Skip to comments.How Accurate is Eyewitness Testimony? (Conspiracy Buffs Take Notice!)
Posted on 12/03/2001 9:11:58 PM PST by Theresa
In many criminal cases eyewitness testimony can be an extremely important part of the evidence which is considered. Police and juries are more likely to believe witnesses who identify someone in a lineup with a high degree of confidence than witnesses who say they are less sure. Unfortunately, how confident people are about making an identification doesn't necessarily reflect how accurate their identification is.
In fact, a witnesss degree of certainty is quite malleable and can easily become "inflated", according to SSHRC-funded research by psychologist Elizabeth Brimacombe.
She found that confidence levels can be influenced by external factors that have nothing to do with the witness's actual memories or perceptions of the event.
In her research, Brimacombe staged a theft for pairs of eyewitnesses, then separately asked each of them to identify the "thief" in a photo lineup. After making their identifications, witnesses who were then told that the other person had identified the same suspect, felt increased confidence in their own identification. Witnesses who were told their co-witness either identified someone else or made no identification expressed less confidence. Interestingly, when some witnesses were subsequently told they had been misinformed about the co-witness statements, their confidence levels did not return to their original level.
"People tend to persevere even if you change the information," said Brimacombe.
Telling subjects whether their identification was corroborated by fingerprint evidence had an even more pronounced effect on their degree of confidence. It went up emphatically when the fingerprints matched and plummeted when they didnt. However, if the subjects were subsequently told the fingerprint information they were given was wrong, their confidence went back up to its original level.
In another experiment, subjects were shown videotaped "testimony" of witnesses who had identified a suspect from a photo lineup. Brimacombe found that eyewitnesses who believed that their co-witness's identification corroborated their own were perceived by those who viewed the videotapes to be more accurate, more persuasive, have had a better view and give better descriptions of the thief than the eyewitnesses who had been told their co-witness's identification differed from theirs. On this basis Brimacombe argues that the malleability of witness confidence has important implications for how judges and juries interpret an eyewitnesss testimony.
Brimacombe noted that an eyewitnesss confidence can also be inflated if the police officer conducting the lineup expresses satisfaction with the choice the eyewitness makes. "What if, immediately following an eyewitness's identification of someone from a lineup, someone says, 'You've got him! That sleazeball has been doing this stuff for months and now we've nailed him.' There is nothing to prohibit such behaviour by police, but it raises the question to simply ask the eyewitness how confident he or she is that he or she identified the right person."
Other factors in the pre-trial period that can inflate witness confidence include media coverage, pre-trial preparations and even just the fact that the person the eyewitness identified in the lineup is actually charged.
"Pre-trial publicity can be quite biased and people are ripe to be influenced," says Brimacombe. She also commented on another study that shows briefing witnesses about what questions to expect on the stand increases confidence. "Before they come to trial, they've retold that story quite a few times, embedding it in their memory," she said. Limiting the number of times they're asked to recount the story would be impractical, says Brimacombe. Instead she suggests making a record of eye witnesses' statements about their confidence immediately after they identify someone in a lineup. "This forces the eyewitness to base his or her confidence assessment on memory alone."
Brimacombe suggests that to prevent witness confidence from getting inflated the following measures be taken: have lineups conducted by police officers who do not know which person is the suspect; record the confidence level expressed by the witness when identification is first made and present that confidence statement at trial; and educate police, witnesses and juries about external factors that can affect eyewitness confidence and about the fact that confidence does not necessarily reflect accuracy.
"This doesnt prevent police or attorneys from introducing confidence-altering information at a later time, but if the initial confidence assessment is recorded, then opposing counsel can ... raise questions as to why the eyewitness's confidence has increased in the meantime."
For example the article in Newsmax about flight 587. An eyewitness was quoted....
"No tail fell off, not before the explosion. I swear to that," Lynch told the paper's Steve Dunleavy. The eyewitness said there was absolutely no doubt about what he saw. "
"Absolutely no doubt." Yeah, uh huh, sure. I imagine he's being sincere...but is he really right? Don't take it to the bank.
I also expect that eyewitness confidence can be inflated if a REPORTER expresses satisfaction with the witnesses side of the story.
That's "channeling", of course, not "canneling".
This has been behind many of the recent cases where inmates were released because of DNA testing. If the real perp is in a lineup, overwhelmingly, the right person is picked..but any witness viewing any lineup that DOESN'T have the perp can lead to disaster.
Contributing to this is the well-documented race problem...."They all look alike" is considered a racist statement, but when people view people of other ethnicities or races, they have huge problems. It's been documented asians have difficulties telling whites and blacks apart, whites have trouble telling asians and blacks apart, and black Africans actually would have difficulty, say, telling swedes and Afghanis apart, though the differences are obvious to us. This is made worse when you have little contact with the race in question. Until I went to college, I would have not been able to tell Vietnamese from Chinese from Japanese from Cambodians...now I can, and all of those groups can tell each other from one another, for example.
If a white woman who has been raped by a black man views a lineup of police suspects picked up based on her description, but the real perp ISN'T in the lineup, there, tragically, is an extremely high chance she'll pick out the guy who looks most like the perp, and she'll very confidently point him out in a trial and say "He raped me."
I gave the reference to Loftus because she is probably the most well-known exponent of the eyewitnesses-are-unreliable viewpoint, so I thought it was relevant. I don't agree with an extreme position on this, although I try to approach all evidence with a healthy skepticism.
Wouldn't dream of it, no way.
As the article states, "Before they come to trial, they've retold that story quite a few times, embedding it in their memory,"
Huh uh. Sounds just like Knowlton. No telling how many hours he spent with Ruddy and Prichard telling his story.
Most trials are not conducted in the courtroom, but on the front steps in front of the media. Beyond that, I doubt what most people say.
I agree. And Newsmax is run by a consipiracy oriented guy. Chris Ruddy. That's why I don't have much trust in that outfit.
Well that is true. And it is the defense lawyers who do it because the prosecution cannot try the case in public. So defense lawyers know how to play the media. I think research in the field of eyewitness testimony is showing more than ever how frought with difficulty our trail system is.
I read a novel called The Truth Machine. The premise was that a genius Bill Gates type invented a machine that was 100% accurate in truth detection. The affect on society was astounding and far reaching. It changed everything. Good book. I recommend it.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.