THE BELOW OF THE SURVEY
Cathie Louvet's chronicle
CRIMINAL INVESTIGATIONS :
Folder n°9: the truth about the lie detector
1. Why a lie detector?
The miscarriage of justice is the nightmare of the police and the courts: everything rests on the statements of the accused. We swear on the Bible to say "the truth, the whole truth and nothing but the truth", when we play his head, all shots are allowed. Yes, but everything rests on the ability to lie and, above all, to make believe his lies; or on the ability to convince jurors that the truth is being told, despite deceptive appearances. So how do you separate the wheat from the chaff? No wonder, for more than a century, researchers have been hard at work inventing THE perfect lie detector, with flawless reliability.
I have the honor to present you with IRMF (functional brain magnetic resonance imaging), a large tunnel-shaped machine able to see in real-time which zone of the brain is working and therefore to read the lie in the brain of the person being interrogated ( at least we hope so very much). IRMF was not invented for the police, but for doctors and scientists conducting brain research. It is based on a sophisticated technique to visualize the most active parts of the brain whose basic principle is the following: to measure the rate of oxygen-depleted red blood cells; in fact, the more the brain cells work, the more they consume oxygen and the more the red corpuscles are locally depleted of oxygen.
This gave Daniel Langleben, a researcher at the University of Pennsylvania, a great idea: what if, thanks to IRMF, it was possible to "light up" the areas of the brains responsible for the lie? In order to achieve this result, Daniel Langleben devised the following experience.
The American researcher has selected a group of student volunteers to participate in the strange experiment conceived as follows: before going to bed in the machine, the volunteer receives an envelope containing two playing cards, a 5 of clover and a 7 of spades, and a twenty-dollar bill. In order to keep the ticket, the "guinea pigs" will have to lie about the possession of one of the two cards. Here is how: the student takes place in the machine (which, I remind you, looks like a big tube); on a screen placed above his eyes, one after the other, the images of 432 playing cards are displayed. The instruction is to systematically lie about one of the two cards he has in his envelope when he sees it appear on the screen. For this, he clicks on the "no" button using the computer mouse he holds in his hand. For other images, he must be sincere.
Analyze the data:
From this data, a computer was able to determine if it was presented to him the image of the brain of a liar or sincere student, detecting the lie of each one with 1% of error close !! Since it was the students whose brains had been used to construct the mathematical model used by the computer, this result is not really impressive.
More importantly, thanks to this model, the computer has been able to classify as "truth" or "lie" the answers of other students who do not belong to the tested group, with 88% of fair results. Same result with a group of students trained to lie shamelessly. While these results are impressive, but Daniel Langleben does not cry the miracle. Because his method has its limits. Especially as the rarity of IRMF and its exorbitant cost, we are not ready to see it settle in all police stations in the country ...
Many researchers around the world are not at all convinced by the results obtained by Daniel Langleben. And for good reason ... For 80 years, they think to have the ideal solution with the famous polygraph, another name of the detector of lies. In the US, the police, counterintelligence, insurance companies and even recruiters use it. More than 1 million people are submitted every year !!
Its operation, very simple, allows to record the intensity of four parameters: a full envelope of air placed on the abdomen of the questioned subject is used to evaluate the pressure it exerts to breathe; his blood pressure is measured by means of a blood pressure monitor placed around his arm; the electrical resistance of the skin is measured at the level of the fingers in order to evaluate whether the subject is perspiring or not; we also measure his breathing in the chest.
A sudden variation of one of its parameters is supposed to betray a strong emotion. And that's where everything goes wrong, because everything is a matter of interpretation. Indeed, how can one be 100% certain that the emotion detected is linked to a lie and not to the stress felt at being accused of a crime that has not been committed?
This is the crucial question to which the polygraph but also all lie detection systems must answer. This is why the specialists have developed questionnaires whose purpose is to distinguish among the suspects an emotion unrelated to any lie. These questionnaires fall into two categories: CQT and GKT.
CQT: Test Question Test (yes Control Question Test): During this test, a relevant question (related to the survey) is asked in the middle of a lot of uninteresting questions. For example: the question "did you kill your husband" will be drowned in a flood of questions such as "Do you have the time?", "Are you sitting on a chair?", "Do you like candy? mint? ", etc ... Attention !! A question of control, supposed to upset the interviewee, like, "Were you sad last year when your mother died of cancer?" is slipped into the CQT questionnaire. The specialists assume that the person concerned will answer them sincerely. It remains to compare the reactions of her body when she answers the question of control and when she answers the relevant question to get an idea of his frankness. The limit of the test is that one must find a control question to which the suspect will answer without lying, and ensure that the control question and the relevant question have the same emotional value for the interviewee, which is far from to be obvious !!
GKT: Guilty Knowledge Test. During this test, the investigators seek to establish the importance of an issue to the suspect. For example, if the killer fled into a red car, which only the culprit and the police know, the investigators will ask the following questions: "Was the assassin's car red?", "The car of Was the assassin blue? " "Was the assassin's car ...?" as well as other neutral issues unrelated to the investigation. It is assumed that the culprit will react more intensely to the question of the red car. The big disadvantage of this test is that it assumes that the investigator is in possession of information on the crime ignored by all but the culprit and himself, a very rare circumstance, which explains why this type of test is very little used.
No bluff in the police stations:
This is the last questionnaire that Daniel Langelben's team used to develop his experiences. Because for users of the polygraph, what matters is not the answer given but the emotion aroused by the question itself. Thus, for a team of researchers working to map the brain of a liar, it must be absolutely certain that the suspect is lying to a given question. Which is far from being obvious in real interrogation situation, in a police station.
What future for the lie detector?
As we have seen, the polygraph is not reliable enough, the fMRI too expensive and impractical: the future of the detectors of lie seems very dark ... Which does not prevent the American investigators from continuing to use the polygraph for the purpose of testing a suspect and possibly cracking him to obtain a confession, although the detection tests are not accepted as irrefutable evidence by the courts for criminal cases, whether in the US or in Canada. ..what is rather a good thing !!