Confessions of a Lie Detector: years of theft, sex, and murder

by Jim Wygant

  I also want to discover any factors that might influence the wording I choose for the test questions. If the client tells me he has done something similar to what he is accused of, or that he did part of what he is accused of, I may need to allow for those events in the questions.

  One of the most important functions of allowing the client to talk is something that probably seems only incidental to most clients. When someone has an opportunity to tell me whatever he wants me to know, even if he is lying, he becomes less nervous. Although the test is not based on how nervous someone happens to be, excessive nervousness can make charts more difficult to read. I generally do all that is possible to minimize nervousness, with the aim of making the client more comfortable and also making my own job easier.

  –––

  The next phase of the procedure is preparation of the actual questions. In the “comparison question” test now used by most examiners, all questions are discussed carefully before the test is begun. The total number of questions will typically range from eight to twelve, with ten being one of the most common formats.

  The number of questions is limited by the discomfort caused by the blood pressure cuff. It is a standard medical cuff that is usually wrapped around an upper arm and inflated with air pressure. I need about four minutes to ask ten questions with the usual silent gaps I leave between each question. Four minutes is about as long as many people can tolerate having a blood pressure cuff squeeze their arm, even at the relatively low pressure used in a polygraph test.

  Not all of the test questions ask about the issue. In fact, the typical ten-question format usually provides for no more than three questions that are directly and specifically about the issue. Those three may be very similar. Most examiners prefer as much redundancy among the issue questions as possible, so two of the issue questions or even all three may make essentially the same inquiry but in slightly different language.

  Other questions include some that are general, but similar to the issue. If the issue is a theft, there will be some general questions about theft. If the issue is sex, there will be some general questions about that. There are also some questions that are neutral in content, in the sense that they do not ask about anything related to the issue or any other accusation, and are likely to be answered truthfully. They may be something like “Is your birthday in [month]?” and “Are you now in the United States?” There is usually a question that asks if the client intends to answer truthfully. In my practice, I also ask whether the person has done anything to try to produce false test results.

  Many examiners include a question that asks if the client is concerned about some other issue or about being asked a question that was not discussed. The purpose is to confirm that no unrelated issue is distracting the client from the questions in the test. I used a common version of that type of question early in my career, but I found that its awkward wording often created more concern in an examinee than it eliminated. Clients who had not even considered that I might inquire into other areas began to worry about it when I presented that possibility as a test question.

  All of test questions are discussed in the exact wording that will be used in the actual examination. I typically go through all of the questions at least twice before starting the test. If a client expresses difficulty with a question or hesitates in offering an answer, we will discuss it further or change it to some form that the client says he can answer. Only when the client has agreed to the question wording will we begin the test. At that point we are well beyond the halfway mark in the ninety minutes the client spends with me, and we have arrived at the event for which everything else we have done has been mere preparation.

  –––

  The client is seated next to a desk or table or in a special chair that provides some kind of arm rest. The intent is to brace the body and the arm on which the blood pressure cuff will be placed. Many examiners use a specially designed chair with extended arms, often compared by clients to the electric chair, which is more a reflection of the client’s apprehension than the true appearance of the furniture.

  Two small metal plates will be placed on the fingertips of one hand, fastened by short lengths of adjustable straps. These electrodes will record variations in electrical resistance or conductance, depending on the instrument.

  Two expandable rubber tubes will be fastened around the chest, one on the upper portion and the other across the abdomen. Two are used to monitor respiration because people vary in where they produce the clearest responses. Only one of the two separate respiration tracings, whichever appears clearest, will actually be scored when the charts are evaluated.

  Those are the basic attachments. Some instruments also provide for alternate monitors of cardio activity, such as thumb sensors. Many examiners also have movement detectors that are either built into a chair or placed underneath the legs.

  In the 1990s the first computerized polygraph instruments appeared. In a remarkable technological advancement, the bulky traditional instrument was almost completely replaced by a laptop computer that analyzed the data and offered its own opinion of truthfulness. Connected to the computer was a small interface box that converted to a digital form all of the analog signals picked up from of the same tubes and cuffs and electrodes customarily used in polygraph testing. Computerized instruments were initially more than three times as expensive as a traditional polygraph, but prices quickly dropped and computerized instruments rapidly replaced the paper and ink variety, which are now rarely seen outside of movies or TV shows.

  With all attachments to the client in place, the blood pressure cuff is inflated with air to tighten it. The pressure in the cuff is likely to be about 65 millimeters of mercury, about a third of the pressure at which a doctor would begin her measure of blood pressure. The cuff remains inflated at a fixed pressure while all of the test questions are recited by the examiner, which generally takes about four minutes.

