That's What it Looked Like to Me Quiz

Answer: Eyewitness misidentification

Elizabeth Loftus and Katherine Ketcham have performed a significant amount of research into eyewitness testimony and it effects on the criminal justice system. In a book on the subject, they estimate that eyewitness testimony alone accounts for 45% of wrongful convictions. This figure, 45%, is the highest of any other single factor involved in wrongful conviction, and other experts have suggested that this may be a conservative estimate.

Indeed, in the first 225 cases where DNA evidence was used to exonerate those wrongly convicted, it was shown that eyewitness testimony had been involved in over 77% of those original convictions - this significantly higher number is probably a result of a combination of factors being involved in the original conviction.

Answer: That pre-trial publicity had contaminated the eyewitness' recall

This case was one of the first instances of a psychologist being used as an expert witness.
The trial was of a man accused of murdering three women. Von Schrenk-Notzing argued that eyewitnesses could not clearly distinguish between what they had actually seen and what they had seen in the press. Von Schrenk-Notzing was a doctor of psychiatry and a qualified psychologist whose primary area of interest was investigating psychical phenomena such as telekinesis.

Answer: The language used during questioning

Loftus & Palmer showed participants videos of a number of car accidents. They then tested whether the use of different words would affect the recall of the incidents. Participants were asked to estimate the speed of the cars using a number of different verbs to describe the collision. The results were significant.
Participants, who were asked 'About what speed were the cars going when they smashed into each other?', gave an average estimated speed of 40.8 mph. Participants, who were asked 'About what speed were the cars going when they contacted each other?', gave an average estimated speed of 31.8 mph.
Loftus and Palmer gave two possible explanations - either the use of the verb (smashed, bumped, contacted etc.) distorted the actual memory or there was a response bias, where participants attempt to give the answer they think is expected, even if they can't recall or are not sure.

Answer: Duration of the event

In the experiment, participants engaged police officers in conversation for either fifteen or thirty seconds. Thirty seconds after the conversations were finished; experimenters approached the police officers with ten simple questions about the person, with whom they had had the conversation, asking about their visual appearance.

It showed that where the conversation had lasted longer, the recall was far greater. Clifford and Richards continued the experiment, to determine whether police officers performed better than civilians when recalling an event.

They found that their recall for specific details was slightly better when asked for those details but in 'free recall' they performed about the same as civilians.

Answer: She removed the eyewitness testimony

This was an decrease of 54% in the rate of conviction. Despite its problems, eyewitness testimony impacts strongly on jurors. Without it, the defendent was perceived as not guilty four out of five times. Even when Loftus did give students a version where the eyewitness was discredited, but maintained that they were being honest, the conviction rate only dropped to 68%.
DNA evidence was not used as forensic evidence until the late 1980s and therefore wasn't part of the experiment.

Answer: They gave a warning the culprit might not be present

They split participants into two groups. One group were given the warning that the culprit 'may or may not be present'. Where the culprit was absent from the line-up but witnesses were not warned, 78% of them attempted and made an incorrect identification.

When the culprit was present and witnesses were given the 'may or may not be present warning', 87% of them still made an accurate identification. This led Malpass and Devine to conclude that the warning didn't just make witnesses reluctant to make an identification but rather that they simply use relative judgment, choosing the person in the line-up who more similar to the culprit than other line-up participants.

Answer: True

In 1998 Gary Wells and Amy Bradfield carried out a study in which participants were asked to watch a low quality video of a crime and then make an identification from a photo line-up. The perpetrator was not in the line-up but participants weren't warned.

The groups were split into three groups, some given no feedback, some given positive feedback, such as 'Yes, that's him' and some given negative feedback and told they were wrong. Less than 15% of those given negative feedback rated their level of confidence in their identification as a 6 or 7 out of 7. Of those given positive feedback, that number jumped to 50%

Answer: Weapon focus

The term 'weapon focus' was demonstrated by an experiment by Johnson and Scott in 1976. Participants were witness to either staged conversations or staged altercations. In the altercations they witnessed a person holding a 'bloody' letter opener. In the conversations they witnessed a person holding a pen.

When questioned, few participants recalled the pen but virtually all of those who saw the 'altercation' recalled the letter opener. Further, those who witnessed no 'weapon' were able to identify the person 49% of the time.

This dropped to 33% when a weapon was present.

Answer: High stress interrogation

The study was carried out by Charles Morgan and involved subjecting trained military personnel to a forty minute interrogation in either a high or low stress manner. The following day, all of the participants were shown photographs and asked to identify their interrogator.

Despite their training, stress has a detrimental effect on memory. Only twelve percent of those in a low stress situation made an incorrect identification.

Answer: False

There are many ways to increase the reliability of an eyewitness testimony. One such way is the warning that a perpetrator may or may not be present in a line-up.
Another is the 'double blind' line-up procedure, where the law enforcement official who is administering the line-up has no information about any of the participants in the line up - in this way there is no chance that they will influence the eyewitness either before or after the fact.
A further suggestion has been to have eyewitnesses make a statement of confidence in their identification, as some witnesses feel they must pick someone but if they are only 10 or 20 percent confident, the police can take this into account.