How Lie Detectors Work

Instead of actually detecting lies, polygraph machines sense physiological variations, ostensibly brought on by guilt. The results are subject to interpretation, and therefore controversial. Join Josh and Chuck as they investigate the polygraph.

Learn more about your ad-choices at https://www.iheartpodcastnetwork.com

See omnystudio.com/listener for privacy information.

Episode Transcript

Speaker 1 (00:00):
Brought to you by the reinvented two thousand twelve camera.
It's ready. Are you welcome to Stuff you Should Know
from House Stuff Works dot com? Hey, and welcome to
the podcast. I'm Josh Clark. With me is always a
child to beat? Chuck Bryant liar you you could tell

(00:25):
there's the ways you can find out Chuck. Sure. Um,
we'll get to that in a minute. There's stuff you
should know. Let me finish. Okay, and it's you lie.
Remember that guy Joe joey Pants or whatever the congressman's
name was. Yeah, Saturday night Life. That is a funny
skit that he had gotten a whole group of people
that all stayed up at once, wasn't the deal? And

(00:46):
well yeah, and he supposed he had a whole group
of senators that we're gonna all stand up and yell
you lie. And then he was the only one that
did it because he was out of the room when
they were like, no, we can't do that. I think
that will say that that's funny. Yeah. So um, we're
talking about a lot of detectors. But let me take
you back to a little place in time and space

(01:09):
called the jazz age. Early nineteen twenties. Yeah, no, that's
the beat Nicks that did that. Um, I'm sure a
jazz person snapped their fingers at one point. Um, but
not like that, right, It was more like like Coltrane style,

(01:29):
just like that. Anyway. This is Chuck in Berkeley, California,
at u c l A, Berkeley, And there is a
place there called um the College Hall, which was a
women's dorm. And in that year there were a theft,
a string of there was a string of um thefts,

(01:50):
cash rings, um, pretty much anything of value went missing
for a little while there. And uh, there is a
man working at the Berkeley Police Department, Okay, called College Hall.
There was a man working in the in the Berkeley
Police Department. His name is John Larson. And John Larson
was the first cop ever to have a PhD. And

(02:14):
he had gotten interested in this device called a cardio
newmo psychograph, which had been invented just a few years
before by another guy named William Marston. And William Marson
was a lawyer in a Harvard shrink and he also,
as an aside, created Wonder Woman with her Lasso of Truth.

(02:38):
He's the guy who invented the what's now called the polygraph.
But what about the Wonder Woman? He created Wonder Woman,
the character he was kind of a renaissance man. But
that's William Marson. John Larson works at the at the
Berkeley Police Department, and he's become interested in this thing,
the cardio neumo psychograph, and he realizes, okay, this is

(02:59):
a perfect chance to apply it. So he rounds up,
you know, some suspects. He does some some normal police
work and finds out who the suspects are in this
in this hall, and he rounds him up, brings him
down the station, and he starts hooking people up to
this um, this machine. And he gets to this one woman,
her name is Helen Graham, and guilty, yes, pretty much

(03:22):
is what he does. He goes U ms Graham, this
machine is saying that you're that you took this, that
you you took the money, you know, did you? Um?
And he said that he noted on the machine a
sharp drop in blood pressure followed by a sudden rise.
And then after that this woman flew into a rate.
She tried to attack the machine. She went crazy, so
they they basically string her along for a few days

(03:44):
and then finally she confesses and it's the first time
that a polygraph was ever used to to um solve
a crime. Ever, that that was probably the heyday because
before the people knew what it was, they could just
say this machine says that you're guilty, and they would be,
oh my god, that's exactly right. Very early on, some
of the early proponents, specifically a guy named a Leonard

(04:07):
keeler Um recognized the placebo effect value before anyone knew
there was a placebo effect. That the placebo effect value
of a polygraph. That just the idea if you believed
in this machine and that it could rude outlies, then
it could force you to confess. Just being hooked up
to it, you weren't going to pass it. They should
have called it the guilt box. They called it the

(04:30):
magic lie detector is one of the one of the
facts that they called it. Yeah, Leonard Keiler called it that.
He worked with John Larson at the Berkeley Police Department,
and eventually, over time John Larson saw the what he
considered the truth behind the lie detector and the fact
that it kept being cold light detector, which is driving
him crazy um and he eventually distanced himself from it.

