phrack/issue30/9.txt

                                ==Phrack Inc.==

                     Volume Three, Issue 30, File #9 of 12
                      ___________________________________
                     |                                   |
                     |   The Truth About Lie Detectors   |
                     |_______                     _______|
                             |  by Razor's Edge  |
                             |                   |
                             | November 10, 1989 |
                             |___________________|

Americans love gadgets, so it is not hard to explain the popularity of the lie
detector.  Many people believe in the validity of lie detectors because the
instruments and printouts resemble those used by doctors and others who collect
scientific data and because lie detectors are simple, convenient shortcuts to
hard complicated decisions.  Polygraphy is fast becoming an American obsession
-- an obsession, incidentally, not shared by the British or the Europeans or,
as far as we know, the Russians.

American industry's increasing dependence on the polygraph reflects an enormous
faith in the rational processes of science.  Each of us can recall a time when
our voices sounded funny as we told a lie.  Surely, if we can "hear" a lie,
science can detect one.  It comes as a disturbing shock, therefore, to learn
how fragile the polygraph's scientific foundations really are.

The roots of the lie detector, more formally known as the polygraph, go back to
the turn of the century, when infatuation with the newly discovered powers of
electricity more than once overcame common sense.  But whereas electric hair
restorers and high-voltage cancer cures have all but vanished, the polygraph
persists and even flourishes.  According to the best estimates, over one
million polygraph examinations are administered each year in the united States.
They are used in criminal investigations, during government security checks,
and increasingly by nervous employers -- particularly banks and stores.  In
certain parts of the country, a woman must pass a lie detector test before the
authorities will prosecute a rape.  In 1983 the television show Lie Detector
added the dimension of home entertainment to polygraph tests.

The National Security Agency (NSA) leads the roster of federal polygraph users;
both it and the CIA rely heavily on polygraph testing for pre-employment and
routine security screening.  The NSA reported giving nearly 10,000 tests in
1982 (CIA numbers are classified).  Those who are labeled "deceptive" often
lose their jobs, even if there is no actual evidence against them.  Moreover,
the polygraph report may become a permanent part of an employee's records, and
it will be extremely difficult to compel a correction.

With the arrest in June 1985 of four Navy men on espionage charges, the issue
of using polygraphs to uncover spies or ferret out dishonest job seekers has
come to the forefront of the debate about what should be done to stem the loss
of defense and company secrets and to dispel potential thieves in the
workplace.

Much the same issue is at the heart of the protracted wrangle between the
Reagan Administration and Congress over plans for expanded government use of
the polygraph.  An executive order issued on March 11, 1983, known as National
Security Decision Directive 84, would have sanctioned for the first time
"adverse consequences" for a federal employee who refuses to take a test when
asked.  The directive authorized tests to investigate candidates for certain
security clearances and to ask any federal employee about leaks of classified
information.  (This directive was issued shortly after Reagan's comment about
being "up to my keister" in press leads.)  Almost simultaneously the Department
of Defense (DOD) released a draft regulation that authorized use of the
polygraph to screen employees who take on sensitive intelligence assignments;
it, too, prescribed adverse consequences for refusal.

Critics of the polygraph maintain that its use represents an invasion of
privacy, especially when the coercive power of the government or an employer is
behind the application.  It is hard for a job applicant to say no when a
prospective employer asks him or her to take a polygraph test; once hooked up
to the machine, the applicant may face questions not only about past criminal
activity but also about matters that an employer may have no business intruding
upon, such as sexual practices or gambling -- questions asked ostensibly to
assess the applicant's "character."  As a result of such abuses, nineteen
states and the District of Columbia have made it illegal for an organization to
ask its employees to take polygraph examinations.

A question more basic than whether the polygraph is an unacceptable invasion of
privacy is, of course, whether it works.  Seeking an answer in the scientific
literature can be a bewildering experience.  A report by the Office of
Technology Assessment (OTA), commissioned in 1983 by Brooks's Committee on
Government Operations, summed up the problem by citing twenty-four studies that
found correct detection of guilt ranging from 35% to 100%.

