A (brief but exciting) history of lie detectors

Lies are inherent in all living things.

Let me clarify. By that, I don't mean that I just broke up with my boyfriend and now I'm going to go write a poem about the cruel deceptiveness of mankind ("The truth is dead, and black roses wilt in the betrayed jail cell of my soul"). Not remotely.

What I mean is that deceptiveness is one of the most commonly used survival tactics. Remember that moth that your third grade teacher was so excited about? The one whose wings looked exactly like tree bark? Well, camouflage is just a very basic form of deception. And as I'm sure you're aware, it's common enough in nature - I would like to draw the reader's attention to rabbits that change color with the seasons, the praying mantis, and high school.

In any case, because lies are such a fundamental part of nature, including human nature, people have trying for thousands of years to develop a way to determine when someone is being deceptive.

The first lie detector tests were more physical than psychological. In order to detect lies, they had trials by ordeal, or tests of physical strength and endurance. Some of the more common trials by ordeal included one-on-one combat (the innocent man would, of course, be stronger in battle), and tests to determine how long the accused could immerse his arm in boiling water. Another favorite involved cutting the accused, and then carefully dressing the wound and setting a 'heal by' date. If the wound hadn't healed by that time, the accused was considered guilty. Time to break out the Neosporin, boys.

Later on, more psychological tests were developed. One, used by the Hindus, required the accused to chew a mouthful of uncooked rice and then try and spit it onto a leaf. If the rice stuck to the inside of the person's mouth, they were considered to be guilty.

The Roman Catholic Church developed a similar type of lie detector specifically for those members of the clergy who were thought to have misbehaved. Not that members of the clergy have ever misbehaved. For this test, the accused priest was made to eat cheese and bread in front of a number of other priests. If the priest wasn't able to swallow his cheese and his bread, then it was assumed that God had personally reached down and stopped the food in the priest's throat in order to indicate deceptiveness.

Now, the reason that these lie detector tests worked (and I use the word 'worked' in the loosest possible sense) was fairly straightforward - they detected nervousness. When a person gets nervous, their mouth gets dry. And when their mouth gets dry, rice and/or a cheese sandwich sticks to it. So what these early lie detectors were actually detecting was fear, and not guilt.

The first real polygraph test was developed in America in the early 20th century. A police officer named Larson invented what is rather wordily referred to as a 'continuously recording interrogation polygraph' in the 1930's. In short, what that means is that he made a device that drew several graphs at the same time. He also developed what is known as the R/I method, which I will get to a little bit later.

When lie detectors are used

The most common use of the modern polygraph test is in the employment industry - that is, the business of determining whether or not an applicant should become an employee, and whether or not an employee should stay an employee. American businesses lose billions of dollars due to employee theft every year (On the night of Wednesday, June 13th, a red BIC pen disappeared from the office of the Assistant Manager of Marketing. Where - think back now - where exactly were you that evening?).

Take, first, the question of an applicant becoming an employee. Often, an employer will hire a polygraph examiner to give an applicant a psychological once-over. To do this, the examiner will force the applicant to swear that he has never done drugs, has never had so much as a strong Irish coffee to drink, has gone to church every day, often twice a day, since the age of three, etc. The polygraph machine then spits out a few graphs for the examiner to interpret. In many cases, the polygraph examiner has the deciding vote in the final decision.

When there has been more employee theft than is considered normal (a rash of mouse ball disappearances recently incited the Great Cubicle Riot of Aught-Three), owners have in recent years begun to turn to polygraph examiners before calling the police. Polygraph examiners are more likely to find someone guilty because, regardless of their intentions, the information on which their decision rests is just more subjective than that of the police. An examiner walks into the building knowing, knowing that someone is guilty. He is being paid to find that someone and, like most people being paid good money to do a job, he feels inclined, even obligated, to succeed.

And now, a Hollywood special: the only other situation in which polygraph tests are commonly administered is in criminal cases.  For reasons that remain hazy, New Mexico is the only state in which polygraph test results are generally admissible. In all other states, both sides need to sign a stipulation agreeing that the results of the test will be considered admissible regardless of the outcome.

This is fine, except for the unfortunate fact that it often results in completely innocent people getting thoroughly screwed.