(04:51):
Later on in his career. UM, but Leonard Keeler ran
around marketing it to anyone and everyone, saying, just having
this is going to not only help you hire um,
more more truthful, forthright people, but it's going to keep
them in line while they're working for you, because they
know you've got access to this thing and you can
strap them to it at any time. Yeah, so that's

(05:13):
where the polygraph came from. Yeah, there's a little prehistory
to just to give them their due. Uh. In e
Cesaire lambroso he's an Italian criminologist. He measured changes in
blood pressure for police cases in in nineteen o four
a device by Vittorio Bernosi measured breathing and so they

(05:34):
were early nineteen hundreds, late eighteen hundreds. They were kind
of on the scene of of measuring these things in.
Dr James Mackenzie in nineteen o six first mentioned the
word polygraph with his instrument that he didn't use to
root out the truth, but for uh he did use
it when giving medical examination. ZO and then right before

(05:56):
the polygraph was the unigraph unigraph, which was part of
what's still used today in the polygraph UM. It measured respiration.
Pretty cool, yeah, Um, but then you add to it
a couple of other things and you got the polygraph.
We could stop here. This is interesting enough right now.
So there's no uh you really, there's no um, no

(06:19):
one walking the planet who has anything to do with
polygraphs that call them lie detectors, And anyone, even the
most ardent defender of polygraph technology would correct you if
you called it a lie detector. They would be like,
it's it's not a lie detector because you can't detect
a lot. It's possible. The whole basis of a polygraph

(06:42):
is that it is a set of um medical instruments
that use to measure changes in things like your heart rate,
your respiration um and sweatiness. Basically, I would fail. Um, well,
a lot of people do fail, and we'll get to that.
But um oh because of your sweat Yeah, that's okay.

(07:04):
They would they would even hook me up. Um. So,
when you're hooked up to this machine. The whole point
is that it measures these physiological changes in the idea
that there you're going to undergo physiological changes based on
the concept that a person hooked this machine who is
guilty will experience fear that they're going to be detected.

(07:28):
So this machine is designed to detect that fear, that's right,
which is really round about. But for a century almost
these things were used um and abused, and it took
a while for people to kind of catch on that.
There's a lot to criticize here with polygraphs, yeah, for sure,
but they're still used, all right, So let's get into

(07:50):
this um. First of all, we need to point out
that analog polygraphs are what you have long seen in
movies and TV when they have the little jit he
uh looks like a seismic graph on the on the
paper scrolling by, and uh, you're hooked up to all
these different things on your chest, in your forehead and
your fingertips. These days they do that digitally, but it's

(08:13):
basically still the same technique. They just don't use the
little scrolling needle. Do They have a name for that.
It's called an inkfield pen. It is, okay. But the
three things that they do, Josh, They measure your respiratory rate,
as you said, They take newmographs, which are rubber tubes
filled with air time around your chest and your abdomen,

(08:36):
and that is going to measure whether or not you're
you know, you start breathing heavy essentially when you get nervous.
It monitors your breathing pattern and any changes to it,
and it does it pretty cleverly. Right. Yeah. With bellows,
it it's they're filled with air. So when you when
you breathe in real deeply or have a change, it's
going to displace that into the bellows and that will

(08:57):
Originally that was attached the bellows were literally attached to
the mechanical arm that showed the change. These days is
a it's a transducer that converts it digitally electronically, right,
it converts it to an electrical pattern, right, Yeah, the
same thing. Yeah, Um no, I think it looks a
lot like it. If you look, there's a picture of

(09:18):
a modern one in the ground. It looks just like
But yeah, it's not a it's not a paper read
out any longer, which is kind of interesting, Like this
technology hasn't hasn't changed on a very fundamental basis for
like a hundred years almost. Yeah, I mean the early
one from Mackenzie in nineteen o six. They say that
a lot of the same components are still very similar today. Right. Um,

(09:40):
you also are going to have so you're gonna have
two tubes, one around your chest, one around your abdomen.
It's um keeping an eye on your breathing. Um. You're
going to have a blood pressure cuff which UM, which
keeps an eye on your heart rate and your blood pressure.
And it does it through sound, right, Yeah, realize this.
So when you're when the blood comes in and out

(10:03):
of your veins, it creates sound. And sound can also
be used to displace air, causing of bellows to contract,
which again moved the arm on the scroll and now
is created or turned into an electrical pattern. Um. But
it's the same thing, but it's sound, which I just
think is very neat. Well, and what's also neat is

(10:25):
the sweat one. I figured they would have some sort
of like a moisturerometer just to detect moisture. But it's
called galvanic skin resistance or GSR or electrodermal activity, that's right.
And they hook up these fingerplates to galvanometers and they
are basically measuring the skin's ability to conduct electricity. And

