Elizabeth Loftus

Q: Why is eyewitness testimony so compelling?

LOFTUS: I would say that eyewitness testimony is amongst the most powerful testimony that can be presented in a courtroom. When a victim gets on the stand and says, "I was there, I saw it, that's the guy, I'll never forget the face, I'm absolutely positive," it's extremely compelling testimony and it's very hard to shake that testimony. Well first of all you have a real victim, you don't want to disbelieve that the crime happened, you don't want to look as if you are undermining this person's experience. And so that's part of the problem of cross examining a victim of a horrible crime. The testimony, especially if it's given with a expression of confidence and if it's given with a lot of details, jurors are going to believe it and that's why innocent people often get--or at least sometimes get convicted.

Q: What about the pressure on the victim to stick with his or her identification?

LOFTUS: By the time a victim does get into court to testify, of course, she's told her story many, many times. She's told it to the police right after the crime, she's talked about it with friends, she's talked about it while she's looking at photographs, she's talked about it when she goes to a lineup. She's talked about it in preparation with the prosecutor before her testimony and now she's telling the story again. So at that point there's been an awful lot of rehearsal and an awful lot of kind of smoothing over of the story, making it often even more coherent, even less inconsistent, even more persuasive.

Q: What can an expert witness, like yourself, do in a trial centered on eyewitness testimony?

LOFTUS: One of the things that expert testimony can do is to educate the jury to recognize the fact that jurors have misconceptions about the workings of memory, about the way eyewitness testimony functions. They believe things that are unsupported by the scientific studies or, at least, sometimes contradicted by the scientific studies and expert testimony can give them accurate information that they can use to decide whether this identification is likely to be an accurate one or possibly a mistaken one.

So just to give you a specific example, many people don't appreciate that we have a cross-racial identification problem. Many people they've heard the saying, "They all look alike," but they think that's just a myth or they think that's only true of prejudiced people. In fact, there are many studies that show that we do have a cross-racial identification problem and it's not just the people who are prejudiced who have it.

Q: Does this kind of expert testimony on eyewitness identification happen often in trials?

LOFTUS: Well, in any particular case a defendant may want to introduce expert testimony about the factors that influence eyewitness accounts about how memory works and it's up to the judge whether to admit that testimony. Some judges let it in and some judges don't.

And when they say, "No, I won't let the jury hear it," they often give a number of reasons. They'll say it's all a matter of common sense or it's going to unnecessarily prolong the trial or it's going to be prejudicial to the state or they might say that it's up to the jury to decide it interferes with the jury's function. They'll give a number of reasons and state those reasons as why they don't want to let the testimony in.

Q: What happens when this expert testimony is not admitted into court?

LOFTUS: When a judge refuses to admit the expert testimony, whatever the reason that might be given, stated on the record, one consequence is that the jurors sit there and they decide based on their gut feelings. They decide that's a confident positive witness, "I'm going to believe her," or they decide based on whatever misconceptions they might have, they think to themselves, "If I were being raped, it would be so frightening I would never forget that face." Not understanding that extreme fright and stress can lead to problems in memory. So what you may have then is a situation where jurors are making decisions based on incomplete or even erroneous information.

Q: What do you say in a situation where a victim might stick with his or her original identification in the face of proof it's someone else?

LOFTUS: Well, first of all, look what happens in a typical situation. You have a crime, the victim and the perpetrator are together for a very short period of time, maybe it's only a matter of minutes, maybe she's really only looked at the face for a matter of seconds, but later on, if she makes a wrong identification, she's looked at the photograph, she's looked at the person in a lineup, she's looked at the person for hours sitting in court, she has that image as the person who committed the crime and so it's not that surprising that when faced with the real person somewhere down the line, when that real person has been finally discovered that she might not recognize him anymore. After all, she's seen the falsely accused person far more often and for far longer.

Q: So should we prosecute somebody based only on eyewitness testimony?

LOFTUS: Sometimes you hear people say that we should never prosecute somebody based only on eyewitness testimony and nothing else, but of course that's going to lead to some problems, because in some situations like rape, for example, often that's all you have. And some identifications in rape situations are very reliable. If people are together for a long period of time, if there's an identification the next day, if you have a same-race situation, if you have an absence of problematic factors, then you have an identification that you can rely on and is perhaps quite reliable.

So you don't want to throw out all eyewitness testimony or all uncorroborated eyewitness testimony, you just want to be careful that you don't do something to turn this situation where you've gone one definite victim into a situation where you now have at least two or more victims, namely, the first victim and now the falsely accused victim.