How lie detectors work

The machine itself has five attachments that must be placed on the subject before the polygraph test begins (“You survived the fire swamp, so you must be very brave, but no one withstands the machine”). These are used to gather the data from which the various graphs are made, and also to ensure that any sense of personal space or comfort has been well and truly violated.

First, there are two rubber tubes filled with air, unnecessarily referred to as pneumographs, and they are placed around the subject’s chest and abdomen. When muscles in said chest or abdomen expand, the air inside the tubes is displaced, allowing the polygraph machine to monitor respiratory rate. A blood pressure cuff is placed around the subject’s upper arm, and is used to measure both blood pressure and heart rate.

Finally, two galvanometers are placed on two of the subject’s fingertips to measure galvanic skin resistance (See also: sweat). These rather cleverly measure the skin’s ability to conduct electricity, which improves with perspiration. Some polygraph machines also record arm and leg movement, and some examiners use these charts in their diagnosis.

And now, I am going to spend roughly thirty minutes painstakingly recreating a real polygraph chart. Mine will be made entirely of hyphens.

                                      DECEPTIVE CHART                                                                             TRUTHFUL CHART
                                                                              --      --                   
                                                                     --   ---          -----             
                                                                  -- --- -                -----                               --                                                                                      ---
       -----  --      ------                          -----                              -----                      ----   --    -----                      ---          ----  --          -------     --   
-- -              ---             ------         -----                                           ---            ---              --         -----        ----     -----              ----           ----                                                   
                                              ---                                                                                                      ---                                                             

Look at the deceptive chart first. Two types of questions have been asked: a control question (“is your name Alice?”), and a relevant question (“did you brutally murder several of your coworkers last week?”). When the subject is lying – that is to say, when they did in fact murder said coworkers– he responds more strongly to the relevant question than to the control question. Scientists everywhere are totally baffled by this behavioral pattern.

On the truthful chart, however, the subject responds about equally to the relevant question and to the control question. Now, if you think about it, this is actually a far more disturbing result – the subject is equally calm discussing his name and discussing the murders that he is suspected of having committed.

I suppose if his name is really Alice, though, that’s a bit more understandable.

Now, the effectiveness of the polygraph test depends almost entirely on the subject’s belief in its infallibility. The idea runs, basically, that if the subject is innocent he will believe so strongly in the abilities of the polygraph machine that he won’t show any signs of nervousness. If the subject is being deceptive, however, he will be certain that the polygraph machine can 'read' him and will therefore give himself away through heightened emotional responses. Also, since the polygraph’s best use is in eliciting confessions, it serves the examiner’s purpose to have the subject believe that the polygraph machine is completely accurate.

America has bought into the polygraph machine wholeheartedly. Europeans, on the other hand, firmly refuse to have anything to do with them. A room full of educated Europeans and Americans was once polled, and while 96% of the Americans said that they were willing to submit to a polygraph test, only 4% of the Europeans agreed.

In the old US of A polygraph tests are right up there with WeightWatchers, the Red Sox, and apple pie. In fact, Americans are so convinced that lie detectors can actually detect lying that some even believe the polygraph test results over their own memories. This is demonstrated rather spectacularly in the case of Peter Reilly.

Mr. Reilly returned home one day in 1974 to find his mother dying from multiple nasty, stabby wounds on the kitchen floor. The police showed up, smoked a few cigarettes while poking around the crime scene, and then had one of their own administer a lie detector test (police forces often keep a trained polygraph examiner on staff). When they hooked the unfortunate (and probably mildly hysterical) Mr. Reilly up to the polygraph machine a few hours later, they discovered that he responded strongly to the question, “Did you stab your mother repeatedly, and then go off to chuckle about it while she lay dying in the kitchen?” Well, maybe they didn’t phrase it quite like that. In any case, Reilly exhibited what the polygrapher referred to as a ‘strong psychological response,’ which indicated deceptiveness. Because there’s nothing at all inherently upsetting about the question, “did you stab your mother?”

