The truth about lying and laughing

Over the years, I have tried to unravel the truth about deception - investigating the telltale signs that give away a liar.

There are some pointers to the evolutionary origins of deceit. A few years ago, animal researcher Maxine Morris observed some rather curious behaviour among a group of Asian elephants at Washington Park Zoo. At feeding time, each elephant was given a big bundle of hay. Morris noticed that a couple of the elephants tended to eat their own hay quickly, sidle up to their slower-eating companions, and then start swinging their trunks from side to side in a seemingly aimless way. Morris's repeated observations suggested that this apparently innocent behaviour masked a duplicitous intent. Once the trunk-swinging elephants were sufficiently close to another elephant, they would grab some of the uneaten hay, and quickly gobble it up.

Was this intentional deception? The only way to know for certain would be to discover what was going on inside an elephant's head - something some researchers believe has been achieved, not with elephants but with gorillas. In the 1970s, as part of a large-scale research programme exploring interspecies communication, Dr Francine Patterson from Stanford University attempted to teach two lowland gorillas called Michael and Koko a simplified version of American Sign Language. According to Patterson, the great apes were capable of holding meaningful conversations, and could even reflect upon profound topics, such as love and death.

Their trainers believe they uncovered instances where the two gorillas were economical with the truth. In one example, Koko broke a toy cat, and then signed to indicate that the breakage had been caused by one of her trainers. In another episode, Michael ripped a jacket belonging to a trainer and, when asked who was responsible for the incident, signed "Koko". When the trainer expressed some scepticism, Michael appeared to change his mind, and indicated that Dr Patterson was actually to blame, before finally confessing. If the elephants might be given the benefit of the doubt, the apes' linguistic skills seem to provide compelling evidence of intentional deceit.

Other researchers have explored the development of deception in children. Some of the most interesting experiments have involved asking youngsters not to take a peek at their favourite toys. During these studies, a child is led into a laboratory and asked to face one of the walls. The experimenter then explains that he is going to set up an elaborate toy a few feet behind them. After setting up the toy, the experimenter says that he has to leave the laboratory, and asks the child not to turn around and peek at the toy. The child is secretly filmed by hidden cameras for a few minutes, and then the experimenter returns and asks them whether they peeked. Almost all three-year-olds do, and then half of them lie about it to the experimenter. By the time the children have reached the age of five, all of them peek and all of them lie. The results provide compelling evidence that lying starts to emerge the moment we learn to speak. Perhaps surprisingly, when adults are shown films of their children denying that they peeked at the toy, they are unable to detect whether their offspring are lying or telling the truth.

A few years ago I carried out a national survey into lying, focusing on adults. Only 8% of respondents claimed never to have lied. Other work has invited people to keep a detailed diary of every conversation that they have, and of all of the lies that they tell, over a two-week period. The results suggest that most people tell about two important lies each day, that a third of conversations involve some form of deception, that four in five lies remain undetected, that more than 80% of people have lied to secure a job, and that more than 60% of the population have cheated on their partners at least once.

Are you a good liar? Most people think that they are, but in reality there are big differences in how well we can pull the wool over the eyes of others. There is a very simple test that can help determine your ability to lie. Using the first finger of your dominant hand, draw a capital letter Q on your forehead.

Some people draw the letter Q in such a way that they themselves can read it. That is, they place the tail of the Q on the right-hand side of their forehead. Other people draw the letter in a way that can be read by someone facing them, with the tail of the Q on the left side of their forehead. This quick test provides a rough measure of a concept known as "self-monitoring". High self-monitors tend to draw the letter Q in a way in which it could be seen by someone facing them. Low self-monitors tend to draw the letter Q in a way in which it could be read by themselves.

High self-monitors tend to be concerned with how other people see them. They are happy being the centre of attention, can easily adapt their behaviour to suit the situation in which they find themselves, and are skilled at manipulating the way in which others see them. As a result, they tend to be good at lying. In contrast, low self-monitors come across as being the "same person" in different situations. Their behaviour is guided more by their inner feelings and values, and they are less aware of their impact on those around them. They also tend to lie less in life, and so not be so skilled at deceit.

What are the telltale signs that give away a lie? Is it possible to teach people to become better lie detectors? I devised a large-scale experiment to be conducted live on the BBC science programme Tomorrow's World. Originally I suggested we ask several politicians to lie or tell the truth on the show, and have the public try to identify the lies.

Politicians were unwilling to participate, allegedly because they were terrible liars (none of us believed them). We looked for a prestigious alternative, and invited the television political interviewer Sir Robin Day to be our guinea pig.

