Deception Detection in Legal Systems

Hello everybody and welcome to today's thrilling, riveting, exciting lecture on detecting deception within the legal system. So just to start off this lecture, what I'd like to do is just to get your opinion. What do you guys think of this statement that uncovering the truth is the central goal of the legal system? True? False? An ideal bill. So it's easier said than done. And when we're talking about the legal system, as you guys could imagine, there are probably situations where certain people, cough-cough criminals, would probably want the actual truth. be hidden. So how do we go about testing whether somebody is lying or telling the truth? And this is going to be a critical process in many different stages of the fantasy world. investigation, and trial proceedings. So I like to start this topic with lie detection in the olden days, lie detection across history, and even the earliest legal systems had techniques that they would use to try to discriminate between truthful and deceptive statements. Although, we can ask, you know, how exactly reliable are the legal systems? these processes were and that might be up for debate so for example there were what were called trials by a trial by ordeals which might sound like something uh in a show like game of thrones or something but back in the olden day people had a wide variety of ways that they could try to prove their innocence so they can have they could undergo something called ordeal by snake where a suspect would put their hand into a clay pot, and the clay pot would have a ring inside of it and a snake on top of the ring. So the suspect would then have to retrieve the ring from beneath the snake, and if the accused was able to do so and not get bitten, then they were innocent. So basically the snakes are lie detector experts here. There's also ordeal by fire, where this is mostly used for women accused of adultery. But these individuals would have to walk across hot metal or hold hot metal, and they were considered to be innocent if, by the end of this, they weren't injured. Which is, you know, hard to tell. What's the odds of not getting injured while doing that? Troublesome. So, if those don't sound appealing to you, if you are a murderer and you want to undergo a different trial, you can undergo what's known as... as an ordeal by blood, where suspects would touch the corpse of a victim, and the victim was expected to bleed fresh blood when the true murderer touched them. So, I guess it's... At the very least, you could say the people back in the olden days were creative, I guess, right? Now we can be incredibly grateful that we don't have to undergo these horrible processes. But a good question is, you know, do we have anything better? And when I say better, I don't mean like, you know, we're not going to get bitten by snakes or anything, hopefully. But are we any more accurate than these, what we would consider to be, largely ridiculous approaches to deception detection? Well, let's find out. So when we are talking about lies, it's important for us to discuss why do we even bother lying? What would you guys think? So when we're talking about society, especially ours, we really seem to value honest or quote-unquote being or keeping it real, right? Wouldn't life just be a lot easier if everybody... told the truth? And the answer is, you know, maybe it might be, but there's also social advantages to lying sometimes. So let's say you have a friend, and normally this friend is very shy. and introverted, but for her new year's resolution, her new year, new me, she says she's going to try coming out of her shell a bit. So she finally agrees to come to a party that you're hosting, but she shows up in this monstrosity of a dress and asks if she looks okay. What are you going- to say you can think about it yeah you know are you gonna say that's the ugliest thing to be burned into your eyes as of recent that she needs to change some people may say yes hashtag keeping it real uh but Most people might not. They might say something like, oh you look fine, or maybe some sort of middle ground. That's an interesting choice, but maybe you look a little extra. Why not borrow this instead or something like that. So rather than crushing your friend's dreams as soon as she takes the leap in coming out of isolation, maybe they try to smooth it over. So one reason that we lie is going to be for that, you know, sometimes it just keeps things working smoothly. Social lubrication, if you will. And sometimes a little lie is going to go a long way in terms of maintaining social bonds. If you think of people... and I know like there's this stereotype with it where it's like when you're just like oh I'm just keeping it real uh usually people kind of use that to defend you know maybe saying really hurtful things and do those people have a lot of people who enjoy hanging out with them and it depends right so in terms of the lies that we may tell socially you know think of some that you might tell you might feign interest in say like if you have a significant other who has like a project or a hobby that they personally love but you think is more boring than like watching somebody cut their toenails, you may pretend to be interested or at least humor them. You probably aren't gonna say just shut up I don't care about that. We also might lie to appear more impressive or to avoid scrutiny. Like for example you may say oh I promise I did the homework it's just at home unfortunately even if you didn't actually do it. Additionally, if you think about it, we perform these white lies quite often as we go out through our day, and often it's the case that it's such a trivial part of our day that we might not remember it. So uh... In addition to these more white lies, people may also lie to, say, hide their wrongdoings. So being able to catch liars is beneficial to society. We can consider this as well from an evolutionary standpoint, where our ability to lie is adaptive and beneficial to us. So from an evolutionary standpoint, one of the special skills that humans have... have is our ability to work together, hunt, and cooperate in groups. And that is an advantage over a lot of other species. And then, in addition to that, our ability to, say, bend, spin, or shade the truth is something that also helps with keeping these groups functioning smoothly. So if you were to tell your significant other, if you talk to me about that boring project one more time, I'm going to scream. That's probably not going to go over so well and you'll have disharmony. in your relationship. So there's a hypothesis known as the Machiavellian Intelligence Hypothesis that a great leap forward in human intelligence was triggered by our need to develop the essential social skills for deception. So basically our development of the ability to lie was a big jump for humankind. awesome natural ability to be able to lie, it can be kind of troublesome when you're trying to decide if you're being lied to or not. Not only is the act of lying ubiquitous or ever-present throughout society, there are many different ways that we can lie. So check out all of these cool words that I found that we can use just to describe the same thing. So we have lie. Lie. We have mislead, we have misinform, trick, fake, betray, fool, cheat, confabulate, delude, distort, fabricate, exaggerate, pretend. misrepresent, and that's just a few of them. There's loads more that we can use, but what's particularly interesting is that each of those words all differ ever so slightly. They have a slight distinction from each other. even though the underlying action is, quote, presenting an untruth. There are many different ways that we can do that. So in addition to these lies, we also have a distinction between what are known as lies of commission and lies of omission. Lies of commission are what people really think of, or what they're more likely to think of, when they think of lying, which is telling somebody something that isn't true. I did my homework even. even though I didn't. Stuff like that. Alternatively, you have lies of omission, and this is going to be when you leave out helpful details, pertinent details that might reveal the truth. So, for example, a cheating spouse may say that they had dinner at a restaurant, but just leave out the not-so-important detail that it was with their affair partner. Who would want to know that, right? Yeah, it's an important detail. Additionally, you can have both well-practiced or well-rehearsed lies, and you can compare those to spontaneous lies or lies on the fly. lies that hurt, you can also have lies that hurt other people, lies that spare people's feelings, so the intent of lies can differ, you can have lies that we feel justified in giving, or we don't feel any type of way. way compared to lies that we feel guilty about. You can have white lies which are, you know, no real consequence. They're not really hurting anybody compared to, say, high stakes lie where the consequences are real and severe. And all of this goes into adding much more difficulty into, say, how do we approach trying to figure out what statement is true or false? If we have so many different types of lies, what on earth do we do? do we do? So give me your opinion on this quote. So this is a quote from one of the researchers in this area. Lying is just so ordinary. It's so much a part of our everyday lives that we hardly notice it. In many cases, it'd be much more difficult, challenging, and stressful for people to tell the truth than to lie. Take a moment and answer, do you actually agree with that statement? Is there a difference, say, between a harmless lie and a harmful lie? This is usually a fun class discussion, but, uh, you know, sometimes people will have the idea, like, no, you should never lie at any cost. But, you know, are there situations where maybe it's easier? Better? Some people may still say no. But, you know, other people would completely disagree. Maybe you don't want to tell your friend that that's the ugliest dress, or maybe you don't want to tell your significant other you're not as attractive as they were however many years ago. It's a troublesome dilemma. So now to the meat and the bones. Can we tell when people are lying? We might like to think that we are, but we aren't. In fact, a meta-analysis using 24,000 participants, more or less, they found that we're only accurate around 54% of the time. of the time. And, if you do the math, you know, if we have two decisions, we can either judge a statement