The police kept at him, hinting that sometimes people blacked out and committed extremely violent acts that they could not later recall. They stressed the total infallibility of the polygraph test. They asked him if he had ever felt angry with his mother. They pointed out that no one else was around to have stabbed her. They waved the murder weapon (a kitchen knife) in his face, and asked him if he had used it to kill his mother. When he said no, they asked him if he had seen it before. When he said no again, they asked him if he had ever seen anything like it... anything vaguely knife-shaped, or at least metal... prior to the incident. And after twelve hours of this, Peter Reilly came to the conclusion that he had murdered his mother in a fit of rage, and then instantly blocked the memory out. He was even willing to sign confessions, pleading guilty to every accusation and tearfully apologizing to anyone who would listen.

A few years later, the real killer was nabbed for something else entirely and confessed to the murder of Peter Reilly’s mother. Eeeesh.

And a dash of irony

One of the ways that polygraph examiners instill this brand of total belief is to use what are called ‘stim tests,’ or stimulation procedure tests. The most common one involves, as do all good magic tricks, a pack of cards. The polygraph examiner hooks the subject up to the machine before the actual lie detector test begins. The examiner then pulls out a pack of cards, and asks the subject to pick one without letting him see it. Lets say it's the King of Diamonds. The examiner first has the subject say that he is holding a face card, and then that he is not holding a face card. After studying the resulting charts for a few moments, the examiner declares that the former is true. Cue astonishment.

The polygraph examiner then decides to go a step further, and has the subject say that it is not in the suit of Spades, not in the suit of Hearts, etc. The examiner states unequivocally that the card is in the suit of Diamonds. Finally they run through the different face cards, and the examiner finishes with a flourish, modestly giving all the credit to the powerful mind-reading machine on the table between them. "The machine says that the card you are holding is the King of Diamonds. Is that correct?"

More astonishment, and a growing feeling of unease on the part of the subject. Even the innocent ones. I actually had the opportunity to perform this particular stim test on an unsuspecting classmate of mine when I gave a presentation on polygraph tests. She described it as both spooky and invasive. Actually, she said that it was "like that thing could, like, totally read my mind. Like it knew what I was thinking, sorta. Like it was inside my fucking head. Whoa. I mean, really whoa."

While they aren't always that articulate about it, most subjects feel something along the same lines. But you want to know the dirty secret? The really filthy irony of it all? Of course you do. The truth is that if you looked through the pack of cards that your smiling polygraph examiner (that is, the one administering the lie detector test) is holding, you would find no less than fifty-two King of Diamonds cards.

They don't let the subject in on that little secret, of course.

So, how to polygraph examiners justify blatantly deceiving just about everyone who walks into their office? Well, the idea runs basically as follows: for the innocent subjects to respond equally to control and relevant questions, they have to believe that the polygraph machine is 100% effective, and never fails. The problem with this convenient line of reasoning? It isn't, and it does.

Different Breeds of Examiner

Polygraph examiners are, on the whole, a rather strange and antisocial bunch. I suppose they would have to be, given their chosen profession. In any case, my attempts to infiltrate their tightly-knit group were largely unsuccessful (one or two even stood up to the 'admiring youngster' act), so I can only describe one major schism in polygraph testing theory.

There are really only two main types of polygraph examiners; clinical examiners, and straight chart examiners (Or 'SCE,' as the boys around the office have taken to calling them). I'm going to cover the SCE first. What that means is that the examiner uses only the polygraph charts to make his decision, and nothing else. The examiner is not supposed to take into account the active signs of nervousness that the subject was exhibiting, or those streaks of dried blood on the subject's lapel. The charts are scored using a standard scale, and the decision is reported exactly according to the charts. Very numerical. Sometimes, in order to ensure that the examiner's decision wasn't tainted by his own damned observant nature, the charts are scored by someone else entirely. Either way, very dangerous. You go first.

The second kind of polygraph examiner considers himself to be as much a lie detector as the machine on the table in front of him. He feels that he is a highly-trained agent of Truth - practically a Honesty Ninja, in fact - and that his observations should therefore heavily influence the final decision. This type of examiner is, if anything, even more dangerous than the SCEs. The truly astounding subjectivity of their decision is partially acknowledged in the fact that what they conclude is called a 'diagnosis,' rather than a 'result.'

A man named Reid developed the aptly named Reid Symptom Chart, which was intended to help clinical examiners determine when someone is being deceptive. I believe the take-home message is something along the lines of "Smile big, sit still, and eat before you go."