The design of the experiment was simple. I would interview Sir Robin twice and in each interview ask him to describe his favourite film. In one interview he would say nothing but the truth, and in the other he would produce a pack of lies. We would then show both interviews on television, and invite the public to telephone in their verdicts.

As he sat down in front of the camera, Day seemed slightly nervous that he was about to receive questions rather than ask them. In the first interview, he plumped for an epic drama:

'So, Sir Robin, what's your favourite film?'

'Gone With The Wind.'

'And why's that?'

'Oh, it's ... it ... it's a classic. Great characters; great film star - Clark Gable; a great actress - Vivien Leigh. Very moving.'

'And who's your favourite character in it?'

'Oh, Gable.'

'And how many times have you seen it?'

'Um ... [pause] I think about half a dozen.'

'And when was the first time that you saw it?'

'When it first came out. I think that it was in 1939.'

Once he had finished, I repeated the questions and he described being a big fan of the Billy Wilder film Some Like It Hot:

'So, Sir Robin, what's your favourite film?'

'Ah ... [pause] er, Some Like It Hot.'

'And why do you like that?'

'Oh, because it gets funnier every time that I see it. There are all sorts of bits in it which I love. And I like them more each time that I see it.'

'Who's your favourite character in it?'

'Oh, Tony Curtis, I think. He's so pretty ... [short pause] and he's so witty, and he mimics Cary Grant so well and he's very funny the way he tries to resist being seduced by Marilyn Monroe.'

'And when was the first time that you saw it?'

'I think when it came out, and I forget when that was.'

Which is the lie?

We received more than 30,000 calls from viewers: 52% thought that Sir Robin had been lying about Gone With The Wind, and 48% voted for Some Like It Hot. We then showed viewers a short clip in which I asked Sir Robin whether he really liked Gone With The Wind. His reply was short and to the point: "Good heavens, no! It's the most crashing bore."

It could be argued that Sir Robin was a skilled liar, and that in everyday life people are much better at detecting deception. Psychologists have been exploring this question for 30 years. The research has studied the lying behaviour of salespeople, shoppers, students, drug addicts and criminals. Some of my own work in this area has involved showing people video tapes of instances in which people have made high-profile public appeals for information about a murder, only later to confess and be convicted of the crime themselves.

The results have been remarkably consistent - when it comes to lie detection, the public might as well simply toss a coin. It doesn't matter if you are male or female, young or old; very few people are able reliably to detect deception. The results suggest that we can't even tell when our partners are being economical with the truth.

We're in good company. Psychologist Paul Ekman from the University of California, San Francisco, showed video tapes of liars and truth-tellers to various groups of experts, including polygraph operators, robbery investigators, judges and psychiatrists, and asked them to try to identify the lies. All tried their best. None of the groups performed better than chance.

So why are people so bad at detecting deceit? The work of psychologists such as Professor Charles Bond from the Texas Christian University provides a clue. He has conducted surveys into the sorts of behaviour people associate with lying. He surveyed thousands of people from more than 60 countries, asking them to describe how they set about telling whether someone is lying. People's answers are remarkably consistent. From Algeria to Argentina, Germany to Ghana, Pakistan to Paraguay, almost everyone thinks liars tend to avert their gaze, nervously wave their hands around and shift about in their seats.

There is, however, one small problem. Researchers have spent hour upon hour carefully comparing films of liars and truth-tellers. On each showing, the observers look out for a particular behaviour, such as a smile, blink or hand movement.

The results are clear. Liars are just as likely as truth-tellers to look you in the eye, they don't move their hands around nervously and they don't shift about in their seats (if anything, they are a little more static than truth-tellers). People fail to detect lies because they are basing their opinions on behaviours that are not actually associated with deception.

So what are the signals that really give away a liar? It is obvious that the more information you give away, the greater the chances of some of it coming back to haunt you. As a result, liars tend to say less and provide fewer details than truth-tellers. Look back at the transcripts of the interviews with Sir Robin. His lie about Gone With The Wind contains about 40 words, whereas the truth about Some Like It Hot is nearly twice as long.

Liars often try psychologically to distance themselves from their falsehoods, and so tend to include fewer references to themselves in their stories. Again, Sir Robin's testimony provides a striking illustration of the effect. When he lies, Sir Robin uses the word "I" just twice, whereas when he tells the truth his account contains seven "I"s. In his entire interview about Gone With The Wind, Sir Robin only once mentions how the film makes him feel ("very moving"), compared with the several references to his feelings when he talks about Some Like It Hot ("it gets funnier every time I see it", "all sorts of bits I love", "[Curtis is] so pretty ... so witty").

