The lie machine
Cheating husbands and cunning fraudsters beware: a new brain-scanning technique apparently makes it easier to unmask a fibber. Simon Hadlington sorts out the truth
Wednesday 13 September 2006
How easy is it to spot a liar? Obviously if it's Pinocchio who's telling the whopper then there's going to be a bit of a giveaway. But in general, lying can be extraordinarily difficult to detect.
The concept of lie detection is mired in controversy, with claim and counter-claim made for a range of techniques, from the classic polygraph - now more than 80 years old - which measures stress and anxiety, to machines that track brain waves and those that analyse tiny movements of the eye.
But now a company in California reckons that it can rumble a fibber with unprecedented accuracy by using a brain scan. No Lie MRI Inc has just started offering its service on the market. Next Monday the first paying customers will be slid into a brain scanner and asked a series of questions to determine if they are telling the truth. All the customers are involved in legal proceedings and have elected to take the lie-detector test to demonstrate that their testimony to their lawyers has been truthful.
Magnetic resonance imaging (MRI) has been used for decades to produce images of the brain, usually to detect abnormalities. More recently, scientists have demonstrated that MRI can show which parts of the brain become active in response to different stimuli. This is called functional MRI, or fMRI. It shows which parts of the brain are consuming most oxygen - and hence working hardest. For different activities - speaking or using the memory, for example - different areas of the brain "light up" under an fMRI scan.
For several years research scientists have been using fMRI to map the functions of different parts of the brain. At the universities of Pennsylvania and South Carolina in the US, researchers began to investigate whether the technique could be used to detect deception. They concluded that when an individual constructs a lie, specific portions of the brain become active.
No Lie MRI is now commercialising this concept. "The MRI machine can be linked up with special computer software, such as neural networks, that can be trained to recognise specific patterns," says Joe Huizenga, the company's founder. The computer is shown hundreds of examples of fMRI scans of people who are known to be lying and gradually gets to know what to look for. "We have trained our system to the extent that we can get 93 per cent accuracy," says Huizenga. "We fully expect to increase that to 99 per cent in due course."
A person being subjected to the test is placed inside the scanning machine - a large hollow cylinder - and provided with earphones and a screen. They are presented with information and asked "yes" or "no" questions, which they answer by pressing a button. Huizenga says that there will be an upfront fee for preparing the test, plus a cost of $30 dollars (about £15) for every minute that the person is in the machine. Further costs would arise if court documents needed to be prepared. No Lie MRI does not propose to buy any machines itself. Instead it will initially lease machines from hospitals. Eventually the company intends to franchise its operation around the world. "Anyone who has spare time on their MRI machine could do these tests," says Huizenga. "No expertise is needed. The data from the images can be sent directly to us via the internet and we can do the analysis."
According to Huizenga, the company already has around 50 people waiting for tests. "These people have heard about us and have approached us directly," he says. "Mostly they are involved in court cases and need to verify that they are telling the truth. Usually it has something to do with sex or money."
So could fMRI silence the lie-detection sceptics? Dr Adrian Owen is a senior scientist at the Medical Research Council's Cognition and Brain Sciences Unit in Cambridge, and an expert on fMRI. "We have had a lot of success in using fMRI to map many of the brain's networks that are responsible for things like aspects of language and memory," he says. "Lying is a lot more sophisticated because it is likely to involve a number of different cognitive processes. It is unlikely that there is one specific region of the brain that is responsible for the act of lying, but there are other well-defined regions that are related."
In laboratory tests, Dr Owen says, certain tasks that could be involved in the process of constructing a lie do display distinct patterns on fMRI scans. "For example, we know that particular regions of the brain are involved in creativity and imagination, so it is conceivable that the process of deception involves these."
Overall, Dr Owen says, it is "perfectly reasonable" that fMRI could detect if someone is telling a lie. "I would add the caveat, however, that there is no infallible neural signature for lying, so the system will be fallible. However, I do think that fMRI should be able to do a much better job than the other methods that provide surrogate measure of lying, such as the polygraph."