                                        Deceptive                                            Truthful
                             Late for appointment                     Eager to take the test
                      Nervous, resentful, aggressive                Feelings of confidence
                         Exhibiting mental blocks            Attitude of sincerity/straightforward
                        Dry mouth/gurgling stomach              Composed/cooperative
                            Overly friendly/polite                 
                                Moving restlessly
                          Claiming to be religious
        Complaining of pain (from blood pressure cuff)
                                Eager to leave
                           Folding hands together

Many similar charts have been created over the years, and lots of seemingly intelligent people believe in them wholeheartedly. A well-respected judge in the higher court system, for example, recently resigned and decided to let the public in on a little secret. For over twenty years, he said, he had been able to determine without question who was guilty and who was innocent. As a judge, he could then treat them accordingly. You can tell, he told the eager listeners, because a guilty person will always touch their nose during the trial.

No bullshit.

Types of Polygraph Tests

Relevant/Irrelevant (R/I) Test

The first type of polygraph test was developed in the 1920's, and it is still commonly used today. It is called the Relevent/Irrelevent test, or R/I test. For this exam, there are two types of questions: relevant questions ("Did you kill her?"), and Irrelevant questions ("Is today Tuesday?"). Three relevent questions are asked, and each one is preceded and followed by an irrelevant question. A subject is considered to be deceptive if their reaction to the relevant questions is much stronger than their reaction to the irrelevant questions. Now, the problem with this should be immediately obvious. Some questions are just more inherently upsetting than others, and just as much so for the wrongfully accused as for the murderers.

Because of this, the chance of an innocent person being classified as deceptive is actually greater than if the judge had just flipped a coin. Downright spooky, if you ask me. Fortunately, those few polygraph examiners who acknowledge this flaw have been working round the clock from the 1920's on to develop more accurate methods.

Lie Control Test (LCT)

The LCT is like the R/I test, except that what are called 'control' questions are used. Before the test, the examiner explains that the control questions are used to 'get to know' the subject. For example, the polygraph examiner might ask if the subject has ever cheated on income taxes. If the subject says no, it is assumed that he is lying. The emotional response to this 'known' lie (on the polygraph charts) can then be compared to their reaction to the relevant questions. If the reaction to the relevant question is much larger than the reaction to the control, the subject is clearly lying.

The problem with this, succinctly, is that not all lies are psychologically equal. Denying something that is commonly done, such as cheating on income taxes, is not the psychological equivalent of denying that you killed someone. Other control questions include "have you ever stolen money from a company while employed there," and "have you ever done drugs?"

There is a second problem with the concept of control questions, of course. What if Jane Smith really hasn't ever cheated on her income taxes? She says 'no' the way she was instructed to, and does so with the warm, proud glow of excessive honesty in her heart. Too bad that she's thrown in prison for murder a few months later. Maybe her attorney will come and visit once in a while.

Truth Control Test (TCT)

Next, the few polygraph examiners left who still maintained a modicum of rationality decided to try the same theory, but in reverse. They developed what is known as a 'guilt complex comparison question,' and then failed spectacularly to put it to the test. For this type of test, the examiner invents an entirely fictitious crime of a similar type and magnitude to the subject's original crime. They pretend to be examining the subject about both. Say, for example, that a nice digital camera went missing from the CEO's office. The bold little punk who may or may not have done it is dragged in front of a polygraph machine, and told that the CEO's nice wrist watch also went missing a few weeks back. The boss didn't report it at the time, they tell him, but now he's looking for the culprit.

Because the subject is known to be telling the truth when they deny involvement in the second crime (remember, it's fictitious), their emotional response to those accusations and the real accusations can be compared. If the subject reacted quite a bit more to the real crime than to the invented crime, they are classified as being deceptive. The good thing about this method is that it allows the examiner to set a baseline; if the subject is that emotionally upset to begin with, a certain amount must be factored out of her final polygraph charts in order to make the readings accurate.