Polygraph theory thrives on a sort of Pinocchio vision of lying, in which
physiological reactions -- changes in blood pressure or rate of breathing or
sweating of the palms -- elicited by a set of questions will reliably betray
falsehood.  Lying, goes the rationale, is deliberate, and the knowledge and
effort associated with it will make a person upset enough to display a physical
reaction like a speedup of the heartbeat.  The variables measured usually
include the galvanic skin response (GSR), blood pressure, abdominal
respiration, and thoracic respiration.  The GSR is measured by fingertip
electrodes that produce changes in the electrical resistance in the palms when
they are sweating.  The blood pressure and pulse are monitored through a system
that uses a sphygmomanometer cuff, which is usually attached to the biceps
(this is similar to the way doctors measure blood pressure).  There is no
"specific lie response."  The polygraph merely records general emotional
arousal.  It does not distinguish anxiety or indignation from guilt.  The real
"lie detector" is the operator, who interprets the various body responses on
the machine's output.

Polygraphers claim that it is the form and mix of questions that are the keys
to their success.  The standard format, known as the Control Question Test,
involves interspersing "relevant" questions with "control" questions.  Relevant
questions relate directly to the critical matter:  "Did you participate in the
robbery of the First National Bank on September 11, 1981?"  Control questions,
on the other hand, are less precise:  "In the last twenty years, have you ever
taken something that did not belong to you?"

In the pretest interview, the polygrapher reviews all the questions and frames
the control questions to produce "no" answers.  It is in this crucial pretest
phase that the polygrapher's deception comes into play, for he wants the
innocent subject to dissemble while answering the control questions during the
actual test.

The assumption underlying the Control Question Test is that the truthful
subject will display a stronger physiological reaction to the control
questions, whereas a deceptive subject will react more strongly to the relevant
questions.  That is the heart of it.  Modern lie detection relies on nothing
more than subtle psychological techniques, crude physiological indicators, and
skilled questioning and interpretation of the results.

Critics claim that polygraphy fails to take the complexities of lying into
account.  For some people lying can be satisfying, fulfilling, exciting, and
even humorous, depending on their reasons for lying.  Other people feel little
or no emotion when lying.  Still others believe their lies and think they are
telling the truth when they are not.  Moreover, the theory holds that deception
produces distinctive physiological changes that characterize lying and only
lying.  This notion has no empirical support.  Quite the contrary:  Lying
produces no known distinctive pattern of physiological activity.

Undeniably, when being dishonest, people can feel great turmoil and a polygraph
can measure this turmoil.  But when apprehensive about being interrogated, they
can give a similar emotional reaction:  When they think they are losing the
chance for job openings or their jobs are on the line, when they reflect on the
judgements that could be made about their answers, or, for that matter, when
they are angry, puzzled, or even amused by the impertinent probing of a total
stranger.  Some control questions may make a person appear guilty.  Such
questions may force a subject into a minor lie or ask about an invented crime
that nonetheless makes the subject nervous.

Lie detectors are especially unreliable for truthful people.  Many more
innocent people test as "deceptive" than guilty people test as "innocent."
Those who run a special risk include people who get upset if someone accuses
them of something they didn't do, people with short tempers, people who tend to
feel guilty anyway, and people not accustomed to having their word questioned.
All of these feelings can change heart rate, breathing, and perspiration and
their heightened feelings are easily confused with guilt.

It has also been shown that polygraphs are easily manipulated.  Four hundred
milligrams of the tranquilizer meprobamate taken an hour or two before a
polygraph session can make it virtually impossible to spot a liar by his
physiological responses.  In fact, some researchers even argue that an examinee
can use simple countermeasures, such as biting one's tongue, gouging oneself
with a fingernail, or stepping on a nail concealed in a shoe, to fake a strong
reaction to the control questions, thus "beating" the test.  According to one
researcher, one prison inmate, who became the jail-house polygraph expert after
studying the literature, trained twenty-seven fellow inmates in the seat
techniques; twenty-three beat the polygraph tests used tons investigate
violations of prison rules.  However, do not try sighing, coughing, or
clenching your fist or arm.  Polygraphers usually are suspicious of those
techniques and may label you "deceptive" for that reason alone.