(10:45):
if your skin is moist it's going to be able
to conduct electricity easier. And that's what they're measuring there.
It's like the ones the little um heart rate monitors
that they clip to your fingertips in the hospital. But
these things as your electricity instead, uh, which if you
are dry, you're going to conduct less of electricity. If

(11:09):
you're wet, you're gonna conduct more so since you have
so many pores on the end of your fingers and
you sweat when you're nervous. There you go. Done. So
you put all this together and um, it paints this picture.
It's the A c l U among other people have
decried as just what are you doing here? Basically is

(11:29):
what the A c l U says, right, Um, what
what you have is a picture of a person who
is undergoing stress, maybe feeling embarrassment, is maybe just scared
to be there. Uh, maybe doesn't like having things wrapped
around his or her chest. Um, maybe doesn't really like

(11:51):
the person asking the questions. The results of these these
um changes in pattern, the data is totally subjective, that's right,
which makes polygraphs totally subjective, which takes it in large
part out of the realm of science. Yeah, voodoo science
is what they call it. And uh. Although proponents will

(12:13):
say that a well trained forensic psychophysiologists, which is the
examiner can get through all that to still get a
good result. They're like, yeah, they know all this stuff,
and if you're good, then you can factor that in
and still get a good result. So let's talk about
what the forensic psychophysiologist does. There's apparently I've seen anywhere

(12:35):
between five thousand and ten thousand of them in the
US at any given time, UM, And some of them
belong to professional organizations. I think probably maybe half or third,
depending on where you are in the on that estimate.
UM belong to any number of professional organizations. Some have

(12:57):
no accreditation whatsoever, UM, but are still able to open
up shops depending on the state there in. Some states
have zero laws about being a forensic psychophysiologist a k
A polygraph examiner. That's right, but there is also some
there are programs out there. Uh. The who wrote this article,
Kevin bondser I think so. He he um interviewed a

(13:21):
guy who founded the Accodon Academy. Exciton is a manufacturer polygraphs,
and they founded this academy as well, where um, you
go through a certain amount of training to become a
forensic psychophysiologist. And he actually interviewed that guy. Yeah, his
name's Bobb Lee Lee uh and Lee says that if

(13:42):
you come to their academy, UM you have to have
a baccalaurea degree bachelor's right, or you have to have
at least five years investigative experience and an associate's degree. UM.
You have to take a ten week course, and after
you complete the ten week course, you have to carry
out twenty five polygraph examinations and submit them to be reviewed. Uh.

(14:07):
So these are like real life ones. I guess you're
working with your local police department or whatever. Maybe you're
already a cop UM and you have to submit it
to the Accident Academy board for a review. And then
once they're all reviewed and everybody's all thumbs up, you
are a licensed I guess. But you're not licensed because
there's no licensing body. UM. You are you graduate, I

(14:31):
guess is is what they call it. So that's as
accredited as it gets. I guess. And like you said, UM.
Proponents of polygraph testing say that there, if you're a
good FP, you're gonna be able to structure everything correctly
so that you can see past somebody who's just who

(14:51):
sweats a lot like you, or who gets stressed out
easily like me, Um, and design your questions appropriately, and
you're gonna able to to figure out whether this person
is deceptive or not. So how how would you do that? Chuck? Well,
we should talk about the test itself. I guess. Uh,
you're gonna you're gonna go in and you're gonna get

(15:13):
a pretest before you get strapped up to anything. Could
take about an hour. This is just you and those
the only two people in the room. You're not surrounded
by folks like in the movies and stuff, although in
the movie sometimes it's just two people, I guess. But
the pretest that you're just gonna get an interview basically
about basically about why you're being investigated. Uh, they're also

(15:35):
gonna be profiling you and checking you out and just
seeing what kind of questions you respond to and what
might make you nervous, just so they'll be better informed
about how to properly question you once you're all strapped in, right,
and the pretest where you're just kind of hanging out
with them casually. The examiners also kind of getting info
out of you that you might not be aware of,

(15:55):
like um, if you are if you talk leisurely about
your favorite beer at one point and how you like
it a lot, and then later on it also comes
up that you have to drive a lot um. They
might come up with it. They might use that for
a control question UM, which could be something like have
you ever driven under the influence of alcohol? And a

(16:17):
control question is something where you would have to admit
guilt um and you may not want to, but it's
such a broad question that just about anybody is guilty
of it, Like have you ever lied to somebody? Have
you ever um stolen anything? That kind of thing to
where if you say no, they now have a baseline
for what it looks like when you lie. That they

