strength of this method is that it was both
questionnaires and interviews were used. The limitation is that it cannot be replicated.
In addition, there was participant attrition - that is, participants who dropped out of the
variables and the study was not done under highly controlled conditions.
The study was naturalistic. Although this is good for ecological validity, it is difficult to
eliminate the role of confounding variables. There was no control over the participants'
behaviour between the first questionnaire and the second. We have no idea how often
this memory was discussed or how often the participants were exposed to media about
the event.
It is possible that the confidence levels were higher than they should have been as a
result of demand characteristics - that is, since the participants were asked to verify
their level of confidence, they could have increased their ratings to please the
researcher or avoid social disapproval for claiming not remember an important day in
As mentioned in the background section above, there are several studies of different
events - like September 11th - which seem to have the same results. This