Forensic psychologist Dr Paul Seager, of the University of Central Lancashire, is an expert in the psychology of deception. He remains cautious of claims of any lie-detection technology. "No method is 100 per cent accurate and I don't think any of them come close to that. Even if a method is 90 per cent accurate it means that there will be an error of one in 10. We should err on the side of caution before we start using these technologies in potentially life-changing situations," he says.
How to spot a liar
Lying is probably as old as language itself and throughout history people have sought new and foolproof ways of exposing a liar. In ancient China, someone under suspicion of a misdemeanour was required to have a mouthful of dry rice while he was being questioned. If the rice remained dry the accused was clearly too anxious to salivate and was therefore guilty. In Africa, the accused person would have to hold an egg in his hand. If it broke it betrayed his nervous tension and by implication his guilt.
The classic lie detector, the polygraph, was invented more than 80 years ago. It works by measuring a person's breathing rate, sweating on the hands, pulse and blood pressure. The measurements together provide a picture of a person's stress. Proponents of the polygraph say it can predict deception with a high degree of accuracy. Its use, however, remains highly controversial. Scientists from both the US National Academy of Sciences and the UK's British Psychological Society have warned against relying on the accuracy of the polygraph to detect guilt because, they say, it is too inaccurate.
Some people believe that the pattern of a person's brainwaves can accurately reflect if they are lying. Electrodes are placed on the skull and face and the electrical activity of the brain is recorded while the person is asked questions. Specific patterns of signal, supporters say, correlate to a lie.
Another technique is "microexpression analysis". This purports that tiny tell-tale signs in the fibber's face give the game away. This idea has been taken a step further with a system that uses infra-red light to track the movement of a person's eyes when they are being questioned. Similarly, lasers have been used to monitor the tremors of a person's carotid artery - the idea being that imperceptible movement of muscles can indicate stress.
Scientists at the University of Pennsylvania have developed a device called the Cognosensor, which consists of a headband that can beam infra-red light into the wearer's brain. The pattern of the light that is reflected indicates the degree of the flow of blood to different parts of the brain. The patterns of blood flow could indicate the relative activity of various regions of the brain, which could indicate if the person is telling a lie.
With security becoming an increasingly important issue, particularly in airports, researchers are looking into low-cost ways of remotely and rapidly monitoring people answering routine questions at check-in desks, for example. One idea is to scan people's faces with a high-definition thermal camera that detects tiny changes in the heat of the capillaries under a person's skin. A computer rapidly analyses any changes, picking up tell-tale signs that someone is fibbing.
The US Department of Defense is seeking ideas for the development of technology that can tell from a distance if someone is exhibiting signs of stress. The idea would be to beam microwaves or lasers at a person. The reflected beams would carry information about the person's physiological state, such as breathing rate, similar to the information gathered during a polygraph test. Experts point out that such a system presents serious practical difficulties, however.
- 1 What happens to your body when you give up sugar?
- 2 Tourist films plane's descent just metres above packed Caribbean beach
- 3 World Book Day: Boy 'excluded' from school after dressing up as Fifty Shades' Christian Grey
- 4 Have sex with your iPad thanks to the new sex toy no-one asked for
Harrison Ford plane crash: Actor will make full recovery thanks to 'beautifully executed' landing
Tourist films plane's descent just metres above packed Caribbean beach
Oxford whistleblower's attempt to protect young victim of a sadistic gang went unheeded
How Homer Simpson discovered the Higgs boson over a decade before scientists
The 'sex selfie stick' lets you FaceTime the inside of a vagina
Nearly 100,000 of Britain's poorest children go hungry after parents' benefits are cut
Durham Free School: 'Creationism taught at' free school facing closure
End of the licence fee: BBC to back radical overhaul of how it is funded
Elif Shafak: Turkish author warns against rise of British nationalism
Ex-head of MI6: 'We shouldn't kid ourselves that Russia is on a path to democracy'
Most people think legal tax avoidance is just as wrong as illegal tax evasion, poll suggests
£20000 - £50000 per annum: Recruitment Genius: A Physiotherapist / Sports Ther...
£8 - £9 per hour: Recruitment Genius: Sales Executives / Advisors are required...
£14000 - £15000 per annum: Recruitment Genius: An opportunity has arisen for a...
£70000 - £80000 per annum: Ashdown Group: A long-established, technology rich ...