Reflection

All rights reserved. This text may not be published online or distributed without written permission. ​Transcripts are generated using a combination of speech recognition software and human transcribers, and may contain errors. Please check the corresponding audio before quoting in print.

Phoebe Judge:​ What about those that have been training to lie? You know, that this is… They know how to do this. They know how to lie in an interrogation. Is that possible? Are we able to train people to lie? Andy Morgan:​ Oh, people lie all the time. I don’t know if you have to be trained. I think people learn as they grow up, there’s all kinds of lies people can tell, right? You can tell lies about the past, the future, yourself, other people. You can lie for fun, you can lie to not be punished. You can lie because it’s your job. If you’re a professional working undercover, you have to lie. You can tell lies about what you did or what you will do. Those are all different kinds of lies. So, in the context of when people are being questioned by the police, there’s a number of ways in which a person could look, if they are deceptive, could look absolutely normal. Whereas a truthful person might actually be frightened of the police. And they, ironically, would look deceptive. Phoebe Judge:​ There’s this thought that if you want to know if someone is lying, you should be able to tell by just looking at them, that people give off signals of dishonesty. This is huge in TV shows like ​Sherlock​ and ​Lie to Me​.

1

Cal Lightman:​ Classic one sided shrug, translation, I have absolutely no confidence in what I just said. The body contradicts the words, he’s lying. Andy Morgan:​ It feels so compelling that you should be able to recognize you know, fear, disgust, anger, sadness, and that these are universal expressions. And I think that, it’s true that around the planet by and large, many emotions are expressed similarly, by the facial muscles, not entirely but similarly. What’s really, I think, more questionable scientifically is whether or not these expressions or micro facial expressions, these quick flashes of human emotions on your face are actually signals of lying. Phoebe Judge:​ So if your body doesn’t matter the way we thought it has for so long, then what does? I’m Phoebe Judge, and this is Criminal. For 20 years, Andy Morgan has been studying human memory and deception. He’s worked with the FBI, CIA, the military, law enforcement groups. And it goes without saying that all those people Morgan worked with, they really like to know when someone’s lying to them. They will do some crazy, expensive, exhaustive things to find out if someone’s lying to them. But what we’re learning now is that a lot of those things they’ve been trying, they just don’t work. We’ve been looking for physiological clues since the 1920s, with the invention of the lie detector. Speaker 1:​ This is the read polygraph, a lie detector. In the past 20 years an estimated 200,000 persons have staked their futures, many of them their lives, on this machine. Andy Morgan:​ The polygraph, as you know, is a device that records how fast you’re breathing, sort of your respiration rate and how much your chest is moving and they put tubes around your chest. And the general belief in law enforcement and within the polygraph community for years has been rooted in the idea that we call the fear and alarm hypothesis. Speaker 1:​ That variations on blood pressure and pulse rate are present, during and after the act of lying. Andy Morgan:​ That telling a lie is threatening in some way. And that threat to you will trigger a difference, a shift. Speaker 1: ​The conscious act of lying creates an emotional disturbance. Andy Morgan:​ And cause your blood pressure to rise or your skin conductance to go up or your respiration rate to change, like holding your breath or maybe breathing more rapidly. Phoebe Judge:​ Now it may not shock you that the polygraph isn’t a perfect tool. But we were actually surprised to learn just how inaccurate the machine is. Andy Morgan: ​We actually know the polygraph in the way it is mainly used is not better than chance at detecting deception, maybe slightly at 52%, 53%. Phoebe Judge:​ You could literally flip a coin and be just as accurate as the polygraph at detecting a lie. But the idea that your body gives you away — it’s still very much alive. It’s still being used in police departments all over the country. Cops are trained to read body language. Some investigators claim they can watch an interrogation with the sound off and know if someone’s lying.

2

But what Andy Morgan says is that when put to the test, people aren’t any better at sensing a lie than chance. And if that’s true, that we can’t see a lie, that we can’t hear a lie or smell a lie, or physically sense in any way when someone’s playing us, then what do we do? About a decade ago, Morgan and some colleagues decided to test a different technique. They started interviewing people from all over the world, about 1,200 of them. Andy Morgan:​ So ranging from white Brits to Chinese, East Indians, Lebanese, Jordanians, Moroccans, Sudanese, Afghans, Iraqis, Vietnamese, Russians. My colleagues and I have tried to study people inside and outside the United States. And we’ve mainly used a particular technique called cognitive interviewing. Phoebe Judge:​ Cognitive interviewing is based on the premise that memory doesn’t work like a video camera. There are sights and sounds and smells stored deep in your brain that can all be recalled if pushed hard enough. So, every time Morgan sat down to interview an American or Jordanian, some of them would be lying, and others will be telling the truth. And he would ask them to do something pretty simple. Andy Morgan:​ To tell us a story about what they’ve been doing. Speaker 2:​ The most memorable concert that I have been to, it’s not the best, it’s the most memorable… Phoebe Judge:​ So obviously, this isn’t actually tape for Morgan’s research. We just put some of our friends to the test. Speaker 2:​ …was Yeasayer, about a year ago. Speaker 3:​ I went to a Foo Fighters concert in 2005 for the In Your Honor tour. Andy Morgan: ​And then say, so if you imagine for a minute that I was there with you, what would I have seen if I’d been there with you the entire time. Speaker 2:​ They had the lights programmed with the music and so the mirrors really, like, reflected the light, and so there was, like — it would form shapes. Andy Morgan:​ Then we say when they’re done. So I think I’m getting a better picture in my head. So imagine for a moment I was with you, but this time I was blind, and I could only listen. What would I’ve heard during that time? Speaker 3:​ So it’s more of a shuffling of like rubber against pavement there. Go down metal steps, go down to the car. Our car door’s open. Andy Morgan:​ You know, depending on the context, you might say, “What would I have smelled?” Speaker 2:​ There was this this girl that was standing next to me who was in a sweater dress, who did not wear deodorant. Andy Morgan: ​And then we say, starting with the very last thing that happened, what you remember happening right before that?