The simple fact is that the real clues to deceit are in the words that people use, not the body language. So do people become better lie detectors when they listen to a liar, or even just read a transcript of their comments? The interviews with Sir Robin were also broadcast on radio and published in a newspaper, and although the lie-detecting abilities of the television viewers were no better than chance, the newspaper readers were correct 64% of the time, and the radio listeners scored an impressive 73% accuracy rate.

Are there no signs of deception that can be detected in people's body language and facial expressions? Not necessarily. Take one of the most common, and frequently faked, forms of non-verbal behaviour - the human smile.

We all smile, but few of us have any insight into the complex psychology underlying this seemingly simple behaviour. Do you smile because you are happy, or to let other people know that you are happy? To help settle the issue, Professors Robert Kraut and Robert Johnston from Cornell University decided to compare the amount people smiled when they were happy but alone, with when they were equally happy but with others. They hit upon an ideal place to conduct their study - a bowling alley. They realised that when bowlers rolled their ball down the lane and obtained a good score, they tended to be alone and happy. When they turned and faced their fellow bowlers, however, they would be just as happy but now interacting with others.

Kraut and his colleagues secretly observed more than 2,000 bowls. Each time, the researchers carefully documented the bowlers' facial expressions and their score.

Their findings revealed that only 4% of the bowlers smiled when they obtained a good score but were facing away from their colleagues; however, when the bowlers turned around and looked at their friends, 42% of them had a huge smile on their faces. Strong evidence that we don't smile simply because we are happy, but rather to let others know that we are happy.

Is a fake smile detectable or is it identical to a genuine smile?

At a science festival in New Zealand, I suggested to the organisers a study that would help unmask the fake smile. The idea was simple. Visitors would be shown several pairs of photographs. Each pair would contain two images of the same person smiling. One of the smiles would be genuine and the other fake, and the public would be asked to spot the genuine grins. Participants were given a questionnaire to fill in. The results revealed that many people couldn't tell the fake from the genuine smiles, and even those that thought they were especially sensitive to the emotions of others scored little better than chance.

In fact, there is a way of distinguishing fake from genuine. As early as the mid-19th century, a French scientist named Guillaume Duchenne de Boulogne set out to develop a system for investigating which muscles are involved in different facial expressions. He decided to photograph living subjects as electricity was applied directly to their faces. After much searching, he found a participant who was willing to put up with the constant, rather painful, stimulation of his face. In his 1862 book The Mechanism Of Human Facial Expression, Duchenne presents a less than flattering description of his guinea pig:

"The individual I chose as my principal subject for the experiments ... was an old toothless man, with a thin face, whose features, without being absolutely ugly, approached ordinary triviality, and whose facial expression was in perfect agreement with his inoffensive character and his restricted intelligence."

In addition, the man had another desirable attribute - almost total facial anaesthesia. This meant that Duchenne could "... stimulate his individual muscles with as much precision and accuracy as if I were working with a still irritable cadaver".

After taking hundreds of photographs, Duchenne uncovered the secret of the fake smile. When electricity was applied to the cheeks of the face, the large muscles on either side of the mouth - known as the zygomatic major - pulled the corners of the lip upwards to create a grin. Duchenne then compared this smile with the one produced when he told his thin-faced participant a joke. The genuine smiles involved not only the zygomatic major, but also the orbicularis oculi muscles that run right around each eye. In a genuine smile these muscles tighten, pulling the eyebrows down and the cheeks up, producing tiny crinkles around the corners of the eyes. Duchenne discovered that the tightening of the eye muscles lay outside of voluntary control, and was "only put into play by the sweet emotions of the soul".

In the 1970s, Monty Python's Flying Circus created a sketch based entirely around the idea of finding the world's funniest joke. Set in the 1940s, a man named Ernest Scribbler thinks of the joke, writes it down, and promptly dies laughing. The joke turns out to be so funny that it kills anyone who reads it and is turned into a weapon of war.

In 2001 I started to think about the possibility of really searching for the world's funniest joke. I knew that there would be a firm scientific underpinning for the project, because some of the world's greatest thinkers, including Freud, Plato and Aristotle, had written extensively about humour.

I got the green light from the British Association for the Advancement of Science (BAAS) for my plan for an international internet-based project called "LaughLab". I would set up a website that had two sections. In one part, people could input their favourite joke. In the second section, people could answer a few simple questions about themselves (such as their sex, age and nationality), and then rate how funny they found various jokes randomly selected from the archive. During the course of the year, we would be able to discover scientifically what makes different groups of people laugh, and which joke made the whole world smile.