as true or a lie. We essentially have a 50-50 on being right. You might as well just flip a coin. like the 54 does that really functionally differ from a 50 percent so one factor that contributes to uh us kind of sucking as lie detectors is what's known as the truth bias what the truth bias is is that you know on average we're just more likely to interpret statements as being truthful rather than interpreting or judging them as deceptive you In fact, if we stick to this, the meta-analysis with the 24,000 people, what the meta-analysis showed is that 61% of truthful statements were in fact judged as true, but only 47% of lies were actually judged as lies. And so you might take that as, wow, well, we're really good at deciding truthful statements, but what the two of those pieces of information mean together is that basically we just kind of say... True, true, true, true, true, true, true. As opposed to actually being any better. So, uh, we're, we just have a slight tendency. This can kind of come from the expectation, well, we don't typically go out into the world expecting people to lie to us. Now, of course, this has exceptions to that rule, but normally we operate under the idea that people... being truthful with us. You may now ask, well, what about people who are trained or who we would hope would be better at spotting lies than the average bear? And again, less than optimistically, Again, no, we're still really bad. So, in a study that had police officers, somebody who we might hope to be better than, say, a college student, researchers had police officers and college students watch true and false confessions from criminals. So, criminals either talked about a crime that they did commit, or they described a crime that they didn't commit. And they had police and college students try to... to assess the veracity of those statements. So college students were slightly better than Chance. What about the police officers? Well, they weren't better than Chance at all, but the police were more confident that they were correct, which is troublesome. You do not want a confident and inaccurate individual. Additionally, having more training seems to kind of bring about what's known as the lie bias. where these statements are, these individuals are now more likely they're biased to say statements are deceptive rather than being truthful. So this kind of goes against the truth bias where they completely switch and they're now just more skeptical of all of these statements. One problem in terms of training is that it might be the training itself that's troublesome. So if I were to ask you, how would you decide if someone was biased? somebody was lying or telling the truth, what would you say? What would you look for? You can take a moment and write some down and think about it. You can pause the video. Okay, so if we would have done this in class, I bet I would have placed money. But some hot answers, some answers we would have gotten would have included things like if they're stuttering, if they're making eye contact, or if their eyes are looking around, if they seem nervous, if they're fidgeting. And, uh, you guys probably played right into this mistaken liar stereotype, which is, you know, we think that we know what a liar looks like, however... as noted by the mistaken portion of that name, it's not actually helpful. So we have a tendency to look for unhelpful behavioral cues when we're trying to assess veracity. But the truth is, there is no... obvious behavior that fully gives away when somebody's lying, so there's no fancy Pinocchio's nose or anything like that. Most behaviors actually have no relationship to deceptive behavior, and a few actually do. best have a weak correlation. Is a weak correlation enough for somebody to go to jail over? Me personally, I would say no, but this is also a problem in training because interrogators are often going to be trained to look. look for the specific verbal and nonverbal behaviors when they're trying to figure out if somebody's lying. So training hasn't been shown to really improve the ability to detect lies, but it just makes interrogators more confident in their abilities to do so, which they shouldn't be. And in fact, like a recent study actually showed that the police officers they surveyed, they said that they're able to detect... around 77% of lies, when, as we just talked about, in reality, their rate is really close to that 50%. This overconfidence is also troublesome, because once an interrogator believes a person's guilty, they can then fall prey to confirmation bias, where they're going to look, they're going to interpret the suspect's behavior as if they're a criminal, meaning that they're going to look. for behaviors that confirm the, oh, this person is guilty notion. And then they're also going to ignore behaviors that, you know, evidence would suggest that this person is not guilty. Additionally, looking for these unreliable behavioral cues has been shown by research to unfortunately occur. Where interrogators perceive innocent suspects as guilty because they didn't act, and I'm quoting, the right way during interrogations. So if people don't act how I expect them to. they must be guilty. Another problem is that if the police believe a suspect is the culprit, even if they're wrong, they're going to be more likely to subject them to harsh interrogation techniques. And when we have these unhelpful cues to deception being used, and then this is going to be paired with that confirmation bias, and then we double down with some high stakes, very threatening interrogations, this can and does lead to false confessions. But we'll get into that on the next video. So, take home point of this slide, we are not really good at this task. Alright, so I just got very doom and gloom on you guys with that deception research, and what about humans as lie detectors? You guys might be feeling a little down, and you might be saying, okay, well, is there anything that we can use? we can't trust ourselves or the police to tell the difference between truths and lies. You may say, what about all those snazzy polygraphs that I hear all about? I mean, wouldn't it just be great and delightful if we could hook suspects up to a machine, have the machine, do the dirty deception detection work. I mean, in theory, if you do that, you could just ask, like, did you kill the person? And I mean, at that point, you wouldn't even need trials, right? What a wonderful world and machine that would be. And as you might have imagined, that magical device that detects those lies perfectly doesn't exist. But we can talk about the polygraph anyway. So the polygraph usually, in terms of like the process, it usually has people hooked up, they have a blood pressure cup around their arm, usually a pneumatic tube around their chest, and then electrodes on the fingers of one hand. The underlying theory behind how the polygraph works is pretty straightforward. So the idea is that lying, the act of lying, causes physiological arousal. Things like increasing like increasing heart beating, sweaty palms, they're nervous, stuff like that. So as the questioner asks their questions, those deceptive statements should be paired with an increase in some sort of physiological arousal. The history of the polygraph is actually really interesting. It was created by a man called William Marston. He was a lawyer, and feel free to insert all your lawyer jokes here. And he was also a psychologist, feel free to insert the jokes there too, who believed that these physiological changes could tell what was happening. when somebody's lying. Also, in his spare time, he created the comic book character Wonder Woman, which is extra fun because, you know, she has that lasso of truth where she wraps people up in it, and once they're tied up, they have to tell the truth. And, you know, maybe he wanted one of his own, except his is a little not as flashy as the lasso. So, you know, If a gavelin does, that would be great, but Marston would probably say that he created one. So just to give a little bit of background, during the 60s, 70s, and most of the 80s, the polygraph was a really, really booming and lucrative business. So there were independent contractors who offered employers quick and easy ways to decide who to hire and who to fire. So people applying for jobs were often asked to submit. to polygraph testing where they would be asked questions like, have you ever used illegal drugs? Have you ever lied to former bosses? Did you tell the truth on all of your job application questions? Yikes. Did you steal money? Did you steal money from your previous employers? You know, all of the juicy stuff. Big old can of yikes, right? And if these applicants... were caught lying, they didn't get the job. But just because you got the job doesn't mean that you are free from the polygraph because employers would then also test people once they were hired as well. And because this was such a booming business, there were a lot of schools opened up at the time to train people in the art of polygraphing. This, however, came to a screeching halt with the Polygraph Protection Act in 1988. And what this act did was it did was it prohibits most private. Employers from using polygraphs to make like job related decisions. The keyword is Private so public employers are exempt so places like the FBI CIA NSA CEA, and Secret Service, you know, no one big, no one famous, they all still rely on polygraph testing. Also, commercial businesses that are involved with supplying like security personnel to facilities. that work with public safety are also exempt. And who does this look like? These are, just think, like nuclear power plants or public water supply facilities. So, gone the way of, you know, it's fallen a little out of favor since its big boom, but you could make a really nice salary from simply trying to figure out if somebody's lying about, say, having stolen too many staplers from a previous job. So... The last thing that we mention are going to be what's known as countermeasures and or self-stimulation. So it's not as hard as we would like it to be to trick the machine. And there are a couple of different ways that you can do this. And this is going to be by altering your level of arousal. So what you can do is you can either make the you try to make the comparison between truths and lies more blurred. Right. So this can either make truthful responses, you can try to pair those with more elevated arousal, or you can try to suppress that arousal through lies. So some techniques