(16:40):
can make a reasonable assumption that you have just lied.
And any of the data UM captured on the polygraph
they're going to use to analyze everything else off of.
That's pretty much it. That's the test, and afterward you
have the post tests where they look at all the
data and chart out whether or not they think you're

(17:01):
deceptive and aware. Like on this question, you're deceptive. On
this question you may have been deceptive. It's kind of
hard to tell. On this one you definitely were deceptive,
so so, and it's all in it's all in relation
to that control question, that baseline. Right, So if you
if your deception, if you if on questions where I mean,
they're gonna have to talk to the police as well

(17:23):
too and say what do you want to know out
of this person? So they'll design questions around that as well. UM.
So they may have a question like, um, are you
wearing a blue shirt? That may be question one, it's irrelevant, right.
Question two is, um, have you ever lied to your boss?
That's the control question, and then question three is something
like you know, did you steal the cookie from the

(17:44):
cookie jar? Like, that's the one that the cops want
them to ask. They'll compare the results of Q three
against Q two and if they're the same or you
can't really tell, that's in uh in that that's an
uh inconclusive tests. So that's it. I mean, like you said,
that's that's polygraph. It's pretty easy. It is. It's uh

(18:08):
it's um jarringly easy, considering that it's used in legal
cases a lot, right, Yes, that's true. Um. People try
to to battle the LDE detector in various ways or
a little tricks that the Internet says works, like taking
a sedative or putting any perspirint on your fingers, which
seems like they would make you wash your hands. Uh,

(18:31):
putting attack in your shoe, and anytime you get asked
a question, every single time you stomp on the tack.
And the idea is that you're just gonna skew the
test so they all look the same, so your body
has the same reaction no matter what's going on, like
they I guess if you press on the tack, your
physiological response could overwhelm any response to the question. Right. Um,

(18:54):
Like I said there, these these things are used in
legal cases, but with caveats. Right, If you undergo a polygraph,
whether you fail or pass, it doesn't really matter legally speaking, um,
because of unless you're in New Mexico. Yeah, this is
the only state that allows it just openly. Like if

(19:16):
you take a polygraph, like, it's admissible in court every
other state. Um. Usually the both sides have to agree
on it being admitted or um. The judge has to say, yeah,
we're gonna admit this one, right. Yeah, And federally the
judge decides whether or not they're going to admit it, right,
And I guess state judges kind of follow that federal

(19:37):
ruling of polygraphing. Yeah, and it's sort of a crapshoot
if a federal judge is gonna allow it or not.
There's no precedent really to where they say we have
to or we don't have to. Right, So, where are
the problems with this um? The problems with a polygraph

(19:57):
um or that it's subjective, Right, that's a one. But
also because there because it's subjective, you can get what
are called false positives and false negatives. Yeah, and you
don't want that because then the test itself is just
not valid. Uh, But I mean that a lot of
people use that as evidence that polygraph polygraphy shouldn't be

(20:20):
done at all, that it's not valid. False positive and
polygraph thing is when you find somebody who is deemed
deceptive but was telling the truth. False negative is when
somebody who wasn't telling the truth um is deemed truthful.
Like Gary Ridgeway, the Green River killer. They had him

(20:41):
for a little while and gave him a polygraph and
he passed and they let him go and he went
and killed a bunch more women. All right, that's right.
I didn't know that actually, Yeah, Well good for him.
There's also the uh, you know, the federal government is
the largest consumer of these exams, and if you work
for the federal govern right, you've probably had one to

(21:01):
get the job. But you can't do that in the
private sector thanks to the Polygraph Protection Act Employee Polygraph
Protection Act in the late eighties, they said, you can't
force your employees to do this. You can request it,
but if they don't want to do it, you can't
fire them because of it. You just can't do it right,
not in private land, right unless you have a contract

(21:24):
with the government, and then that's not valid. But yeah,
the the federal government is the largest opponent to them
in court, but also the largest consumer. I imagine that
UM and there's been a lot of cases that shaped
its admissibility or not. But the polygraph it seems like
it's kind of honest way out. I wrote an article

(21:44):
about um M R I being used as lie detectors,
and that's starting to kind of come into fashion. The
more we start to understand, like how lies are born
in the brain, being able to see it and saying
this is the pattern that will happen, um it if
this person is lying, and then that pattern happens. They say,
what we know, you're lying. We just saw that live
form in your brain. That makes sense. Yes, But at