  The examiner indicates by some means on the chart where he began reading each question, where he finished, and where the client answered. In computerized instruments those marks are inserted by pressing keys on a computer keyboard. The examiner also identifies each question on a chart by number. The client’s answer is usually designated with a plus or minus sign to indicate “yes” or “no.” After each answer, the examiner waits about fifteen seconds before asking the next question. When he has read all of the questions once each, he deflates the blood pressure cuff and allows the client a brief rest between charts. The attachments are not removed, but the release of pressure from the cuff does permit circulation to recover in the arm. After a minute or so, the examiner inflates the cuff again and repeats the process of asking all of the questions, usually in a different sequence than the previous chart.

  Three charts – three times through the questions – comprise most polygraph tests. Some tests require more charts to produce clear results. A few tests are completed with two charts. There is no such beast as a valid one-chart polygraph examination. TV programs that show a single chart have generally completed a full series of charts off-camera, although those entertainment tests have many other faults and are not representative of what occurs in the more serious environment of criminal case testing.

  Repetition of the questions exposes patterns in the examinee’s physiological responses. Polygraph results are not based upon a one-time response to any particular question. In most tests, the client will establish a pattern of physiological responses that will produce definite results. A definite result is a pattern that is sufficiently clear to permit the examiner to reach one of two conclusions. The examiner can conclude that the client’s body responded in a way that is typical of people lying, or that the responses matched those of most people who are telling the truth. The only other possible polygraph result is an inconclusive one, which generally arises from inconsistent responses, a lack of a clear pattern. The inconclusive category is a kind of safety factor in the test procedure, and it will account for anywhere from fiv
e to twenty per cent of all test results, depending on the examiner.

  Most examiners evaluate their charts immediately, while the client is still present. If the test indicates lying, the examiner wants to offer the client a chance to explain – or confess.

  The chart evaluation process bears little resemblance to what most people have seen in television dramas or movies. There was a time when examiners simply looked at their charts, pulling six to eight feet of paper through their fingers and trying to maintain a mental record of what they were seeing. When I first began giving tests I knew several examiners who had been trained before the benefits of chart scoring had been thoroughly demonstrated. One popular method of evaluating charts was to tack them to the wall and analyze them from the opposite side of the room in order to see “the big picture.”

  Today most examiners who test people on specific issues, such as sex offenses or thefts, use some method of numerical scoring. It is a simple procedure with obvious benefits. It was once resisted only by examiners who did not want to surrender their reliance upon personal observations of client behavior. Numerical scoring forces the examiner to build his conclusion from his charts. It imposes a greater degree of objectivity than found in decisions influenced by an examiner’s regard for the client’s conduct, spoken responses, and attitude. And because what’s on paper can be reviewed by other examiners, numerical scoring encourages a greater degree of standardization than is otherwise possible.

  Numerical scoring was originally developed in the 1960s by examiner and teacher Cleve Backster as a training aid for his students. The federal government adopted the same system at its own polygraph school, and some form of scoring is now included as a routine part of most examiners’ education.

  There is nothing particularly brilliant about scoring, and it is not some magical solution to unclear charts. It is simply a note keeping system, so that the examiner accumulates a record of what he is seeing on the charts as he methodically analyzes responses on individual questions and pairs of questions. By the time he has gotten to the last question on the last chart, he still has his record of what he observed at the beginning of the first chart. In a typical test with three issue questions, the examiner will probably need to devote his closest attention to comparisons among six questions, analyzing respiration, skin resistance, and cardio function for each of those six. The questions and answers are commonly recorded three separate times (three charts), with separate evaluation required for each chart. There is simply too much data to analyze without some system that permits the examiner to record his observations as he makes them.

  Most examiners use a system that assigns values that range between plus and minus two or three, including zero. The scores are accumulated and are regarded as conclusive indications of truthfulness or lying when they exceed certain minimum levels. Another examiner looking at the same charts, recognizing the same kinds of common physiological reactions, should arrive at the same conclusion. This is true even if the second examiner uses a different scoring system, provided that he is consistent in applying his system and has cutoffs for conclusive results that realistically correspond to whatever range of numbers he uses when scoring.

  This seems like such an obvious approach to evaluating polygraph charts that I can only wonder why someone did not “discover” it much sooner.

  –––

  Polygraph results technically fall into three categories: truth, lie, and inconclusive. “Truth” means no more than a match to the kinds of physiological responses that most people produce when telling the truth. “Lie” means that physiological responses matched what most people produce when lying. We are really asserting a probability rather than a certainty. While many tests reveal strong, consistent responses that leave little doubt about the results, many also contain inconsistent responses. Those are often the tests that examiners agonize over, trying to decide whether the charts are strong enough to support a determination or so weak and inconsistent that the results must be called inconclusive. Polygraph examiners fear mistakes, not only for the harm they might do to suspects or victims, but for the sake of the examiner’s own credibility. Even police examiners, who are often mistakenly regarded by defense attorneys as prejudiced against suspects, desire foremost to be accurate in their work.