people will do is they may bite their tongues, which hurts, and that raises arousal. They may press their toes into the floor. And these countermeasures are actually quite effective, reducing detection by around 50%. And what's particularly troublesome is that examiners won't know. know when these countermeasures are being done. So in terms of the polygraph, you know, maybe there are other things. Maybe there are other things that are more beneficial, but a reasonable question is, well, what do the courts think about this evidence? And if you watch shows like Law and Order SVU, they're like, oh no, this is inadmissible. And we like to think think that polygraphs are inadmissible in court, but the truth is actually a little bit more nuanced than that. So 23 states have the polygraph banned. Others will allow it under certain circumstances, and then, like, New Mexico routinely allows polygraphs. Polygraphs fly in New Mexico. So, uh, it's not extinct, and it actually does occur in courts more often than you might think. So when jurors hear about the polygraph reports, when they come out in the trial, what do jurors think about this sort of evidence? And for the most part, jurors find polygraphs... and it results incredibly compelling or persuasive. So for example, here's the case of a man known as Buzz Fay. Fay was sent to prison for aggravated murder and was convicted for shooting a man called Fred Erie. during a robbery of a liquor store. After being shot, the victim was rushed to the hospital. The victim was given pain medication, and then he was asked who shot him. So while the victim was suffering from blood loss and dying, they said, I think it looked like Buzz, but it couldn't have been. And then, later on, right before he died, he said, Buzz did it. The case was very, very weak because there was no physical evidence linking Faye to the crime. And the police knew that their case was really weak. So what they did was they made an offer to Faye's attorney that they would drop all of the charges if Faye was able to pass a lie detector test. And Faye was great. Terrific. He knew he was innocent. And this seemed like a great way to get this over and done with. Right. He's just going to take the test. He's not guilty. guilty, why should he worry if he's innocent, right? Wrong. Or at least according to the polygraph. So Fay ended up failing the test, and then he failed the test again, and then he was convicted by a jury who sent him to prison based largely on those lie detector results. Additionally, this isn't the only instance of polygraphs. Pretty much changing the court outcome either, where that might be the only piece of evidence or the best piece of evidence. And research has shown that, you know, it's backed up this notion that jurors find it compelling because... say one study they asked law students to provide they read a case and then they provided is this person guilty or not guilty 85% of law students who originally said that the defendant wasn't guilty changed their vote to guilty when they heard the defendant had failed their lie detector test and these are law students so so So although the polygraphs do not have a specific legal precedent, Justice Clarence Thomas expressed two concerns of the polygraph for the legal system. One, there's this lack of consensus for scientific validity. And then second, it reduces the role of the jury. So, you know, is science really sold on the lie detector as a use of on the polygraph for detecting lies? No. Not really. And then the view that, well, the jury should be the lie detector, which may still also have some caveats with that as well, too. Additionally, how effective the polygraph might be is going to depend on a couple of things that are going to vary case from case as well. So, for example, how persuasive the polygraph or testimony is. Remember we talked about kind of the... A polygrapher might be more or less intimidating depending on, say, how they look. Similar here, so some individuals just come off as more compelling, believable, persuasive, and we really want jurors' opinions hinging on something as unrelated as that. Sophistication of jurors, so whether they are, say, up to snuff on how reliable polygraphs actually are, and then whether or not judges include, say, instructions that might warn. the jurors about, well, you might want to consider this. Those are all things that are going to change and impact how likely those polygraphs are to be persuasive to jurors. In terms of what the experts think, experts are a little bit more skeptical than the average person in terms of, you know, their thoughts on whether this is an effective tool or not. So if you look at the experts, most of them... are more likely to say that polygraphs are questionable, some might say it's useful, and then non-experts are going to say that it's useful, but you never have an instance of experts saying that they are reliable. So, you know, science is meh on it, but people seem to really find it compelling. A different approach is known as the guilty knowledge test, or the concealed knowledge test. sealed information test. And this test uses the same polygraph equipment, but it's not trying to detect lies. Instead, what it's trying to detect is whether somebody knows facts that only the criminal would know, or should know. And the idea behind this approach is that the guilty suspects will recognize details to a crime that an innocent person shouldn't be able to. And what this, uh, the GKT is doing is it's looking for that recognition because that recognition would be paired with an increased physiological arousal. So not as detection, but more as recognition. So in terms of what the process is, the suspect is given a question and then they get multiple choice answers on what, you know, they have to answer based on these multiple choice. And the idea is when they hear the correct answer, that guilty people are going to have a spike in that arousal. So I have an example of what this would look like. Here are some questions that could have been used in the O.J. Simpson case. You know, if the glove doesn't fit, you must acquit. So here are some examples. So, you know Nicole has been found murdered, Mr. Simpson. How was she killed? Was she drowned? Was she hit on the head with something? Was she shot? Was she beaten to death? What was the crime? Was she stabbed? Was she strangled? And question two. Where did we find her body? Was it in the living room? In the driveway? By the front gate? In the kitchen? In the bedroom? By the pool? Or something else? Like, Nicole was dressed in one main color when we found her. What color was she dressed in? Was it white, black, pink, blue, green, tan? So each question that I just read had six multiple choice answers, and each one should be plausible. So if Nicole's house didn't have a movie theater, they probably shouldn't include Mood of B Theater as an option. Additionally, they usually throw out the first option because people typically have a strong physiological response to that first item. But that leaves five remaining options, and the remaining odds should be that an innocent person would have a... a 1 in 5 chance of responding to the correct detail. Additionally, you should have multiple questions as well. This is currently probably the most promising technique in terms of using polygraphing, and researchers have recently found that incorporating things like reaction time, how long it takes for them to ask a question, might even enhance that accuracy level further. There are some important limitations, though. So first, there needs to be enough crime facts from the crime scene available to create valid GKT questions. Second, this is important, these facts can't be known to the public. So only the law enforcement officers and the culprit should know these true details. So you can't have media leaking all of this information out there. Third, the guilty person must remember details surrounding the crime. You might exist. like a culprit would know say if they stabbed or shot the victim but they might not know like what the victim was wearing memories kind of a shaky sketchy thing in normal life our memories are far from perfect and especially so and say something that's so emotionally and psychologically arousing like committing a crime additionally culprits can also you know have issues remembering if say they were under the influence of drugs or drunk. Additionally, the GKT might not be helpful in all types of crime. So, say for example, a husband comes home and he sees his wife dead. Well, now the husband knows just as much as the guilty culprit would because they both saw the crime scene. And a final limitation on this is professional polygraphers aren't huge fans of the GKT because, you know, they have a lot of... When they are doing lie detection, they have a lot of control. They come up with the questions. They have to put on a good show. And sometimes they feel like the GKT is just reducing them to being technicians. So somewhat of a personal component involved as well. Alright, so if I were to ask you whether it's harder to say provide a truthful or deceptive statement What would you think? Is it easier to tell the truth or to lie? The answer according to some research is that lying is actually very very cognitively demanding a lot of cognitive resources Going to work to tell a good and believable lie take up, it's a lot harder to provide a deceptive rather than a truthful statement. Why? Well, all of those tiny pesky details that make a statement believable, they gotta come from somewhere, right? just fall from the sky. You have to make this up and you have to make it believable. And you have to do so while, say, looking believable. So don't have shifty eyes, don't fidget, don't go um, um, um. Now, of course, truth tellers do that as well. But, you know, there's a lot of going on when you're trying to provide a deceptive statement. So in terms of trying to detect lies, some strategies that researchers have identified, um, to detect lies. detect lies is they can use processes or approaches like, say, rather than having your suspect tell the story in correct chronological order, they ask them to tell it backwards. And that sounds kind of silly, like why? But for a lot of these, for these approaches, what they're trying to do is to amplify the difference between the amount of cognitive load for truthful and deceptive individuals. So if liars are already having a harder time, let's make it even harder for them. It might make it more troublesome