Eyewitness Testimony
Eyewitness testimony takes place when someone witnesses a crime such as a bank
robbery or a major road accident and is then asked to give testimony / describe what
they saw either as part of a police investigation or in court.
The accuracy of eyewitness testimony is crucial as it is used as evidence in trials. If it is
inaccurate, guilty defendants may be released and innocent ones may be convicted.
Many people assume that eyewitness testimony is accurate as the witness was actually
present when the event took place unlike the investigating police officers, the judge, jury
and lawyers for the prosecution and defence.
Psychologists have identified a number of factors which affect the accuracy of EWT:
- The age of the witness
- Schemas
- Misleading information (leading questions and post-event discussion)
- Anxiety
- How the witness is interviewed
^ Bold: on spec
Problems with EWT can occur during or after the event and can affect all three stages of
the memory process:
- During the encoding stage (e.g., anxiety and the “weapon focus”)
- During the storage stage (e.g., misleading information and post-event discussion)
- During the retrieval stage (e.g., through the use of leading questions or the
cognitive interview)
Misleading Information
The effects of leading questions and post-event discussion on the accuracy of
eyewitness testimony
Misleading information refers to incorrect information given to an eyewitness
following an event. This can take the form of leading questions or can happen during post-
event discussion.
A leading question is a question that is worded in such a way that it might influence
how a respondent answers. Research by Elizabeth Loftus has shown that even subtle
changes to the wording of a question can affect eyewitness testimony.
,Post-event discussion refers to discussion either between co-eyewitnesses (the
eyewitness and other eyewitnesses) or between the eyewitness and interviewing police
officers. Research by Gabbert suggest that post-event discussion can influence the
accuracy of eyewitness testimony as the testimony of an individual eyewitness can be
contaminated by being exposed to the memories of other eyewitnesses or by the
questions asked by police officers.
Loftus and Palmer (1974): Leading questions / post-event discussion
Aim: to investigate the effects of leading questions about car accidents on the
subsequent recall of speed and damage caused.
Experiment 1: Loftus and Palmer
Procedure:
45 students were shown clips of traffic accidents. They were then asked a series of
questions about the film, including one that read ‘About how fast were the cars going
when they ... each other?’ The missing verb was either smashed, collided, bumped, hit
or contacted.
Findings:
More severe verbs resulted in higher estimations of speed
Verb Mean speed estimate
(mph)
Smashed 40.5 = 41
Collided 39.3
Bumped 38.1
Hit 34
Contacted 31.8 = 32
The severity of the verb clearly affects the accuracy of EWT.
Experiment 2: Loftus and Palmer
Procedure:
150 students were shown a short film that included a 4 second scene of a car accident,
which did not include any broken glass, and asked some questions.
- One third of the participants were asked the critical question ‘About how fast were
the cars going when they smashed into each other?’
- A third were asked the same question but with the word ‘hit’
- The final third was not asked about speed.
, A week later the participants were given another questionnaire including the critical
question ‘Did you see any broken glass?’
Findings:
With a more severe verb, participants were more likely to report having seen (non-
existent) broken glass.
Verb condition Response
Yes No
Smashed 16 34
Hit 7 43
Control 6 44
This shows that participants were more than twice as likely to say yes if the leading
question was used.
Conclusion:
The combined findings of the two experiments show that the leading information of the verb in the
speed question affects the way in which the event was represented in memory. Over time, the
misleading information from the question was integrated with the original memory so that, at
retrieval, the memory was reconstructed to include this new information. This resulted in the
inclusion of false memory of the presence of broken glass as well as speed.
Experiment 3: Loftus and Zanni
Procedure:
Participants were shown footage of a car accident and asked either:
- “Did you see a broken headlight?”
- “Did you see the broken headlight?” -> leading question
Participants who had seen a film with no broken headlight were more than twice as
likely to recall having seen one when asked about “the” broken headlight.
Again, this suggests that leading questions can influence answers given by eyewitnesses.
Experiment 4: Loftus
The effects of misinformation that is blatantly incorrect:
Procedure: Loftus showed participants a set of slides that showed the theft of a large red
purse from a handbag. They then read an account of theft that was allegedly written by a
professor of psychology (this information was designed to suggest that the account was
accurate). The account contained a glaringly inaccurate statement that the purse was
Eyewitness testimony takes place when someone witnesses a crime such as a bank
robbery or a major road accident and is then asked to give testimony / describe what
they saw either as part of a police investigation or in court.