It should be obvious that the interpretation of the results of any polygraph
test will certainly be very difficult.  Also, not all responses on the machine
will agree.  What are the present qualifications for a polygrapher?  Most of
the twenty-five or more schools that train examiners provide only an eight-week
course of instruction and require two years of college for admission.  This is
about one-sixth the study time of the average barber college.  Perhaps as many
as a dozendy time of contemporary polygraphers do hold Ph.D's, but the vast
majority of the 4,000 to 8,000 practicing examiners had no simple significant
training in physiology or in psychology, even though lie detection demands
extremely subtle and difficult psychophysiological interpretations.  There are
no licensing standards for polygraph operators, and, with so many poorly, who
trained operators, thousands of tests are conducted hastily and haphazardly,
resulting in highly questionable accuracy.  For many innocent people, their
judge and jury are these unskilled operators.

Honesty is also difficult to predict because it tends to be situation-
specific.  Therefore, it is more dependent on motivation and opportunity than
on some personality trait.  As Bertrand Russell once said, "Virtue is dictated
by results of circumstance."

Proponents of the polygraph sometimes cite "correct guilty detections":  The
percentage of guilty subjects who are caught by the polygraph.  This figure can
be very impressive:  In one study that does not suffer from the failings
already mentioned, it was 98% correct.  But the same study found that 55% of
innocent subjects were also diagnosed as "deceptive."  The handful of studies
that used a truly random selection of cases and scored them blind produced
similar results:  Overall, 83% of guilty subjects were diagnosed as
"deceptive," as were 43% of innocent subjects.  It's no trick to push the rate
of correct guilty detections to 100% -- just call everyone "deceptive."  You
don't even need a machine to do that!

Nature published its conclusions last year.  Their aggregated findings were
based on the polygraph charts of 207 criminal suspects, which 14 polygraphers
scored independently.  On the average, they erroneously diagnosed 43% of
innocent suspects as deceptive. Such errors, called false positives, ranged as
high as 50%.  The corresponding errors of deceptive persons "passing the test,"
or false negatives, were as high as 36%.

The accuracy rates of "failed" and "passed" depend, of course, on the
proportion of dishonest persons in the group tested.  Thus, if 800 of 1,000
persons tested are truthful, a test that is 72% accurate overall will accuse
144 liars and 224 truthful persons.  This is not an impressive accuracy record.

These numbers suggest that the polygraph test is biased against innocent
people.  The problem is accentuated when the test is used in the screening
situations envisioned in the Reagan Administration proposals (and already
established at the NSA and the CIA).  Everyone is tested, but presumably only a
very small proportion has done anything wrong.  If we assume that one employee
in a hundred is a spy (probably a gross overestimate), and if we use the 83%
correct-guilty-detection rate, we find that 51 innocent persons will flunk the
polygraph test for every real spy who flunks.  Any test, whether it is for
truth or for cancer, has to be extremely accurate to detect a rare phenomenon
without setting off a lot of false alarms in the process.  Even if the test
were 99% accurate for both guilty and innocent detections, one innocent person
would be falsely branded for each spy caught.  Because of this "case rate"
problem, the FBI forbids the use of polygraph dragnets:  The tests can be used
only after an initial investigation has narrowed the field of suspects.

Given all the doubts about their validity, why does the government persist in
using polygraph tests?  Some clues are found in the DOD 1983 report on
polygraph testing -- even in its title, "The Accuracy and Utility of Polygraph
Testing" which suggests that accuracy and utility are two different things.
The most that report concludes about accuracy is that it is "significantly
above chance."  Utility, however, is quite another matter.  Perhaps the most
telling statement about lie detectors comes from former president Nixon, who
declared on one of the White House tapes, "I don't know anything about lie
detectors other than they scare the hell out of people."
_______________________________________________________________________________