for truth-tellers, but since they're telling the truth, it shouldn't be as detrimental to their performance. So asking people to tell the account or the story of events backwards can make it more troublesome for liars and enhances cues. that allow people to detect that they're lying. Insisting on eye contact is also something that increases that cognitive load. Asking them to draw out spatial detail, so that's additional information. Somebody's telling the truth. might not necessarily think it's important to provide that, but they have that information where somebody who's lying now has to come up with one more additional piece of information. And then asking unanticipated questions. So liars, when they're providing a count, they have kind of a gist of what the big points are going to be, what they really need to hammer on and practice. But if you catch them off guard with something, this really heightens the ability to detect truthful and deceptive statements based on the type of content and non-verbal behaviors they have as well. One interesting thing to note though is that there's a notion, so students will often ask me, how can I lie effectively? And sometimes I answer with, follow-up question or kind of a statement which is like, well, sometimes it's not the case that you want to lie effectively. Sometimes it's the case that you need to tell the truth effectively. So innocent people or people who are telling the truth have what's known as the presumption of innocence or they think that I'm telling the truth and if I just speak my truth, it'll be so obvious and so clear that there's no way that they'll think I'm lying. and that's not always the case. Especially say if you're comparing a truthful statement to somebody who's lying but knows that they're gonna come under a lot of scrutiny. So liars gonna practice, they're gonna have a rehearsal. story, whereas if I were to ask you what you did Sunday, you're not going to be able to say, well, 6.30 p.m. I started out over here for a walk, and then after that I went over here and I did this, and then I watched my favorite show at 8.30 p.m. or whatever. Instead, you're probably going to say something like, um, so, I think I, like, I had this for breakfast, and then I, um, did this with a friend, and, um, oh, I guess before that I did this. And so sometimes truth-tellers' accounts can just look really shaky or questionable because they, you know, they don't feel like they need to practice. They think that the truth will set them free, but it's easy to confuse truthful statements with deceptive statements. There are some other approaches that are also used to detect lies, and these don't need, say, expensive equipment. So one technique is called the Criteria-Based Content Analysis, or CPCA. And this uses systematic analysis of written descriptions about events that's going to try to assess the truthfulness of witnesses. And so what happens for this? Train coders examine transcribed statements that are provided by witnesses, and then they write the statements based on criteria like, say, how logical was the structure, the amount of detail, and context. Another approach is known as reality monitoring, and this analyzes written accounts of alleged crime, but this looks at things like clarity, realism, and then emotionality of the description. It also considers spatial and temporal information, and in terms of what the research says about this, research says it's accurate around 63% of the time, which is still better than that 54% of... It's like the everyday they are trying to detect accuracy, right? And a different approach is to rely on people who are supposedly better than the average person at detecting deception. They're known as lie detection wizards, and supposedly they detect lies through the power of observation. So far, there have only been 42 of these lie detection wizards. identified and these individuals seem to be accurate around like 80% of the time they supposedly rely on more subtle verbal and On verbal cues, things like microexpressions, maybe you've heard that term if you've watched the show Lie to Me, but basically like slight movements of certain areas can give away deception, some people say, or... Or they might think differently about the verbal and non-verbal cues, but then remember we said that those cues are quite wishy-washy. The explanation behind these wizards, how they are supposedly... performing their magic isn't clear. But, you know, people like to talk about the lie detection wizards, but not everybody in the lie detection literature accepts these findings. And, uh, there are people who criticize these wizards as being a statistical fluke. So, uh, they wouldn't actually be better than discriminating, just odds are that some people might be better. Uh, so, the wizards are controversial. I guess. Alright, so it's easy enough to say trick a lie detector, trick a person, trick your mom, trick your friends. So some researchers have focused on, say, whether the brain is more truthful, so to speak. Can we detect lies by looking at the person? One approach to this is through fMRI. And fMRI works by giving a video of the brain and what it's looking at is brain activity. Brain activity is going to be imaged by how much energy is in the brain. much oxygen is being used at a given site or given area, and it allows scientists to see what areas of the brain are active during different tasks. So when participants do this for deception, when participants are in these sorts of studies, what they're going to do is they lie on their back in the fMRI machine, it's about the size of a car, and then they're asked to either lie or tell the truth while the machine scans their brain. And fMRI research has shown that this, in fact, can be used to detect lies. So certain areas of the brain, like the prefrontal and the parietal cortex, are more active when people lie. And additionally, and interestingly, fMRIs can also differentiate between different types of lies. So different areas of the brain are used for spontaneous and then rehearsed lies, which I think is just delightful. There are some limitations of fMRI. So... So obviously, it's an expensive piece of equipment, not particularly portable. You have to lie very, very, very still to get these readings. And then when we're talking about the science behind it, fMRIs are tedious to run those sorts of studies. So usually there are only like 10 or 20 participants in a study, which is not really... enough to say generalized to the whole population. fMRI has also been excluded by a Tennessee federal court, so it wasn't allowed to be incorporated in a court proceeding. So those are our fMRIs. Another approach is through EEGs, which measures brain activity through numerous electrodes that are placed all over the scalp of the person. The benefit EEG has over fMRI is that it's going to be more... time sensitive so you can almost see it like immediately as the changes are happening. The fMRI on the other hand it's not as sensitive temporally but it is more detailed in exactly where is the activation happening. So fMRIs for location, EEG more for time, so there's a trade-off. But readings of these electrical patterns are measured and recorded. and this is happening when they are, say, responding to a stimulus or an event. This approach has shown evidence of their, it seems that these EEG readings are able to detect, like, after you're asked a question, there's a period where you try to suppress your initial response of telling the truth to a question, right, before the answer is then going to be substituted with a deceptive state. This also raises... is an interesting question though so you know if we have the ability to say have our brains betray our truths and our lies does using things like fmri and eeg violate like unreasonable search and seizure can we be forced to undergo an eeg or an fmri it raises ethical questions and legal questions as well you So the last thing that we'll talk about, which I think is really fun, are eyes. Eyes are particularly fun when we talk about deception because, you know, we're already biased into how we interpret eyes naturally. We're less likely to trust somebody with, say, shifting eyes. beady eyes, they want eye contact when people are talking to us, but can our eyes actually betray our lives? Some people, including myself, think that it is a very worthy investigation. And researchers can look for... changes in eye movements, they can look for things like pupil size and plinks per second, and prior research has shown that, you know, changes in, say, pupil movement size and what have you, can reveal important pieces of information. So one example of this is that when you put more effort into a task, your pupils expand. Would this not also be true for lying, if lying is supposed to be effortful? lie rather than tell the truth. So some other approaches are to like track small heat changes near the skin by the eyes. Usually at this point somebody says, well like, what about the idea that I could tell people are looking from the left to the right and that's like, if people look in the left they're supposedly lying but they're in the right and they remember or whatever it is. The answer is science does not support that. So another example of how something that we might try to rely on to decide if somebody's lying or telling the truth is being not particularly accurate. There's other work being done on ways that lies can work too, you know, that we can use eye movements to discriminate between truthful and deceptive suspects. One's going to be known as the eye movement memory assessment, which tracks visual attention attention to a scene and then they consider things like eye movement, pupil dilation, scanning pattern, and gaze fixation, or how long they look at something to indicate guilty knowledge. So this is something that's in progress. There's also what's known as laser doppler vibrometry, which tries to monitor physiological stress through an almost infrared beam aimed at the neck, and this can be done from a few feet away. So if a person may not necessarily know that they're being investigated. or that they're undergoing a lie detector test, which also raises ethical concerns, right? So we will end here with this topic for the day. I hope you guys are doing well. If you guys have any questions on anything that we've talked about or anything that I can do to help and make life easier, please feel free to reach out. I'm always happy to hear from you all. Have a great day and talk to you next time.

