L17 - Applying the research - legal Flashcards
Success stories in aiding decisions in the legal system
What impact does eyewitness identification evidence have in the courtroom?
Large increase in conviction rates when ID evidence is present.
Increased when the witness is confident.
What is the main problem with eyewitness testimony?
Eyewitnesses can be mistaken, particularly when no suspect is present
(100% mistaken ID rate in lab tests)
Courts are more likely to convict if the witness is confident in their decision.
Do eyewitness researchers believe that confidence is a reliable indicator of accuracy?
What is the highest correlation between confidence and accuracy that researchers have been able to show?
No
Confidence correlation accuracy = .29
What is the overconfidence bias?
The tendency to be more confident than warranted by accuracy.
When is the overconfidence bias most prevalent?
Overconfidence is greatest when accuracy is at chance levels and the decision is hard
When the question is hard but perceived to be easy.
What was the accuracy compared to confidence in the Lichtenstein & Fischoff (1980) experiment?
Accuracy = 45%
Confidence = 65%
At what level of confidence does overconfidence diminish?
50% to 80%
Do people become under-confident in their decisions at some point?
Yes, at over 80% confidence they start to become less confident.
What is the hard-easy effect?
If people receive easy questions they will report ‘under-confidence’.
If they are given hard questions they will report ‘over-confidence’
How can the overconfidence bias be reduced?
If people are given immediate accurate feedback.
What was the difference in the subjective likelihood of accuracy vs. objective likelihood of accuracy of eyewitness testimony when foils were present compared to when foils were absent? (Brewer & Wells, 2006)
15% Target absent = .01 difference in accuracy
50% Target absent = .04 difference in accuracy
Subjective accuracy is much more likely to be wrong when targets are absent.
Can witnesses give accurate ratings about confidence at the trial?
No, in order to get accurate confidence ratings it must be right after identification.
If a witness takes 30 minutes to identify a suspect, does that mean it is more likely to be accurate?
No, means less likely as memory does not need 30 minutes and more likely they rationalised who was closest to suspect.
What instructions must be given to witnesses and when must results be taken in order to get accurate confidence levels?
They must be given a warning that the perpetrator may not be in the lineup.
Witnesses must be asked immediately after identification for best accuracy.
What is the ‘post-identification feedback effect’?
The tendency for eyewitnesses to alter their confidence after being given either explicit or subtle cues to the accuracy of their identification.