This sounds quite a bit more promising, but there are two problems with it. The first is that the subject has to actually believe - completely, 100% bought in - that they are in just as much danger from the fictitious crime as from the real one. If they are innocent their emotional response to the control question, which they had to have answered honestly, should be the same as their emotional response to the relevant question. Unfortunately, this kind of belief in a secondary crime can be difficult to instill in someone - I mean, how many watches does the boss really have? Seems kinda fishy, if you ask me. If the person has even some doubt as to the validity of the second accusation, then their emotional responses to questions regarding that crime will be significantly smaller. So they seem guilty, when really all they are is a good, home-grown, modern day skeptic.

The second problem with this type of lie detector test is that it requires deceiving the subject, which is supposed to be some kind of moral issue, or something. Fortunately, most polygraph examiners don't let that bother them.

Positive Control Test (PCT)

Subjects taking the PCT are asked only relevant questions, and they are instructed to answer each one twice; the first time, they are supposed to answer 'yes,' and the second time, they are supposed to answer 'no.' Since the question is the same, theoretically the only factor that could account for different reactions is that one answer is known to be deceptive. So, if the subject responds more strongly when they say that they did not commit the murder, it would imply that they did, in fact, commit it.

Do you see the problem?

As with some of the previous, the problem falls squarely in the interpretation. Different factors could be responsible - maybe the person reacts more strongly to the 'did not' reply because they're worried that the authorities don't believe them. Alternatively, this might get a real killer pronounced innocent, when he is more upset by the admission of true guilt. Also, again, some sentences are just more disturbing to say or think than others. Admitting to rape, for instance, would cause a rather large emotional response in a rape victim.

Echhh. Gets pretty muddy, don't it?

Relevant Control Test (RCT)

This is more of a general test, used when there's no single crime being discussed - for example, during job application screening. The subject is asked vague relevant and irrelevant questions such as "is your last name Tong," followed by "have you ever used drugs before coming to work?" Theoretically (and I have a homework assignment for you - go back and count the number of times the word 'theoretically' appears in the writeup), since there's no specific crime, a reaction to any one of the relevant questions should indicate deceptiveness in that area.

What we have here, kids, is another relatively useless style of polygraph exam. Suppose, just for a fun experiment, that a friend of yours recently died of a drug overdose. You yourself had never used drugs (we-ell... not since Harvard, anyway), and you had been trying to get him help for months now. You apply for a new job, and they begin asking you all kinds of questions about drugs, drug use, etc. Pop Quiz: are you exhibiting a strong emotional reaction to the questions?

(A) No, because I never liked that guy anyway.
(B) Yes, but I'm responding even more strongly to the questions about child molestation because of the... 'incidents'
(C) Yes.

There are a thousand other reasons for someone to respond more strongly to one type of question than to another, and most of them don't indicate any deceptiveness on the part of the subject. 

Searching Peak of Tension (SPOT) Test / Guilty Knowledge Test

Instead of detecting lying, this type of lie detector test detects guilty knowledge. As you might have guessed from the absence of the word ‘theoretically’ in that last sentence, this is by far the most promising technique developed so far. What this type of test does is look for a ‘peak of tension,’ or point where the emotional reaction is the highest by a considerable amount, rather than just comparing responses. 

This is best used, for example, in a kidnapping case. Say that seven suspects are brought in. Each is shown a series of photographs of children of about the right age, one of which is the kidnapped child. Now, only the kidnapper would know which child had actually been kidnapped, and so seeing that photograph in particular would (and here, unfortunately, I must use the t-word) theoretically trigger a stronger emotional reaction.

The other good thing about this type of test is that it provides a baseline; if one of the suspects is reacting very strongly to all the photographs, it can be assumed that he is just generally upset, and not necessarily guilty.

The other time that this test is useful is even more morbid: crime scenes. The cops are instructed to take pictures of the scene with the dead body dragged to different locations, and these pictures are then shown to the suspects. Though all are gruesome, only the real murderer should know which one shows where the victim actually died.

Now… the problems. The first problem is that the media tends to get involved, splashing pictures of the kidnapped kid all over the newspapers in the admittedly reasonable hope that he will turn up. Once pictures of the child are widely available, however, the police can no longer use the SPOT test to determine who is the kidnapper.

The second problem is, simply, psychopaths. It seems fairly intuitive that the kind of person capable of committing a grisly murder is likely to be the kind of person who isn’t overly affected by it. Not always, certainly, but it does make the test slightly pointless.