3

Speaker 3:​ Leave dorm. Andy Morgan:​ Right before that? Speaker 3:​ Go walk backwards to car. Andy Morgan:​ Right before that? Speaker 3:​ Get in car, come back to the dorm. Andy Morgan:​ Kind of having them almost walk us backwards through their memory. Speaker 3:​ Crunching guitars and awesome drum stuff. Andy Morgan:​ What’s been found over the years in many studies of cognitive interviewing is that using those mnemonic prompts, those sensory prompts—what you would have seen, heard, smelled, thought, touched or tasted—they trigger more memory recall, you get more detail. The picture becomes much more rich and complex without suggesting anything specific to the person you’re interviewing. Phoebe Judge:​ But here’s where things get interesting. Turns out, if you’re telling a lie, a made up story, even one that’s well rehearsed, you can’t complete this interview, not without giving yourself away. Andy Morgan:​ If you tell a very simple lie, there’s not much work to, you go, I don’t know. And you have nothing to say. Kids do that all the time. But if you’re telling… If you’re telling a story that’s reasonably complicated and supposed to be believable, because if I’m lying to you, my goal is to sell you the story I’m telling you, and then to leave before you figure out it’s not true. Or not to bring up anything that might lead you to suspect that it’s not true. I’m trying to tell my story and stick to it. We find that liars are often worried that if they’re inconsistent, they will be thought to be lying. So what they do is they tell you a story. So if I say, “Tell me everything that happened.” They might tell me quite a wonderful story. Speaker 4: ​I went to see Florence and in the Machine, it was like in the spring, Andy Morgan: But then when I say, so let’s go back to the beginning. So, imagine I was with you, what would I have seen, you know, heard, thought, smelled, touched, tasted, or walk through it backwards. The overall result is they have very little to say. Speaker 5:​ Do you remember the car ride home? Speaker 4:​ Uh… Not really. Andy Morgan:​ They will say things like, well, pretty much just like I told you before, and they will repeat very closely the same thing they’ve already told me, so that I’m not learning anything new over time. Speaker 4: ​The thing is, I don’t even really remember. Andy Morgan:​ It’s sort of like comparing a digital photograph of your house with the tree in the front yard. And a child’s picture of it where there’s a house and there’s a tree, and there’s clouds

4

and there’s birds, but there’s not a lot of detail. And what we tend to do when people tell us lies, is I think we fill in the blanks. Phoebe Judge: ​And that Morgan says is where the problem lies. We fill in the blanks, because of course we want to believe people. So Andy Morgan’s challenge was to find a way to take these interviews and analyze them without filling in the blanks. To do this, he had to put a little distance between himself and the interesting person telling the story. Andy Morgan: ​The way we’ve analyzed that, in most of our studies, is we record the interviews and we do a transcript. And we let the computer just count the number of words that comes out of a person’s mouth in the interview, and the number of unique words that comes out of their mouth. That — so, if you think about the phrase, one small step for man. Neil Armstrong: ​It’s one small step for man. Andy Morgan: ​One giant leap for mankind. Neil Armstrong:​ One giant leap for mankind. Andy Morgan:​ There are 10 words, but you’ve used “one” and “for” twice. So there’s only eight unique words. As you’re thinking harder about monitoring what you’re saying, it has the side effect of reducing the richness of what you have to say and shortens the thing that you have to say. All the computer is doing is counting those two variables. And when we sort people based on those two variables, response length and unique word count, the computer’s right, typically 80% to 85% of the time. Whereas our professional raters, whether they come from Homeland Security or the FBI or the DIA or other intelligence groups, or law enforcement, or human raters are rarely better than 54% correct. Phoebe Judge:​ It’s kind of a crazy idea that someone with a transcript and a tally sheet could tell us whether or not what we’re remembering is accurate, while someone with decades of FBI interrogation training could be flying blind. And this idea could have all sorts of huge implications. Think about a jury. A jury is just a bunch of people trying to figure out who to believe. But people, according to Andy Morgan, are terrible at figuring out who to believe. Isn’t it possible that a juror with a transcript of the trial, who’s never laid eyes on the accused, might actually do a better job at figuring out who’s lying, who to trust? For Andy Morgan’s part, he’s glad our bodies aren’t the end-all, be-all. Andy Morgan:​ It’s reassuring to know that, although there’s so many books on nonverbal behavior and people say 90% of communication is nonverbal, based on the science, I think that the best way to sort out the truth is to listen to what people have to say. Phoebe Judge:​ And to know that maybe our fate won’t hang on sweaty palms or a pounding heart, that our words really do matter. Thanks to Andy Morgan, who’s a forensic psychologist at Yale. And just in case you were curious, during his interview with us, he used more than 7,200 words and enough unique words to make us trust he wasn’t lying.

5

The show is produced by Lauren Spohrer, Eric Mennel, and me. I’m Phoebe Judge, and this is Criminal.