The evolution of laughter

A British researcher believes that he has found the Darwinian key to comedy: laughter expresses pattern recognition, a critical element to survival and attracting a mate.

Charlie Chaplin has comedic appeal because his routines turn recognisable patterns upside down, according to a new scientific theory.

Stop me if you have heard this one before. 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 a silence, then a shot is heard. Back on the phone, the guy says: "OK, now what?"

Whether or not this was a mighty knee-slapper for you, the gag - submitted among more than 40,000 others to the LaughLab website - was nevertheless crowned in 2002 as "the funniest joke in the world" by British psychologists at the University of Hertfordshire. But why so many people found that particular punchline so amusing has less to do with the content of the joke and much more to do with something quite humdrum - patterns.

That is the foundation for a new theory that purportedly answers centuries-old questions about the evolutionary function of humour - one of mankind's defining characteristics that separates the species from the animal kingdom. "People have presumed humour is an emotion that exists in laughter in order to rectify a situation," said Alastair Clarke, the UK-based science writer whose new research into the mechanism of humour has been described in the periodical Science as the "first analytically justifiable explanation of any instance of humour".

But he argues that humour is a cognitive, intellectual process, not an emotional one. "The research before now has tended to be based on the idea that if you tickle a chimp, it laughs," Mr Clarke said. "But laughter due to tickling is not primarily related to humour; it's just in the same way that crying while peeling an onion is not related. Laughter is an automatic, knee-jerk response." Mr Clarke boiled down his Pattern Recognition Theory of humour in simple terms: "In a nutshell, what it says is that we find something amusing when the brain recognises a pattern that surprises it."

The lighthearted "separated at birth?" gag, in which photos of two lookalikes are compared side-by-side is an illustration of spatial, two-term repetition. In 2005, the American TV host Conan O'Brien discovered that his red wavy hair and fair Irish complexion resembled that of Finland's first female president, Tarja Halonen. In a long-running comedy bit, he popularly matched his face with Ms Halonen, who was herself amused by their resemblance.

Viewers find the gag amusing because they are surprised by their recognition of the similarities between the two people, Mr Clarke says. The same principle applies when it comes to fun-house mirrors that magnify, minimise or exaggerate certain parts of the body, or "one extra pattern". Pattern Recognition Theory is also evident in what Mr Clarke describes as the "it's so true" brand of observational stand-up comedy famously popularised by American comics such as Jerry Seinfeld.

"It's two stages," said Mr Clarke. "The comedians' words depict something first, then the second stage is the mental image the observer retains of the matter being portrayed. If the audience recognises a similarity and it surprises them, then it's amusing." So it is not the content, but the structure of a joke that determines what we find amusing, he argues. Mr Clarke's humour hypothesis, which was published last month as the Pattern Recognition Theory of Humour, seeks to explain humour's evolutionary function as a "weapon in the cognitive arsenal" of human beings.

The ability to recognise patterns instantly was possibly a remarkable asset for early Homo sapiens, Mr Clarke said. The broadcast of this recognition - in the form of laughter - is a "reward" associated with the release of fuzzy-feeling endorphins. Laughter also communicates the recognition of patterns. "We are attracted to people who laugh. It's an important facet for searching for mates," he said. "Those who make us laugh are showing us a pattern they've recognised and which still surprises us. They'll recognise things that we haven't yet, and that enables us to choose people who can help us in situation of survival.

"This humorous response has been a driving force in evolution. By rewarding the brain for instantly recognising patterns, it has played a major role in the advancement of the species." As for why two people sometimes find different things funny, or why two people can laugh at the same stimulus for different reasons, Mr Clarke said it came down to the individuals and the patterns they recognised. Cultural assumptions have a role in shaping these patterns.

The development of humour may also be linked to the development of an infant's cognitive abilities. In the future, researchers may use MRI scans to monitor brain activity during a simple process of surprise repetition - such as with the "peek-a-boo" game - to study whether early life instances of humour contribute to the child's cognitive development. Mr Clarke's theory has, as he expected, met some opposition from scholars who believe science has no place in trying to explain what makes us fundamentally human.

"Some people feel that science should not be allowed to explain human behaviour, which is potty of course," he said. "There have been various, knee-jerk reactions when many of these people haven't actually examined the theory. I expected those reactions." Having felt that he has now identified the mechanism for humour, Mr Clarke is turning his attention towards artificial intelligence. He noted that a "lack of sense of humour" was often used as criticism to describe something as cold and robotic.

But that could change, he said. "We have the best chance ever now of producing something that would seem less robotic because it would be able to exhibit a sense of humour," Mr Clarke said. "We're trying to design an AI model that can attempt humour." The problem is that AI cannot absorb information in the same way as humans can, so people will have to go online and input jokes and comments to build up the linguistic computer programme's linguistic experience.

"What we're looking at in years to come is producing an inanimate being with a sense of humour," he said. "The possibility is very real that we can develop AI capable of doing that, making it seem less robotic as a result." @Email:[e-mail]