False? An ideal bill. So it's easier said than done. And when we're talking about the legal system, as you guys could imagine, there are probably situations where certain people, cough-cough criminals, would probably want the actual truth. be hidden.

So how do we go about testing whether somebody is lying or telling the truth? And this is going to be a critical process in many different stages of the fantasy world. investigation, and trial proceedings.

So I like to start this topic with lie detection in the olden days, lie detection across history, and even the earliest legal systems had techniques that they would use to try to discriminate between truthful and deceptive statements. Although, we can ask, you know, how exactly reliable are the legal systems? these processes were and that might be up for debate so for example there were what were called trials by a trial by ordeals which might sound like something uh in a show like game of thrones or something but back in the olden day people had a wide variety of ways that they could try to prove their innocence so they can have they could undergo something called ordeal by snake where a suspect would put their hand into a clay pot, and the clay pot would have a ring inside of it and a snake on top of the ring. So the suspect would then have to retrieve the ring from beneath the snake, and if the accused was able to do so and not get bitten, then they were innocent. So basically the snakes are lie detector experts here.

There's also ordeal by fire, where this is mostly used for women accused of adultery. But these individuals would have to walk across hot metal or hold hot metal, and they were considered to be innocent if, by the end of this, they weren't injured. Which is, you know, hard to tell.

What's the odds of not getting injured while doing that? Troublesome. So, if those don't sound appealing to you, if you are a murderer and you want to undergo a different trial, you can undergo what's known as...

as an ordeal by blood, where suspects would touch the corpse of a victim, and the victim was expected to bleed fresh blood when the true murderer touched them. So, I guess it's... At the very least, you could say the people back in the olden days were creative, I guess, right? Now we can be incredibly grateful that we don't have to undergo these horrible processes.

But a good question is, you know, do we have anything better? And when I say better, I don't mean like, you know, we're not going to get bitten by snakes or anything, hopefully. But are we any more accurate than these, what we would consider to be, largely ridiculous approaches to deception detection?

Well, let's find out. So when we are talking about lies, it's important for us to discuss why do we even bother lying? What would you guys think? So when we're talking about society, especially ours, we really seem to value honest or quote-unquote being or keeping it real, right?

Wouldn't life just be a lot easier if everybody... told the truth? And the answer is, you know, maybe it might be, but there's also social advantages to lying sometimes. So let's say you have a friend, and normally this friend is very shy.

and introverted, but for her new year's resolution, her new year, new me, she says she's going to try coming out of her shell a bit. So she finally agrees to come to a party that you're hosting, but she shows up in this monstrosity of a dress and asks if she looks okay. What are you going- to say you can think about it yeah you know are you gonna say that's the ugliest thing to be burned into your eyes as of recent that she needs to change some people may say yes hashtag keeping it real uh but Most people might not.

They might say something like, oh you look fine, or maybe some sort of middle ground. That's an interesting choice, but maybe you look a little extra. Why not borrow this instead or something like that. So rather than crushing your friend's dreams as soon as she takes the leap in coming out of isolation, maybe they try to smooth it over.

So one reason that we lie is going to be for that, you know, sometimes it just keeps things working smoothly. Social lubrication, if you will. And sometimes a little lie is going to go a long way in terms of maintaining social bonds.

If you think of people... and I know like there's this stereotype with it where it's like when you're just like oh I'm just keeping it real uh usually people kind of use that to defend you know maybe saying really hurtful things and do those people have a lot of people who enjoy hanging out with them and it depends right so in terms of the lies that we may tell socially you know think of some that you might tell you might feign interest in say like if you have a significant other who has like a project or a hobby that they personally love but you think is more boring than like watching somebody cut their toenails, you may pretend to be interested or at least humor them. You probably aren't gonna say just shut up I don't care about that. We also might lie to appear more impressive or to avoid scrutiny.

Like for example you may say oh I promise I did the homework it's just at home unfortunately even if you didn't actually do it. Additionally, if you think about it, we perform these white lies quite often as we go out through our day, and often it's the case that it's such a trivial part of our day that we might not remember it. So uh...

In addition to these more white lies, people may also lie to, say, hide their wrongdoings. So being able to catch liars is beneficial to society. We can consider this as well from an evolutionary standpoint, where our ability to lie is adaptive and beneficial to us.

So from an evolutionary standpoint, one of the special skills that humans have... have is our ability to work together, hunt, and cooperate in groups. And that is an advantage over a lot of other species. And then, in addition to that, our ability to, say, bend, spin, or shade the truth is something that also helps with keeping these groups functioning smoothly.

So if you were to tell your significant other, if you talk to me about that boring project one more time, I'm going to scream. That's probably not going to go over so well and you'll have disharmony. in your relationship. So there's a hypothesis known as the Machiavellian Intelligence Hypothesis that a great leap forward in human intelligence was triggered by our need to develop the essential social skills for deception. So basically our development of the ability to lie was a big jump for humankind.

awesome natural ability to be able to lie, it can be kind of troublesome when you're trying to decide if you're being lied to or not. Not only is the act of lying ubiquitous or ever-present throughout society, there are many different ways that we can lie. So check out all of these cool words that I found that we can use just to describe the same thing.

So we have lie. Lie. We have mislead, we have misinform, trick, fake, betray, fool, cheat, confabulate, delude, distort, fabricate, exaggerate, pretend.

misrepresent, and that's just a few of them. There's loads more that we can use, but what's particularly interesting is that each of those words all differ ever so slightly. They have a slight distinction from each other. even though the underlying action is, quote, presenting an untruth. There are many different ways that we can do that.

So in addition to these lies, we also have a distinction between what are known as lies of commission and lies of omission. Lies of commission are what people really think of, or what they're more likely to think of, when they think of lying, which is telling somebody something that isn't true. I did my homework even.

even though I didn't. Stuff like that. Alternatively, you have lies of omission, and this is going to be when you leave out helpful details, pertinent details that might reveal the truth. So, for example, a cheating spouse may say that they had dinner at a restaurant, but just leave out the not-so-important detail that it was with their affair partner. Who would want to know that, right?

Yeah, it's an important detail. Additionally, you can have both well-practiced or well-rehearsed lies, and you can compare those to spontaneous lies or lies on the fly. lies that hurt, you can also have lies that hurt other people, lies that spare people's feelings, so the intent of lies can differ, you can have lies that we feel justified in giving, or we don't feel any type of way. way compared to lies that we feel guilty about.

You can have white lies which are, you know, no real consequence. They're not really hurting anybody compared to, say, high stakes lie where the consequences are real and severe. And all of this goes into adding much more difficulty into, say, how do we approach trying to figure out what statement is true or false? If we have so many different types of lies, what on earth do we do?

do we do? So give me your opinion on this quote. So this is a quote from one of the researchers in this area. Lying is just so ordinary. It's so much a part of our everyday lives that we hardly notice it.

In many cases, it'd be much more difficult, challenging, and stressful for people to tell the truth than to lie. Take a moment and answer, do you actually agree with that statement? Is there a difference, say, between a harmless lie and a harmful lie? This is usually a fun class discussion, but, uh, you know, sometimes people will have the idea, like, no, you should never lie at any cost.

But, you know, are there situations where maybe it's easier? Better? Some people may still say no. But, you know, other people would completely disagree. Maybe you don't want to tell your friend that that's the ugliest dress, or maybe you don't want to tell your significant other you're not as attractive as they were however many years ago.

It's a troublesome dilemma. So now to the meat and the bones. Can we tell when people are lying? We might like to think that we are, but we aren't.

In fact, a meta-analysis using 24,000 participants, more or less, they found that we're only accurate around 54% of the time. of the time. And, if you do the math, you know, if we have two decisions, we can either judge a statement as true or a lie.