The success of the project hinged on being able to persuade thousands of people worldwide to participate. To help spread the word, we launched LaughLab with an eye-catching photograph based on perhaps the most famous (and, as we would go on to prove scientifically, least funny) joke in the world: "Why did the chicken cross the road? To get to the other side."

Within a few hours of opening the website, we received more than 500 jokes and 10,000 ratings. Then we hit a major problem. Many of the jokes were rude; absolutely filthy, in fact. Other jokes cropped up again and again ("What is brown and sticky?" "A stick" was submitted more than 300 times).

Participants were asked to rate each joke on a five-point scale ranging from "not very funny" to "very funny". Initially most of the material was pretty poor, and so tended to obtain low percentages. Around 25-35% of participants found the following jokes funny, and at that point they came towards the top of the list:

Did you hear about the man who was proud when he completed a jigsaw within 30 minutes, because it said 'five to six years' on the box?

Texan: Where are you from?

Harvard graduate: I come from a place where we do not end our sentences with prepositions.

Texan: OK - where are you from, jackass?

The top jokes had one thing in common - they create a sense of superiority in the reader. The feeling arises because the person in the joke appears stupid (like the man with the jigsaw), or pricks the pomposity of another (like the Texan answering the Harvard graduate).

We were not the first to notice that people laugh when they feel superior. The theory was described by Plato in The Republic. Plato was not a fan of laughter. He thought it was wrong to laugh at the misfortune of others, and that hearty laughter involved a loss of control that resulted in people appearing to be less than fully human. In the middle ages, dwarves and hunchbacks caused much merriment. In Victorian times, people laughed at the mentally ill in psychiatric institutions.

Very early in LaughLab, we could see the superiority theory appearing in the age-old battle of the sexes. This joke was rated as being funny by 25% of women, but just 10% of men:

A husband stepped on to one of those penny scales that tell you your fortune and weight, and dropped in a coin.

'Listen to this,' he said to his wife, showing her a small white card. 'It says I'm energetic, bright, resourceful and a great person.'

'Yeah,' his wife nodded, 'and it has your weight wrong, too.'

One obvious possibility for the difference in ratings between the sexes is that the butt of the joke is a man, and so appeals more to women. Or it could be that women generally find jokes funnier than men. A year-long study of 1,200 examples of laughing in everyday conversation revealed that 71% of women laugh when a man tells a joke, but only 39% of men laugh when a woman tells a joke. To help try to tease apart these competing interpretations, we studied the LaughLab archive to find jokes that put down women, such as:

A man driving on a highway is pulled over by a police officer. The officer asks: 'Did you know your wife and children fell out of your car a mile back?' A smile creeps on to the man's face and he exclaims: 'Thank God! I thought I was going deaf!'

On average, 15% of women rated as funny jokes putting down women, compared with 50% of men.

Research suggests that men tell a lot more jokes than women. People with high social status tend to tell more jokes than those lower down the pecking order. Traditionally, women have had a lower social status than men, and thus may have learned to laugh at jokes, rather than tell them. Interestingly, the only exception to this status/joke-telling relationship concerns self-disparaging humour, with people who have low social status telling more self-disparaging jokes than those with high status. Researchers examining self-disparaging humour produced by male and female professional comedians found that 12% of male scripts contained self-disparaging humour, compared with 63% of female scripts.

Quite quickly we downloaded 10,000 jokes, and the ratings from 100,000 people. The top joke at that early stage had been rated as funny by 46% of participants. It involved the famous fictional detective Sherlock Holmes and his long-suffering sidekick Dr Watson:

Sherlock Holmes and Dr Watson were going camping. They pitched their tent under the stars and went to sleep. Sometime in the middle of the night Holmes woke Watson up and said: 'Watson, look up at the stars, and tell me what you see.'

Watson replied: 'I see millions and millions of stars.'

Holmes said: 'And what do you deduce from that?'

Watson replied: 'Well, if there are millions of stars, and if even a few of those have planets, it's quite likely there are some planets like earth out there. And if there are a few planets like earth out there, there might also be life.'

And Holmes said: 'Watson, you idiot, it means that somebody stole our tent.'

It is a classic example of two-tiered superiority theory. We laugh at Watson for missing the absence of the tent, and also at the pompous way in which Holmes delivered the news to Watson.