(22:07):
the same time, people who understand Mr. Eyes say, it
is way too early to be doing that. And even
if we can do it with accuracy, there are a
lot of moral and ethical questions to it as well
that we need to address. First always UH and then
penile plus demography. What's that? So? Remember the the UM

(22:30):
newmegraphs that go around the chest in the abdomen. Imagine
one of those that goes around the penis and it
does the same thing if the text changes in traction
and girth. It's a perfect way to put it UM,
and it's used to detect arousal. They use it for
UM sex offenders. It's under at least as much attack

(22:51):
as regular UM polygraphs. But I wrote this blog post
called UM using science to root out late in homoset
quality among homophobes. A study at U g A used
UH penile plesim plethismography UM to find if anyone who
they had deemed homophobic became aroused when exposed to homosexual pornography. Yeah,

(23:18):
it's it's one of the better posts I've ever written cheese.
All right, that's our future. I guess penole plethismography for everyone, um,
everyone with a penis at least and then chuck. Lastly,
I want to encourage everybody, if this has piqued your
interest about lie detectors, to go watch the Shoe Court

(23:39):
shoe store job interview clip from Mr Show on YouTube.
You remember that one. That's very good bather friend Paula
Thompkins and he has a breakthrough. See that one. Yeah,
good old p f T. And that's it for lie detectors, right, Yeah,
I want to take a test. If there's someone in
the Atlanta area that administers these and would be willing

(24:00):
to give me a lie detector test, I would love
to do that, Okay, and I'll watch yes, um as
long as I can, you know, approve the questions or
not approve them. But I don't want to be like
rooted out is some miscreant. It's a little late for that.
Just keep it above bared. Um. If you want to

(24:22):
know more about lie detectors and play with some lie
detector flash animation, you can do that by typing in
lie detector on this in the search bar on how
stuff works dot Com. And um, that means it's time
for listener. Now that's right, Josh, this is from Brad
and Brad. If you remember, we had a list of

(24:42):
suggestions from a listener not too long ago that thought
our podcast could be a lot better if we changed
a few things. Brad has some suggestions of his own
of how we can make the podcast better. We should
both have nicknames zazz up the actual name like welcome
to stuff you should know with j C and the
Ingo sit back while getting a big helping of knowledge

(25:04):
from CHUCKO and the Duck. I second the suggestion to
remove the personal anecdotes should be moved to a separate
podcast called the Josh and Chuck Memoirs. Daily one hour
podcast can recount your lives from birth to present, focusing
on depressing stories that are marginally factual. In development is Chuck,

(25:24):
please your raise your voice one octave, Josh, lower yours
one octave? What okay? So now this is how I talk.
The opening of the podcast should be a description of
what each of you eight that day and the number
of trips to the bathroom. This allows the listener to
keep track at home, hedgehogs, brain surgeons, arcades, and Bolivian

(25:45):
politics are underrated, under sorry, underrepresented on your podcast at
least should be about these subjects. Do not exclude listener mail. Instead,
create a quieter audio track reading the listener mail and
overlay it on the rest of the podcast. That way,
listeners can hear both the mail and the main content
at the same time. This is a pretty good idea.

(26:08):
Why not set the pod we literally drive people insane?
If we did, then why not set the podcast to
a backdrop of tribal drums and jungle animal noises. Would
give it an exotic feel. That's over the listener mail track,
over the whole thing. So that would be three tracks deep. Yes,
and it would lead to suspense for the listener to
wonder if you'll be eaten by jaguars uh. And it

(26:31):
was clear from the podcast on mummies neither of you
had ever been mummified. Please refrain from explaining topics that
you don't have personal experience with. And then the final suggestion,
just retail episodes of this American line that went that
last one went down like the Staplerment top family. So
that's Brad, Thanks brand. Those are all great ideas, three

(26:52):
tracks all in one, streaming at once together listener mail
quietly the podcast and tribal drumming and jungle noises. Yes, um,
let's see. If you have access to a polygraph and
want to hook chuck up to it, let us know.
You can let us know on Facebook at Facebook dot com,
slash stuff you should know. You can tweet to us

(27:13):
at s y s K podcast, and you can send
us a regular old email at Stuff Podcast at how
stuff works dot com. Be sure to check out our
new video podcast, Stuff from the Future. Join how Stuff
Work staff as we explore the most promising and perplexing

(27:34):
possibilities of tomorrow, brought to you by the reinvented two
thousand twelve camera. It's ready, are you