We essentially have a 50-50 on being right. You might as well just flip a coin. like the 54 does that really functionally differ from a 50 percent so one factor that contributes to uh us kind of sucking as lie detectors is what's known as the truth bias what the truth bias is is that you know on average we're just more likely to interpret statements as being truthful rather than interpreting or judging them as deceptive you In fact, if we stick to this, the meta-analysis with the 24,000 people, what the meta-analysis showed is that 61% of truthful statements were in fact judged as true, but only 47% of lies were actually judged as lies. And so you might take that as, wow, well, we're really good at deciding truthful statements, but what the two of those pieces of information mean together is that basically we just kind of say... True, true, true, true, true, true, true.

As opposed to actually being any better. So, uh, we're, we just have a slight tendency. This can kind of come from the expectation, well, we don't typically go out into the world expecting people to lie to us.

Now, of course, this has exceptions to that rule, but normally we operate under the idea that people... being truthful with us. You may now ask, well, what about people who are trained or who we would hope would be better at spotting lies than the average bear? And again, less than optimistically, Again, no, we're still really bad. So, in a study that had police officers, somebody who we might hope to be better than, say, a college student, researchers had police officers and college students watch true and false confessions from criminals.

So, criminals either talked about a crime that they did commit, or they described a crime that they didn't commit. And they had police and college students try to... to assess the veracity of those statements.

So college students were slightly better than Chance. What about the police officers? Well, they weren't better than Chance at all, but the police were more confident that they were correct, which is troublesome. You do not want a confident and inaccurate individual. Additionally, having more training seems to kind of bring about what's known as the lie bias.

where these statements are, these individuals are now more likely they're biased to say statements are deceptive rather than being truthful. So this kind of goes against the truth bias where they completely switch and they're now just more skeptical of all of these statements. One problem in terms of training is that it might be the training itself that's troublesome.

So if I were to ask you, how would you decide if someone was biased? somebody was lying or telling the truth, what would you say? What would you look for?

You can take a moment and write some down and think about it. You can pause the video. Okay, so if we would have done this in class, I bet I would have placed money.

But some hot answers, some answers we would have gotten would have included things like if they're stuttering, if they're making eye contact, or if their eyes are looking around, if they seem nervous, if they're fidgeting. And, uh, you guys probably played right into this mistaken liar stereotype, which is, you know, we think that we know what a liar looks like, however... as noted by the mistaken portion of that name, it's not actually helpful.

So we have a tendency to look for unhelpful behavioral cues when we're trying to assess veracity. But the truth is, there is no... obvious behavior that fully gives away when somebody's lying, so there's no fancy Pinocchio's nose or anything like that. Most behaviors actually have no relationship to deceptive behavior, and a few actually do.

best have a weak correlation. Is a weak correlation enough for somebody to go to jail over? Me personally, I would say no, but this is also a problem in training because interrogators are often going to be trained to look.

look for the specific verbal and nonverbal behaviors when they're trying to figure out if somebody's lying. So training hasn't been shown to really improve the ability to detect lies, but it just makes interrogators more confident in their abilities to do so, which they shouldn't be. And in fact, like a recent study actually showed that the police officers they surveyed, they said that they're able to detect...

around 77% of lies, when, as we just talked about, in reality, their rate is really close to that 50%. This overconfidence is also troublesome, because once an interrogator believes a person's guilty, they can then fall prey to confirmation bias, where they're going to look, they're going to interpret the suspect's behavior as if they're a criminal, meaning that they're going to look. for behaviors that confirm the, oh, this person is guilty notion.

And then they're also going to ignore behaviors that, you know, evidence would suggest that this person is not guilty. Additionally, looking for these unreliable behavioral cues has been shown by research to unfortunately occur. Where interrogators perceive innocent suspects as guilty because they didn't act, and I'm quoting, the right way during interrogations.

So if people don't act how I expect them to. they must be guilty. Another problem is that if the police believe a suspect is the culprit, even if they're wrong, they're going to be more likely to subject them to harsh interrogation techniques.

And when we have these unhelpful cues to deception being used, and then this is going to be paired with that confirmation bias, and then we double down with some high stakes, very threatening interrogations, this can and does lead to false confessions. But we'll get into that on the next video. So, take home point of this slide, we are not really good at this task.

Alright, so I just got very doom and gloom on you guys with that deception research, and what about humans as lie detectors? You guys might be feeling a little down, and you might be saying, okay, well, is there anything that we can use? we can't trust ourselves or the police to tell the difference between truths and lies.

You may say, what about all those snazzy polygraphs that I hear all about? I mean, wouldn't it just be great and delightful if we could hook suspects up to a machine, have the machine, do the dirty deception detection work. I mean, in theory, if you do that, you could just ask, like, did you kill the person? And I mean, at that point, you wouldn't even need trials, right? What a wonderful world and machine that would be.

And as you might have imagined, that magical device that detects those lies perfectly doesn't exist. But we can talk about the polygraph anyway. So the polygraph usually, in terms of like the process, it usually has people hooked up, they have a blood pressure cup around their arm, usually a pneumatic tube around their chest, and then electrodes on the fingers of one hand. The underlying theory behind how the polygraph works is pretty straightforward.

So the idea is that lying, the act of lying, causes physiological arousal. Things like increasing like increasing heart beating, sweaty palms, they're nervous, stuff like that. So as the questioner asks their questions, those deceptive statements should be paired with an increase in some sort of physiological arousal. The history of the polygraph is actually really interesting. It was created by a man called William Marston.

He was a lawyer, and feel free to insert all your lawyer jokes here. And he was also a psychologist, feel free to insert the jokes there too, who believed that these physiological changes could tell what was happening. when somebody's lying.

Also, in his spare time, he created the comic book character Wonder Woman, which is extra fun because, you know, she has that lasso of truth where she wraps people up in it, and once they're tied up, they have to tell the truth. And, you know, maybe he wanted one of his own, except his is a little not as flashy as the lasso. So, you know, If a gavelin does, that would be great, but Marston would probably say that he created one. So just to give a little bit of background, during the 60s, 70s, and most of the 80s, the polygraph was a really, really booming and lucrative business.

So there were independent contractors who offered employers quick and easy ways to decide who to hire and who to fire. So people applying for jobs were often asked to submit. to polygraph testing where they would be asked questions like, have you ever used illegal drugs?

Have you ever lied to former bosses? Did you tell the truth on all of your job application questions? Yikes. Did you steal money?

Did you steal money from your previous employers? You know, all of the juicy stuff. Big old can of yikes, right? And if these applicants...

were caught lying, they didn't get the job. But just because you got the job doesn't mean that you are free from the polygraph because employers would then also test people once they were hired as well. And because this was such a booming business, there were a lot of schools opened up at the time to train people in the art of polygraphing.

This, however, came to a screeching halt with the Polygraph Protection Act in 1988. And what this act did was it did was it prohibits most private. Employers from using polygraphs to make like job related decisions. The keyword is Private so public employers are exempt so places like the FBI CIA NSA CEA, and Secret Service, you know, no one big, no one famous, they all still rely on polygraph testing.

Also, commercial businesses that are involved with supplying like security personnel to facilities. that work with public safety are also exempt. And who does this look like? These are, just think, like nuclear power plants or public water supply facilities. So, gone the way of, you know, it's fallen a little out of favor since its big boom, but you could make a really nice salary from simply trying to figure out if somebody's lying about, say, having stolen too many staplers from a previous job.

So... The last thing that we mention are going to be what's known as countermeasures and or self-stimulation. So it's not as hard as we would like it to be to trick the machine. And there are a couple of different ways that you can do this. And this is going to be by altering your level of arousal.

So what you can do is you can either make the you try to make the comparison between truths and lies more blurred. Right. So this can either make truthful responses, you can try to pair those with more elevated arousal, or you can try to suppress that arousal through lies.

So some techniques people will do is they may bite their tongues, which hurts, and that raises arousal. They may press their toes into the floor. And these countermeasures are actually quite effective, reducing detection by around 50%.

