How to detect lies with a machine?

In the early 1980s at least a million people every year in the USA were subjected to lie-detector tests – most of them applicants for jobs. However, the tests resulted in some people being falsely accused of dishonesty. One such victim was a college student, Shama Holleman, who was dismissed by a New York department store after a test indicated that she might be a drug dealer and might have served a prison sentence. Both were untrue.

Since then, US companies have been prohibited from using lie-detector tests to screen employees. However, government agencies and some security services and drugs firms are still allowed to use them.

Other major users of lie detectors are police forces, who tests suspects and witnesses in criminal cases, through not in Australia where their use is illegal. More than 60 years after the invention of the lie detector its use remains controversial.

The lie detector works on the assumptions that people who tell a lie become stressed emotionally, and the stress speeds up their pulse and breathing rate, and makes them sweat. These effects can be detected by sensitive instruments.

The first person to use instruments to detect stress through bodily changes was an Italian criminologist, Cesare Lombroso, in the 1890s. He measured variations in pulse rate and blood pressure.

In 1921 the first modern lie detector using continuous monitoring was developed, in conjunction with the local police, by a medical student, John A. Larson, at the University of California, Berkeley.

Larson’s machine recorded simultaneously a person’s blood pressure, pulse rate and a rate of breathing. The results were recorded by three pens making ink traces on a continuous roll of paper. The machine, called a polygraph, was soon dubbed a lie detector.

Later the polygraph evolved into the modern instrument with the addition of a fourth measurement – that of the skin’s ability to conduct electricity, which varies according to the amount of perspiration.

Ideally, lie-detector tests are conducted under strictly controlled conditions. The person to be tested is wired to the machine and asked a series of innocent questions (called control questions), such as: ‘Is your name John Smith?’ (which it may or may not be). This is to elicit a response from the subject that will provide reference traces on the polygraph.

Then if the person lies, the polygraph should be able to detect the changes caused by the stress of lying, and record them.

One drawback is that some people are so nervous that they may appear to lie even though they are telling the truth.

Other people may be so much in control of their emotions that they will be able to lie without affecting their polygraph traces. But this is exceptional.

 

Picture redit : Google