Black Sheep Syndrome (BSS)

Voice Stress Analysis is sort of the uncle that nobody talks to of lie detector tests. The uncle that's always getting drunk at three in the afternoon, and then shouting bad jokes that he forgets the endings to while simultaneously trying to flirt with one of his niece's friends.

Err... that is, voice stress analysis is what the polygraph examiners keep around to make them feel better about their own jobs. Voice Stress Analysis operates on the basic and largely flawed principle that emotional arousal can be detected through vibrations in the throat and larynx muscles. It’s measuring stress instead of lying, and not even doing that terribly well.  

This technique was, of course, developed in the 1960’s. It was immediately used on old clippings from the Lee Harvey Oswald trial (the person doing the testing came to the conclusion that Oswald was innocent, which I think tells you all you need to know about voice stress analysis).

In a masochistic sort of reverse case study, one supporter of voice stress analysis asked a group of people to look at a number of slides, half of them pastoral landscapes and half of them mutilated corpses. The subjects were then asked to say “yes, I like this picture” in response to each. Three judges studied the results independently, and tried to guess in each case which type of picture the subject was observing. Sorting at random, let me remind you, would have produced 50% correct guesses for each judge.

Judges One, Two, And Three were correct about, respectively, 50%, 40%, and 35% of the pictures. From this, we must conclude one of two things; voice stress analysis is worse than useless, or there are an awful lot of sick people wandering around out there.

The Second Half of this Node (Oh, God! Will it never end?) 

The real problem with lie detectors (and forgive me if you have already drawn this conclusion yourself) is that they don’t work. Polygraph examiners claim accuracy ranging from 95% to 99%, but there are three problems with this.

(1) Polygraph examiners want the machines to work, and so they will creatively interpret the data. Not lie, precisely, but… well, I like ‘creatively interpret.’ They are more likely to remember the good (times when they were right) than the bad. This is of course perfectly normal (because who wants to acknowledge the fact that their chosen profession is complete bunk?), but it does lead to a certain bias
(2) Polygraph tests are extremely intimidating (at least, if you’re American they are), and so they tend to elicit confessions. A polygraph examiner considers a confession to be a correct diagnosis, even if the charts said differently. Some polygraphers never ever bother to score the charts and determine what they would have decided on their own.
(3) Finally, for the vast majority of their cases, polygraph examiners never know whether they were right or wrong. If a subject who was diagnosed as ‘deceptive’ goes to jail (even if it’s on the strength of the diagnosis), then the examiner can feel free to chalk it up as another brilliant success in the name of Science.

As in the case of Peter Reilly, however, many Americans believe so strongly in the results that they can convince themselves that their own memory is flawed, rather than acknowledge that the machine might be. Polygraph machines actually have a validity that varies between 70%, for the good ones, and closer to 50%, for the bad ones. That’s only a little better than a coin toss. Since polygraph machines offer a solid answer (a guilty/innocent verdict) rather than testimony, they need an accuracy far better than 70%.

How to Approach Taking the Polygraph Test

Did you note the careful wording? Is it too late in the node to take up political correctness as a hobby?


How to Beat a Polygraph Test

There, that’s better.

Now, chances are fairly good that you will, at some point, be asked to take a polygraph test. If not for murder, then probably for a job application.

The first step in beating a polygraph test is to completely refuse to take one. The test is never obligatory, and in many cases it can be turned down flat. Sometimes, however, social pressure rears its ugly head. Refusing to submit to a polygraph exam can be interpreted as an admission of guilt, and for that reason it can’t always be avoided.

There are two types of countermeasures to take if you are forced to take a polygraph test.

Behavioral Manipulation

First of all, make no admissions, except if you’re asked if you’ve ever committed some extremely minor crime such as lying to your parents in high school (‘extremely minor’, right, dad?), or accidentally stealing pens from work. In any case, sign absolutely no statements. Even if it’s something you admitted to during the pre-test interview or as part of a control question – anything you sign can quite easily be used against you later in a variety of vicious and diabolic ways.