And what's particularly troublesome is that examiners won't know. know when these countermeasures are being done. So in terms of the polygraph, you know, maybe there are other things. Maybe there are other things that are more beneficial, but a reasonable question is, well, what do the courts think about this evidence?

And if you watch shows like Law and Order SVU, they're like, oh no, this is inadmissible. And we like to think think that polygraphs are inadmissible in court, but the truth is actually a little bit more nuanced than that. So 23 states have the polygraph banned.

Others will allow it under certain circumstances, and then, like, New Mexico routinely allows polygraphs. Polygraphs fly in New Mexico. So, uh, it's not extinct, and it actually does occur in courts more often than you might think. So when jurors hear about the polygraph reports, when they come out in the trial, what do jurors think about this sort of evidence?

And for the most part, jurors find polygraphs... and it results incredibly compelling or persuasive. So for example, here's the case of a man known as Buzz Fay. Fay was sent to prison for aggravated murder and was convicted for shooting a man called Fred Erie.

during a robbery of a liquor store. After being shot, the victim was rushed to the hospital. The victim was given pain medication, and then he was asked who shot him. So while the victim was suffering from blood loss and dying, they said, I think it looked like Buzz, but it couldn't have been. And then, later on, right before he died, he said, Buzz did it.

The case was very, very weak because there was no physical evidence linking Faye to the crime. And the police knew that their case was really weak. So what they did was they made an offer to Faye's attorney that they would drop all of the charges if Faye was able to pass a lie detector test. And Faye was great.

Terrific. He knew he was innocent. And this seemed like a great way to get this over and done with.

Right. He's just going to take the test. He's not guilty. guilty, why should he worry if he's innocent, right? Wrong.

Or at least according to the polygraph. So Fay ended up failing the test, and then he failed the test again, and then he was convicted by a jury who sent him to prison based largely on those lie detector results. Additionally, this isn't the only instance of polygraphs.

Pretty much changing the court outcome either, where that might be the only piece of evidence or the best piece of evidence. And research has shown that, you know, it's backed up this notion that jurors find it compelling because... say one study they asked law students to provide they read a case and then they provided is this person guilty or not guilty 85% of law students who originally said that the defendant wasn't guilty changed their vote to guilty when they heard the defendant had failed their lie detector test and these are law students so so So although the polygraphs do not have a specific legal precedent, Justice Clarence Thomas expressed two concerns of the polygraph for the legal system. One, there's this lack of consensus for scientific validity.

And then second, it reduces the role of the jury. So, you know, is science really sold on the lie detector as a use of on the polygraph for detecting lies? No. Not really.

And then the view that, well, the jury should be the lie detector, which may still also have some caveats with that as well, too. Additionally, how effective the polygraph might be is going to depend on a couple of things that are going to vary case from case as well. So, for example, how persuasive the polygraph or testimony is.

Remember we talked about kind of the... A polygrapher might be more or less intimidating depending on, say, how they look. Similar here, so some individuals just come off as more compelling, believable, persuasive, and we really want jurors' opinions hinging on something as unrelated as that. Sophistication of jurors, so whether they are, say, up to snuff on how reliable polygraphs actually are, and then whether or not judges include, say, instructions that might warn.

the jurors about, well, you might want to consider this. Those are all things that are going to change and impact how likely those polygraphs are to be persuasive to jurors. In terms of what the experts think, experts are a little bit more skeptical than the average person in terms of, you know, their thoughts on whether this is an effective tool or not.

So if you look at the experts, most of them... are more likely to say that polygraphs are questionable, some might say it's useful, and then non-experts are going to say that it's useful, but you never have an instance of experts saying that they are reliable. So, you know, science is meh on it, but people seem to really find it compelling. A different approach is known as the guilty knowledge test, or the concealed knowledge test.

sealed information test. And this test uses the same polygraph equipment, but it's not trying to detect lies. Instead, what it's trying to detect is whether somebody knows facts that only the criminal would know, or should know. And the idea behind this approach is that the guilty suspects will recognize details to a crime that an innocent person shouldn't be able to. And what this, uh, the GKT is doing is it's looking for that recognition because that recognition would be paired with an increased physiological arousal.

So not as detection, but more as recognition. So in terms of what the process is, the suspect is given a question and then they get multiple choice answers on what, you know, they have to answer based on these multiple choice. And the idea is when they hear the correct answer, that guilty people are going to have a spike in that arousal.

So I have an example of what this would look like. Here are some questions that could have been used in the O.J. Simpson case.

You know, if the glove doesn't fit, you must acquit. So here are some examples. So, you know Nicole has been found murdered, Mr. Simpson.

How was she killed? Was she drowned? Was she hit on the head with something?

Was she shot? Was she beaten to death? What was the crime? Was she stabbed? Was she strangled?

And question two. Where did we find her body? Was it in the living room? In the driveway? By the front gate?

In the kitchen? In the bedroom? By the pool? Or something else?

Like, Nicole was dressed in one main color when we found her. What color was she dressed in? Was it white, black, pink, blue, green, tan?

So each question that I just read had six multiple choice answers, and each one should be plausible. So if Nicole's house didn't have a movie theater, they probably shouldn't include Mood of B Theater as an option. Additionally, they usually throw out the first option because people typically have a strong physiological response to that first item. But that leaves five remaining options, and the remaining odds should be that an innocent person would have a...

a 1 in 5 chance of responding to the correct detail. Additionally, you should have multiple questions as well. This is currently probably the most promising technique in terms of using polygraphing, and researchers have recently found that incorporating things like reaction time, how long it takes for them to ask a question, might even enhance that accuracy level further. There are some important limitations, though.

So first, there needs to be enough crime facts from the crime scene available to create valid GKT questions. Second, this is important, these facts can't be known to the public. So only the law enforcement officers and the culprit should know these true details.

So you can't have media leaking all of this information out there. Third, the guilty person must remember details surrounding the crime. You might exist.

like a culprit would know say if they stabbed or shot the victim but they might not know like what the victim was wearing memories kind of a shaky sketchy thing in normal life our memories are far from perfect and especially so and say something that's so emotionally and psychologically arousing like committing a crime additionally culprits can also you know have issues remembering if say they were under the influence of drugs or drunk. Additionally, the GKT might not be helpful in all types of crime. So, say for example, a husband comes home and he sees his wife dead.

Well, now the husband knows just as much as the guilty culprit would because they both saw the crime scene. And a final limitation on this is professional polygraphers aren't huge fans of the GKT because, you know, they have a lot of... When they are doing lie detection, they have a lot of control. They come up with the questions.

They have to put on a good show. And sometimes they feel like the GKT is just reducing them to being technicians. So somewhat of a personal component involved as well.

Alright, so if I were to ask you whether it's harder to say provide a truthful or deceptive statement What would you think? Is it easier to tell the truth or to lie? The answer according to some research is that lying is actually very very cognitively demanding a lot of cognitive resources Going to work to tell a good and believable lie take up, it's a lot harder to provide a deceptive rather than a truthful statement.

Why? Well, all of those tiny pesky details that make a statement believable, they gotta come from somewhere, right? just fall from the sky.

You have to make this up and you have to make it believable. And you have to do so while, say, looking believable. So don't have shifty eyes, don't fidget, don't go um, um, um.

Now, of course, truth tellers do that as well. But, you know, there's a lot of going on when you're trying to provide a deceptive statement. So in terms of trying to detect lies, some strategies that researchers have identified, um, to detect lies.

detect lies is they can use processes or approaches like, say, rather than having your suspect tell the story in correct chronological order, they ask them to tell it backwards. And that sounds kind of silly, like why? But for a lot of these, for these approaches, what they're trying to do is to amplify the difference between the amount of cognitive load for truthful and deceptive individuals. So if liars are already having a harder time, let's make it even harder for them.