According to Freud, jokes act as a kind of psychological release valve that helps prevent the pressure of repression from becoming too great. Many of the jokes submitted to LaughLab supported Freud's ideas. Time and again, we would get jokes about the stresses and strains of loveless marriage, inadequate sexual performance and, of course, death:

I've been in love with the same woman for 40 years. If my wife finds out, she'll kill me.

A patient says to his psychiatrist: 'Last night I made a Freudian slip. I was having dinner with my mother-in-law and wanted to say: "Could you please pass the butter." But instead I said: "You silly cow, you've completely ruined my life."'

We also examined another source of humour - computers. LaughLab attracted lots of jokes about this topic ("The software said it needed Windows 98 or better, so I bought a Mac."). However, it also contained a few jokes actually written by a computer.

We were keen to discover whether computers were funnier than humans, and so entered into LaughLab several of the computer-composed jokes. The majority of them received some of the lowest ratings in the archive. One example of computer comedy, however, was surprisingly successful, and beat about 250 human jokes:

What kind of murderer has fibre? A cereal killer.

The results from another mini-experiment we conducted during LaughLab supported the widely held theory that some words and sounds are funnier than others, notably the mysterious comedy potential of the letter "K". Early on in the experiment, we received the following submission:

There were two cows in a field. One said: 'Moo.' The other one said: 'I was going to say that!'

We decided to use the joke to test the words/sounds theory. We re-entered the joke into the archive several times, using a different animal and noise: two tigers going "Gruurrr", two birds going "Cheep", two mice going "Eeek", two dogs going "Woof" and so on. At the end of the study, we examined what effect the different animals had on how funny people found the joke. In third place came the original cow joke, second were two cats going "Meow", but the winning animal-noise joke was:

Two ducks were sitting in a pond. One of the ducks said: 'Quack.' The other duck said: 'I was going to say that!'

By the end of the project we had received 40,000 jokes, and had them rated by more than 350,000 people from 70 countries. We identified our top joke. It had been rated as funny by 55% of the people who had taken part in the experiment:

Two hunters are out in the woods when one of them collapses. He doesn't seem to be breathing and his eyes are glazed. The other guy whips out his phone and calls the emergency services.

He gasps, 'My friend is dead! What can I do?'

The operator says, 'Calm down. I can help. First, let's make sure he's dead.'

There is silence, then a shot is heard. Back on the phone, the guy says, 'OK, now what?'

The year-long search for the world's funniest joke concluded. Did we really manage to find it? In fact, I don't believe such a thing exists. If our research into humour tells us anything, it is that people find different things funny. Women laugh at jokes in which men look stupid. The elderly laugh at jokes involving memory loss and hearing difficulties. Those who are powerless laugh at those in power. There is no one joke that will make everyone guffaw. Perhaps we uncovered the world's blandest joke - the gag that makes everyone smile but very few laugh out loud.

Five years after the study, I received a telephone call from a friend. He explained that he had just seen a television documentary film about Spike Milligan, comedian and co-founder of the Goons, and that the programme contained a very early version of our winning joke. The documentary contained a brief clip from a 1951 BBC programme called London Entertains with the following early Goon sketch:

Michael Bentine: I just came in and found him lying on the carpet there.

Peter Sellers: Oh, is he dead?

Michael Bentine: I think so.

Peter Sellers: Hadn't you better make sure?

Michael Bentine: All right. Just a minute.

(Sound of two gunshots.)

Michael Bentine: He's dead.

Spike Milligan died in 2002, but with the help of the documentary-makers, I contacted his daughter Sile, and she confirmed that it was highly likely that her father would have written the material. We announced that we believed we had identified the author of the world's funniest joke.

I am often asked what was my favourite joke from the thousands that flooded in through the year. I always give the same reply:

A dog goes into a telegraph office, takes a blank form and writes: 'Woof, Woof, Woof, Woof, Woof, Woof, Woof, Woof, Woof.'

The clerk examines the paper and politely tells the dog: 'There are only nine words here. You could send another "Woof" for the same price.'

The dog looks confused and replies, 'But that would make no sense at all.'

© Richard Wiseman, 2007.

· This is an edited extract from Quirkology, by Richard Wiseman, to be published next month by Macmillan at £14.99. To order a copy for £13.99, with free UK mainland p&p, call 0870 836 0875 or go to theguardian.com/bookshop.

ncG1vNJzZmivp6x7tbTEoKyaqpSerq96wqikaKuTnrKvr8RoaWloZ2Susb6Oa2hor5WauKa6w6aYoJmqnrum