鈥淚鈥橠 LOVE to give you that pay rise,鈥� says your boss, 鈥渂ut we鈥檙e not in a financial position to make that happen right now. Maybe next year.鈥� Or your spouse is explaining why they have got home so late: 鈥淚t was the secretary鈥檚 leaving do, and I had to show my face.鈥� Or perhaps your children are fighting again: 鈥淗e hit me first!鈥�
When you think about it, a surprising number of our social interactions involve trying to deceive each other 鈥� and spotting if we in turn are being deceived. Working out if we鈥檙e being duped often isn鈥檛 easy. Psychologists, however, are starting to get a handle on what it takes to be a good deception detector. Some are studying people who are particularly good at spotting lies, to see if the rest of us can learn their skills. Others are investigating the circumstances under which we turn up our lie-detecting radar. 鈥淲e hope the [research] will provide information that we can use to detect deception more accurately,鈥� says Maureen O鈥橲ullivan, a psychologist at the University of San Francisco in California. 鈥淲hat we are learning is being incorporated into training programmes that the government is using.鈥�
Humans are not the only primates to practise intentional deceptive behaviour (see 鈥淣atural born liars鈥�) but with our unique intelligence and language abilities, we are the only ones to have made it such a fine art. Whether we are trying to attract a mate or gain wealth or status, lying can be an effective strategy, especially since humans are so bad at detecting deception.
Advertisement
How do scientists test lie-detecting abilities? The guru of deception research is Paul Ekman, a psychologist at the University of California, San Francisco, who has studied the subject for some 40 years. One of his standard tests is to show subjects films of people talking about their emotions as they watch a video of either gruesome close-ups of burn patients or an innocuous nature documentary, although the subjects can鈥檛 see what the speakers are viewing. All the speakers claim to be watching the nature film, but only half are telling the truth. The subjects have to decide who is lying.
A succession of studies using tests like this have shown that most of us are not very good at spotting if someone is lying. Even people whose job it is to detect deception 鈥� police officers, FBI agents, therapists, judges, customs officers, and so on 鈥� perform, on average, little better than if they had taken a guess. An as-yet unpublished review of 253 studies found that overall accuracy hovers at around 53 per cent. That鈥檚 hardly better than flipping a coin, point out the authors, Charles Bond of Texas Christian University, and Bella DePaulo of the University of California, Santa Barbara.
Law enforcers can鈥檛 even rely on the polygraph lie-detector. Although its fans claim high accuracy rates, studies have shown that innocent subjects are classed as guilty in 47 per cent of cases.
But a few people seem to be the exceptions that prove the rule. In the mid-1980s Ekman discovered someone who got 100 per cent on all his deception detection tests. J J Newberry was an agent at the US Bureau of Alcohol, Tobacco, Firearms and Explosives, based in Washington DC, and was, unsurprisingly, one of the agency鈥檚 top investigators. Newberry鈥檚 talents had not gone unrecognised, and he was responsible for training fellow agents in interview techniques.
After discovering Newberry, Ekman collaborated with O鈥橲ullivan on a mission to uncover further prodigies in the world of deception detection. In a range of studies that totalled about 14,000 people, they investigated individuals鈥� abilities to spot lies about emotions and simulated criminal activity. The team designed the experiment to ensure that the odds of passing the test by chance were less than 25 in 1 million.
鈥淭he researchers identified 29 鈥榳izards鈥� of deception detection, who are now the subject of intensive study鈥�
The researchers identified 29 鈥渨izards鈥� of deception detection, who are now the subject of intensive study by O鈥橲ullivan and Ekman. The research is still ongoing and has not yet been published. But preliminary analyses confirm some of their earlie findings: that fleeting facial expressions leaking emotions, such as anger or guilt, are key indicators of lying. The wizards seem to be able to spot these 鈥渕icro-expressions鈥�, which may last less than one-fifth of a second. 鈥淥ur wizards are attuned to detecting the nuances of facial expressions,鈥� O鈥橲ullivan says.
Ten of the 29 wizards were women, a higher proportion than would be expected, given the low fraction of females taking part in all the studies, although the researchers don鈥檛 yet have that exact figure. O鈥橲ullivan is surprised by this, as none of their previous studies had shown that women were better lie detectors. She points out it may be due to women being more conscientious about completing the tests. But she also notes that women often perform better in non-verbal communication tasks, such as gauging people鈥檚 emotions through their expressions.
Indeed, our normal reliane on verbal communication may make it harder for us to spot lies. A previous study by Ekman found superior lie-detecting ability in people with damage to the left hemisphere of the brain that had impaired their ability to understand sentences (Nature, vol 405, p 139). It seems that these people were forced to rely more on non-verbal cues, such as facial expressions.
Unusual childhoods
Another intriguing finding from the most recent project is that the wizards tended to have had difficult childhoods, in which sensitivity to the emotional temperature at home could have been useful. Some, for example, had alcoholic parents. Others had slightly unusual family backgrounds, such as parents who were immigrants, or mothers with demanding careers (at a time when this was less common). Such experiences could also have made it valuable to be extra-sensitive to non-verbal cues valuable.
Ekman and O鈥橲ullivan are not the only psychologists to take an interest in the idea of motivation for lie detection. Work by researchers at Montclair State University in New Jersey has suggested that people may be able to turn their lying radar up or down depending on the circumstances.
One of the studies, published last year, investigated women鈥檚 skills at detecting men who were pretending to have appealing attributes 鈥� sometimes called 鈥渇aking good鈥� (Personality and Individual Differences, vol 37, p 1417). An example of this kind of lie would be a man claiming he owned the Ferrari outside, rather than admitting he had borrowed it from a friend for the night.