It might make it more troublesome for truth-tellers, but since they're telling the truth, it shouldn't be as detrimental to their performance. So asking people to tell the account or the story of events backwards can make it more troublesome for liars and enhances cues. that allow people to detect that they're lying.

Insisting on eye contact is also something that increases that cognitive load. Asking them to draw out spatial detail, so that's additional information. Somebody's telling the truth.

might not necessarily think it's important to provide that, but they have that information where somebody who's lying now has to come up with one more additional piece of information. And then asking unanticipated questions. So liars, when they're providing a count, they have kind of a gist of what the big points are going to be, what they really need to hammer on and practice. But if you catch them off guard with something, this really heightens the ability to detect truthful and deceptive statements based on the type of content and non-verbal behaviors they have as well. One interesting thing to note though is that there's a notion, so students will often ask me, how can I lie effectively?

And sometimes I answer with, follow-up question or kind of a statement which is like, well, sometimes it's not the case that you want to lie effectively. Sometimes it's the case that you need to tell the truth effectively. So innocent people or people who are telling the truth have what's known as the presumption of innocence or they think that I'm telling the truth and if I just speak my truth, it'll be so obvious and so clear that there's no way that they'll think I'm lying.

and that's not always the case. Especially say if you're comparing a truthful statement to somebody who's lying but knows that they're gonna come under a lot of scrutiny. So liars gonna practice, they're gonna have a rehearsal.

story, whereas if I were to ask you what you did Sunday, you're not going to be able to say, well, 6.30 p.m. I started out over here for a walk, and then after that I went over here and I did this, and then I watched my favorite show at 8.30 p.m. or whatever.

Instead, you're probably going to say something like, um, so, I think I, like, I had this for breakfast, and then I, um, did this with a friend, and, um, oh, I guess before that I did this. And so sometimes truth-tellers' accounts can just look really shaky or questionable because they, you know, they don't feel like they need to practice. They think that the truth will set them free, but it's easy to confuse truthful statements with deceptive statements. There are some other approaches that are also used to detect lies, and these don't need, say, expensive equipment. So one technique is called the Criteria-Based Content Analysis, or CPCA.

And this uses systematic analysis of written descriptions about events that's going to try to assess the truthfulness of witnesses. And so what happens for this? Train coders examine transcribed statements that are provided by witnesses, and then they write the statements based on criteria like, say, how logical was the structure, the amount of detail, and context. Another approach is known as reality monitoring, and this analyzes written accounts of alleged crime, but this looks at things like clarity, realism, and then emotionality of the description. It also considers spatial and temporal information, and in terms of what the research says about this, research says it's accurate around 63% of the time, which is still better than that 54% of...

It's like the everyday they are trying to detect accuracy, right? And a different approach is to rely on people who are supposedly better than the average person at detecting deception. They're known as lie detection wizards, and supposedly they detect lies through the power of observation.

So far, there have only been 42 of these lie detection wizards. identified and these individuals seem to be accurate around like 80% of the time they supposedly rely on more subtle verbal and On verbal cues, things like microexpressions, maybe you've heard that term if you've watched the show Lie to Me, but basically like slight movements of certain areas can give away deception, some people say, or... Or they might think differently about the verbal and non-verbal cues, but then remember we said that those cues are quite wishy-washy.

The explanation behind these wizards, how they are supposedly... performing their magic isn't clear. But, you know, people like to talk about the lie detection wizards, but not everybody in the lie detection literature accepts these findings.

And, uh, there are people who criticize these wizards as being a statistical fluke. So, uh, they wouldn't actually be better than discriminating, just odds are that some people might be better. Uh, so, the wizards are controversial.

I guess. Alright, so it's easy enough to say trick a lie detector, trick a person, trick your mom, trick your friends. So some researchers have focused on, say, whether the brain is more truthful, so to speak.

Can we detect lies by looking at the person? One approach to this is through fMRI. And fMRI works by giving a video of the brain and what it's looking at is brain activity. Brain activity is going to be imaged by how much energy is in the brain.

much oxygen is being used at a given site or given area, and it allows scientists to see what areas of the brain are active during different tasks. So when participants do this for deception, when participants are in these sorts of studies, what they're going to do is they lie on their back in the fMRI machine, it's about the size of a car, and then they're asked to either lie or tell the truth while the machine scans their brain. And fMRI research has shown that this, in fact, can be used to detect lies. So certain areas of the brain, like the prefrontal and the parietal cortex, are more active when people lie. And additionally, and interestingly, fMRIs can also differentiate between different types of lies.

So different areas of the brain are used for spontaneous and then rehearsed lies, which I think is just delightful. There are some limitations of fMRI. So... So obviously, it's an expensive piece of equipment, not particularly portable.

You have to lie very, very, very still to get these readings. And then when we're talking about the science behind it, fMRIs are tedious to run those sorts of studies. So usually there are only like 10 or 20 participants in a study, which is not really...

enough to say generalized to the whole population. fMRI has also been excluded by a Tennessee federal court, so it wasn't allowed to be incorporated in a court proceeding. So those are our fMRIs. Another approach is through EEGs, which measures brain activity through numerous electrodes that are placed all over the scalp of the person.

The benefit EEG has over fMRI is that it's going to be more... time sensitive so you can almost see it like immediately as the changes are happening. The fMRI on the other hand it's not as sensitive temporally but it is more detailed in exactly where is the activation happening.

So fMRIs for location, EEG more for time, so there's a trade-off. But readings of these electrical patterns are measured and recorded. and this is happening when they are, say, responding to a stimulus or an event. This approach has shown evidence of their, it seems that these EEG readings are able to detect, like, after you're asked a question, there's a period where you try to suppress your initial response of telling the truth to a question, right, before the answer is then going to be substituted with a deceptive state. This also raises...

is an interesting question though so you know if we have the ability to say have our brains betray our truths and our lies does using things like fmri and eeg violate like unreasonable search and seizure can we be forced to undergo an eeg or an fmri it raises ethical questions and legal questions as well you So the last thing that we'll talk about, which I think is really fun, are eyes. Eyes are particularly fun when we talk about deception because, you know, we're already biased into how we interpret eyes naturally. We're less likely to trust somebody with, say, shifting eyes. beady eyes, they want eye contact when people are talking to us, but can our eyes actually betray our lives?

Some people, including myself, think that it is a very worthy investigation. And researchers can look for... changes in eye movements, they can look for things like pupil size and plinks per second, and prior research has shown that, you know, changes in, say, pupil movement size and what have you, can reveal important pieces of information.

So one example of this is that when you put more effort into a task, your pupils expand. Would this not also be true for lying, if lying is supposed to be effortful? lie rather than tell the truth. So some other approaches are to like track small heat changes near the skin by the eyes.

Usually at this point somebody says, well like, what about the idea that I could tell people are looking from the left to the right and that's like, if people look in the left they're supposedly lying but they're in the right and they remember or whatever it is. The answer is science does not support that. So another example of how something that we might try to rely on to decide if somebody's lying or telling the truth is being not particularly accurate. There's other work being done on ways that lies can work too, you know, that we can use eye movements to discriminate between truthful and deceptive suspects. One's going to be known as the eye movement memory assessment, which tracks visual attention attention to a scene and then they consider things like eye movement, pupil dilation, scanning pattern, and gaze fixation, or how long they look at something to indicate guilty knowledge.

So this is something that's in progress. There's also what's known as laser doppler vibrometry, which tries to monitor physiological stress through an almost infrared beam aimed at the neck, and this can be done from a few feet away. So if a person may not necessarily know that they're being investigated.

or that they're undergoing a lie detector test, which also raises ethical concerns, right? So we will end here with this topic for the day. I hope you guys are doing well. If you guys have any questions on anything that we've talked about or anything that I can do to help and make life easier, please feel free to reach out. I'm always happy to hear from you all.

Have a great day and talk to you next time.

Transcript for:Deception Detection in Legal Systems

Transcript for:
Deception Detection in Legal Systems