  –––

  I have described a “typical” examination. What I have failed to convey is what is occurring during that process in the mind of the client. I know that the majority of the people I see are afraid, just as they would be if presenting themselves for an important medical examination. Most people who take polygraph tests have never taken one before. They do not know what to expect. Unfamiliar procedures tend to make any of us uncomfortable. Besides that, I am a stranger to nearly everyone I test – there is no polygraph equivalent to the family doctor.

  Among the clients who are telling the truth, some must fear that perhaps no one will ever believe them. All of the people I see in my work have already been disbelieved by someone – usually at least the police. The innocent know the nightmare of being unable to convince others of their truthfulness. Some of them must worry that I, too, will turn against them.

  The liars suffer a different fear, that of humiliation and punishment. When I explain to a client in simple, direct terms that his polygraph test indicated he was lying, I often see defeat sweep over him. His shoulders may slump, he may become quiet and reflective. He may tell me that he thinks he “might have done something like” what he is accused of. When he leaves he will probably thank me, a response which does not fail to puzzle me. Some even apologize for the trouble they think they might have caused me.

  It is a sad business, dealing with guilty people. I often regret their mistakes, wishing they’d had more sense when they did whatever it was that brought their current trouble. With many of them, I regret their inability to see what damage they have done to others and to themselves; and what additional damage they are likely to do in the future. A few of the liars I just plain dislike, but for most I simply feel some measure of regret. Even when they don’t reveal a sense of guilt, they often appear virtually crippled with the misery of their failure and dread of the consequences.

  Liars come in all shapes and sizes. Not all of them have the tough, street look of someone who can’t keep a job. Some are just ordinary people who had been enjoying success, as measured by such usual indicators as occupation, possessions, and personal relationships. You might encounter these people in a supermarket line or in an office. They are brought down by such elemental vices as greed, pride, revenge, and lust. Their commonality suggests that many others who have not yet been subject to the accident of discovery have committed the same misdeeds. Whatever the crime of these ordinary people, it often seems to have arisen out of their own arrogance. At some point they must have reached a decision, however carelessly taken, that they deserved to get away with something they knew was wrong. In that regard the most prestigious client is little different from the unemployed laborer whose test fee is paid by the State because he is indigent.

  None of them ever really believed they would get caught.

  3. The Great Soda Pop Hoax

  At the beginning of summer in 1993, a time of the year when soft drink companies are looking forward to a seasonal surge in sales, Pepsico was hit nationwide by a bizarre form of hysteria. Consumers started reporting syringes in their Pepsi Cola cans. A syringe is about the size of a short pencil but has a sharp needle on the end. Pepisco said it was virtually impossible for syringes to be inserted into cans during the rapid process of filling and sealing that occurred non-stop on automated production lines.

  The first complaint surfaced in a news report from the state of Washington. TV broadcasts showed pictures of the syringe with its sharp hollow needle, an implement that most of us never see outside of a medical lab. Quickly a second Washington consumer reported a similar find. The national news media picked up the story, and within only a few days there were fresh claims about syringes in Peps
i cans from across the United States. Pepsico went on the offensive, pointing out that there was as much as a six-month difference in bottling dates among suspected cans, and that the bottling plants were scattered from coast to coast. At the very least, a single disgruntled employee dropping syringes into all the suspect cans was impossible.

  Beyond that, the assembly line technology created its own doubts about the authenticity of so many complaints. In the bottling plant topless cans sped along a miniature version of an amusement park roller coaster track. The track flipped the empty cans over, theoretically dumping out any contaminant, then righted them, squirted in the soda, and pressed on a lid. All of this happened while the cans continued to race along their track without pause. Any employee trying to put anything into a can had only a few seconds to do it and risked losing fingers in the high speed operation. The possibility of putting something into a can after it had been sealed was equally remote. As anyone knows who ever opened a pop-top can, the lid itself is virtually tamper-proof. The news media did not concentrate on this side of the story until numerous consumers – probably ignorant of the bottling process – had already made their complaints.

  The Food and Drug Administration regarded initial complaints with appropriate gravity and publicized a “hot line” phone number. Consumers reporting syringes to the FDA were urged to file formal reports. Most did.

  In Oregon, where I live, three complaints surfaced in the Portland area. Two were from young couples; the third was from a young handicapped man. None of them was acquainted with the others.

  Before the hysteria had become even one week old, a man in the mid-west was arrested for filing a false report with the FDA. The news media talked of a maximum possible penalty of five years in jail and a fine of $250,000. On that same day, I was contacted by a Portland television channel with a request that I test one of the Oregon complainants. She had agreed to a polygraph test.