In the Montclair team鈥檚 test of 34 female undergraduates, the single women seemed to be better at detecting men who were faking good than those who were in a committed relationship. 鈥淲omen have a kind of radar for deception in men, which they switch on or off, depending on the context,鈥� says Julian Paul Keenan, who led the research team.
Mixing your genes with a man who borrows rather than owns a Ferrari could have serious implications. As Keenan points out, pregnancy from a deceptive male could have 鈥渉uge negative consequences鈥�. On the other hand, if you have already mixed your genes or are committed to doing so, then you are less motivated to see deception.
As well as looking at what makes someone a good lie detector, Keenan鈥檚 group investigated what makes a good liar. Several previous findings have led psychologists to speculate that ability to deceive is linked with high self-awareness, or understanding of one鈥檚 feelings, thoughts and motives. Children鈥檚 ability to deceive emerges shortly after the development of self-awareness, as indicated by self-recognition, self-pronoun use, and self-conscious emotions. And people with poor self-awareness, such as schizophrenia patients, appear to be bad at lying and spotting lies.
Keenan鈥檚 latest study, which is also due to be published in Personality and Individual Differences, used a group of people who varied in their scores on a self-awareness test and got them to lie on video. The videos were played to undergraduate students, who were asked if the speakers were lying or telling the truth. The researchers found that those with higher self-awareness were better deceivers.
鈥淲omen have a kind of radar for deception in men, which they switch on or off depending on the context鈥�
The study鈥檚 findings are intriguing but it is notable how little research has been done on what makes a good liar compared with the plethora of work on deception detection. Perhaps there is a problem: who would want it known that they were being studied for their outstanding talents at lying?
Natural born liars
Deception is rife in nature. From the harmless snake that looks like a poisonous one to the orchid shaped like a female wasp to attract pollinating male wasps, all may not be what it seems. And deception extends beyond appearances. Some animals use deceptive behaviour to keep one step ahead of the competition.
Primates, with their complex social groupings, have a particularly well-developed Machiavellian streak. Their large repertoire of subterfuge includes sneaky matings, hiding food and manipulative misdirection. Such duplicity takes brains, as primatologists Andrew Byrne and Nadia Corp from the University of St Andrews in Scotland have shown. Their study reveals that the size of the cortex, the outermost layer of the brain, which is responsible for advanced cognitive functions, is a good predictor of the degree of deception to be found in a primate species. This suggests that the benefits of hoodwinking neighbours have helped favour the evolution of cognitive prowess.
Since nobody wants to be cheated, animals have also found ways to avoid being duped. When a rhesus monkey comes across some food, for example, it usually calls to the others, so all can share. But sometimes it will remain silent and eat the food itself. Psychologist Marc Hauser of Harvard University has shown that rhesus monkeys spotted discovering a food stash, but failing to publicise it, find themselves on the receiving end of aggressive behaviour from other members of the community. If the costs of getting caught outweigh the benefits of deception, would-be deceivers soon learn to curb their cheating ways.
As far as we can tell, rhesus monkeys lack 鈥渢heory of mind鈥� 鈥� the ability to understand that another individual may hold a different perspective on things. They deceive only by learning that certain tactics work in particular situations 鈥� an ability that has surprisingly ancient origins. 鈥淭hat sort of tactical deception is found in all primates, implying that learning to deceive is at least 50 million years old,鈥� Byrne says.
Only our closest animal relatives are thought to share theory of mind with us, allowing individuals to plan to deceive others intentionally. 鈥淭hat is found in all great apes, dating it to around 12 to 14 million years ago,鈥� says Byrne. That鈥檚 a long history of fibbing.
Dan Jones
To catch a liar
How do you know if someone is lying to you? Deception detection is an inexact science, but there can be some tell-tale signs, say Bella DePaulo of the University of California, Santa Barbara, and Wendy Morris of the University of Virginia in The Detection of Deception in Forensic Contexts, from Cambridge University Press.
Contrary to folklore, liars are not more fidgety, nor do they blink more or look less relaxed. Liars do tend to seem more nervous than truth-tellers, however, perhaps because their voices are pitched higher. There is also an association between lying and larger pupil size, a signal of tension and concentration.
The kind of lie determines the signals. When liars are highly motivated 鈥� when the stakes are high 鈥� they become unusually still and make notably less eye contact with listeners. When the lie is planned, deceivers start their answers more quickly than truth-tellers. If taken by surprise, however, the liar takes longer to start answering questions, and they talk less.
The content of speech can be another tip-off. Liars seem more negative than truth-tellers 鈥� more complaining and less cooperative. They also tend to withhold information, either from guilt or to make it easier to get their stories straight, and to repeat words and phrases. 鈥淟iars鈥� answers sound more discrepant and ambivalent; the structure of their stories is less logical,鈥� the researchers write.
It may even be possible to spot a liar if a written statement is all there is to go on. Content-based criteria analysis is a technique used by forensic scientists to systematically analyse witness statements to work out if they are true or fabricated. The technique involves checking the statement for 19 criteria thought to be hallmarks of truthfulness. True statements are supposed to include more superfluous details, spontaneous self-corrections and speculation about other people鈥檚 mental states. The truthful witness is also more likely to be self-deprecating and to make comments that go some way towards pardoning the alleged perpetrator.
The success rates reported for content-based criteria analysis seem to vary depending on the circumstances. But one small study of its use in testing teenagers鈥� claims of school bullying found a success rate of 95 per cent. That compared with 55 per cent success for judgements made by the school鈥檚 teachers.