Q; What is unconscious transference?

LOFTUS: Unconscious transference is a general term that refers to the mistaken identification of a person, who was seen in one situation, with a person who was seen in a different situation. You confuse people--you think you saw this person here, when you actually saw him over here. In a case where a victim goes, looks at some photographs, later on goes to a lineup, sees some individual standing there, sometimes people will recognize somebody in the lineup, make an identification, but they're making that identification, he looks familiar because his photograph was seen earlier, not because he was the person who committed the crime. That is a kind of an unconscious transference.

Q: What about DNA evidence and its impact?

LOFTUS: Well, one of the things that the DNA teaches us is that, for sure, there are mistakes that have been made. We can do all the laboratory studies of eyewitness testimony we want and we still have some critics out there who say, "OK those are just studies, those are just college students, those aren't real crime victims, real crime victims wouldn't make those kinds of mistakes." But the DNA analysis that exonerates these people, and ultimately proves their innocence after they've been convicted based on faulty eyewitness testimony, shows us clearly that real victims do make mistakes and once we acknowledge that and once we bring these cases together and try to analyze them and say what went wrong, hopefully we can design a system, fix our system, modify our system so that we minimize the chances of this happening again.

Q: What should police understand about eyewitness identification, memory, etc.?

LOFTUS: I think police need to be aware that the things they say and the things they do can change the witness, can make that witness identify the wrong person, can make that witness more and more confident about an erroneous identification. So they have to try to not ask leading questions, to not communicate and convey information to a witness that can contaminate the witness's memory, the accuracy of that memory or the confidence of that memory.

Q: And what happens when a victim is confronted with the fact he or she made a false identification?

LOFTUS: I've seen all kinds of situations where victims are confronted with the real guy or confronted with the fact that they appear to have made a mistake and some of them will accept the fact that they've made a mistake and will back off of their identification. But others insist that it hasn't been a mistake. There's something wrong with your new evidence, someone changed the DNA, there's still a way to maintain your memory if, in fact, you want to.

Q: How do prosecutors view the value of eyewitness identification?

LOFTUS: Prosecutors, as a general rule, tend to trust eyewitness testimony. Perhaps they over-believe in the value of eyewitness testimony and they play on the jury's misconceptions about eyewitness testimony so they'll, first of all, through rehearsing their witnesses and getting the witness's story down, and getting rid of some of the inconsistencies in the story, they'll create a witness who is more coherent, more consistent, more confident, more credible, more impressive to the jury. And then in their argument, some of them will push some of these misconceptions. They'll say things like she was so frightened, she'll never forget that face as long as she lives. And in those arguments, that prosecutor can help insure that the jury believes that testimony and convicts the defendant.

Q: Do they put a certain kind of value on eyewitness testimony that they don't on some physical or circumstantial evidence ...

LOFTUS: Well, prosecutors have been around for a long time, they have learned through their experience that there is almost nothing more valuable than an absolutely positive confident eyewitness and if that's what they have and they see that they have a witness who is credible, who is sympathetic, who really was a crime victim, and who is going to be making an identification in a confident way, that's an easy case for them.

Q: But we can't dismiss eyewitness testimony, can we?

LOFTUS: Often times, eyewitness testimony is accurate and in many cases has proven to be accurate, because you have circumstantial evidence. You know the victim's belongings are found in the defendant's home. The defendant's fingerprints are found in the victim's bedroom. You have corroborating evidence that helps to confirm that the eyewitness testimony is accurate. So it's no wonder that we see many situations where eyewitness testimony is accurate, we see it's accurate, we know it's accurate, it's just that fraction of the time where it's uncorroborated, unsubstantiated, that's all we have and it could be mistaken especially if things weren't done right.

Q: Is DNA opening up our eyes?

LOFTUS: Well we have throughout history collections of case of wrongful conviction. We, of course, know that we have real cases of wrongful conviction and we know it because a perpetrator comes forward who ultimately confesses and he knows details only the real criminal would know. So we know that one was a mistake. We have had collections before of wrongful convictions. But the DNA tells us pretty much, for sure, we have a whole new class of cases where wrongful convictions have occurred. We have rape cases because often times when you have these DNA analyses later on, it's because there has been a rape. So even when they're this close, the victim and a perpetrator, you can still get a mistaken identification.

the photos | cotton's wrongful conviction | interviews | faqs | re-evaluating procedures | song of an innocent man | links | tapes & transcripts | reactions | explore frontline | wgbh

web site copyright 1995-2014 WGBH educational foundation