The accuracy of eyewitness testimony is crucial as it is used as evidence in trials. If it is
inaccurate, guilty defendants may be released and innocent ones may be convicted.
Many people assume that eyewitness testimony is accurate as the witness was actually
present when the event took place unlike the investigating police officers, the judge, jury
and lawyers for the prosecution and defence.
Psychologists have identified a number of factors which affect the accuracy of EWT:
- The age of the witness
- Schemas
- Misleading information (leading questions and post-event discussion)
- Anxiety
- How the witness is interviewed
^ Bold: on spec
Problems with EWT can occur during or after the event and can affect all three stages of
the memory process:
- During the encoding stage (e.g., anxiety and the “weapon focus”)
- During the storage stage (e.g., misleading information and post-event discussion)
- During the retrieval stage (e.g., through the use of leading questions or the
cognitive interview)
Misleading Information
The effects of leading questions and post-event discussion on the accuracy of
eyewitness testimony
Misleading information refers to incorrect information given to an eyewitness
following an event. This can take the form of leading questions or can happen during post-
event discussion.
A leading question is a question that is worded in such a way that it might influence
how a respondent answers. Research by Elizabeth Loftus has shown that even subtle
changes to the wording of a question can affect eyewitness testimony.
,Post-event discussion refers to discussion either between co-eyewitnesses (the
eyewitness and other eyewitnesses) or between the eyewitness and interviewing police
officers. Research by Gabbert suggest that post-event discussion can influence the
accuracy of eyewitness testimony as the testimony of an individual eyewitness can be
contaminated by being exposed to the memories of other eyewitnesses or by the
questions asked by police officers.
Loftus and Palmer (1974): Leading questions / post-event discussion
Aim: to investigate the effects of leading questions about car accidents on the
subsequent recall of speed and damage caused.
Experiment 1: Loftus and Palmer
Procedure:
45 students were shown clips of traffic accidents. They were then asked a series of
questions about the film, including one that read ‘About how fast were the cars going
when they ... each other?’ The missing verb was either smashed, collided, bumped, hit
or contacted.
Findings:
More severe verbs resulted in higher estimations of speed
Verb Mean speed estimate
(mph)
Smashed 40.5 = 41
Collided 39.3
Bumped 38.1
Hit 34
Contacted 31.8 = 32
The severity of the verb clearly affects the accuracy of EWT.
Experiment 2: Loftus and Palmer
Procedure:
150 students were shown a short film that included a 4 second scene of a car accident,
which did not include any broken glass, and asked some questions.
- One third of the participants were asked the critical question ‘About how fast were
the cars going when they smashed into each other?’
- A third were asked the same question but with the word ‘hit’
- The final third was not asked about speed.
, A week later the participants were given another questionnaire including the critical
question ‘Did you see any broken glass?’
Findings:
With a more severe verb, participants were more likely to report having seen (non-
existent) broken glass.
Verb condition Response
Yes No
Smashed 16 34
Hit 7 43
Control 6 44
This shows that participants were more than twice as likely to say yes if the leading
question was used.
Conclusion:
The combined findings of the two experiments show that the leading information of the verb in the
speed question affects the way in which the event was represented in memory. Over time, the
misleading information from the question was integrated with the original memory so that, at
retrieval, the memory was reconstructed to include this new information. This resulted in the
inclusion of false memory of the presence of broken glass as well as speed.
Experiment 3: Loftus and Zanni
Procedure:
Participants were shown footage of a car accident and asked either:
- “Did you see a broken headlight?”
- “Did you see the broken headlight?” -> leading question
Participants who had seen a film with no broken headlight were more than twice as
likely to recall having seen one when asked about “the” broken headlight.
Again, this suggests that leading questions can influence answers given by eyewitnesses.
Experiment 4: Loftus
The effects of misinformation that is blatantly incorrect:
Procedure: Loftus showed participants a set of slides that showed the theft of a large red
purse from a handbag. They then read an account of theft that was allegedly written by a
professor of psychology (this information was designed to suggest that the account was
accurate). The account contained a glaringly inaccurate statement that the purse was