Remember not to trust your polygraph examiner. It isn’t an interview, it’s an interrogation, and you should be able to recognize the common interrogation techniques. The most common is a projection of a false sense of empathy, trying to make you believe that your polygraph examiner is your friend, and is there to help you. He isn’t. Don’t believe his lies. Paranoid? Why, no. What on earth would make you think that?

Other examiners will take an aggressive approach and try to appear threatening. These two techniques are sometimes combined, for extra fun and excitement. Think ‘good cop, bad cop.’

Finally, a particularly cruel brand of job applicant screening examiners will use an egotistical approach, during which they play up your accomplishments and good points. This is designed to make you feel that you aren’t just some ordinary applicant, and that the company will go to any lengths to hire you.

Other ways you can help your case is to make a good first impression by arriving on time, dressing professionally, and keeping eye contact.

Remember – you are being watched. In the waiting room, during the pre-test interview, and during the testing. Some polygraph machines even have extra sensors that detect motion. During the pre-test interview it is important to be friendly and polite, to admit to nothing more than some minor childhood misdeeds, and to keep your answers fairly short. Avoid, for example, saying “yes, basically” when you can just say “yes.”

Why do I feel like I’m leading some kind of morally ambiguousLearn to Lie in 10 Easy Steps” type seminar?

No matter. Now, polygraph examiners love mind games. Some of the most common ones are below:

(1) Upon entering the room, you may find your chair facing the wrong direction. By making you move it, the examiner is (subtly?) indicating that he is in control of the situation.
(2) Here’s an especially diabolical one. You are instructed to go into the next room and wash your hands. According to the examiner, this is so that the machine can get a better reading. A filthy lie. You are then observed (told you). If you fail to wash your hands, it is interpreted as a desire to fool the machine, which is a strong indicator of deceptiveness.
(3) And then, for the slightly dimmer bulbs out there, polygraph examiners have developed another commonly used mind game. You are left alone in the room on some pretext. As the examiner is leaving, he instructs you not to touch the machine, because even the slightest change will completely invalidate the results. You’re then watched, to see if you will try and loosen any wires, etc. Of course, you won’t try. You’re much too bright for that. But it’s terrifying how many people do.

One other behavioral manipulation. Most polygraph examiners will ask you at some point how much you know about polygraph testing. Don’t admit to knowing much, because that will immediately make the examiner suspicious. Instead, say something like “I heard on TV that they’re pretty much always right.” Improvise.

Chart-recording Manipulation

First off, there are some breathing countermeasures that will help steady your respiratory rate during the exam. Practice maintaining a baseline breathing rate of 15-30 shallow breaths per minute. Keep this up from the moment the pneumograph tube is placed around your chest. Remember not to change the pattern once the last question has been asked – an especially sneaky examiner may continue recording for several minutes.

When you recognize a control question (“Is you name…?”), change your breathing pattern to one of the ones below:

-Hold your breath for 4-5 seconds after breathing out.
-Breathe more slowly for the next 5-15 seconds, ending before the next question is asked. 
-Take several shallower breaths for 5-15 seconds, ending before the next question is asked.  
-Allow your breathing to get gradually shallower for 5-15 seconds, then return immediately to your baseline

That takes care of breathing, but you have more than just the pneumograph strapped to you during the exam. The second type of chart-recording manipulation are Cardio/Electrodermal countermeasures. Use only one of the following:

(1) Mental countermeasures, such as doing mathematical calculations in your head as quickly as you can. Pick an arbitrary number and count backwards by sevens. Alternatively, think about something upsetting, like falling off a cliff.
(2) Physical countermeasures, such as biting down on the side of your tongue. Do this as soon as you recognize a control question or just after answering it, and continue biting for no longer than 20 seconds, pausing to answer the question. A potentially harmful alternative is placing a tack in your sock, and pressing your toe against it at the right moment.

Finally, during the stim test, make sure to augment your responses so that when you respond strongly to control questions (respond, in fact, as though you had just stepped on something sharp), the examiner will assume that the heightened response is normal. And... well, that's about it.


Sources (only the best for Everything2):
<http://www.antipolygraph.org> (if you have to take a polygraph test, I recommend downloading their free book on how to... ahem... approach the lie detector exam).
Lykken, David T. A Tremor in the Blood: Uses and Abuses of the Lie Detector. 1998.

Log in or register to write something here or to contact authors.