Apes love being tickled, rats giggle: Hannah Fry on the science of laughter | The Formula To Life With Hannah Fry

While the British sense of humour is something every Brit holds dear, it’s fair to assume many British jokes would leave other cultures mystified. I think there is something quite interesting in that: it means that humour is not universal. The things we find funny are not innate, they’re cultural.

That makes humour very different from laughter, which every human (even the grumpy ones) will have experienced at some point or another. Unlike humour, laughter is much easier to understand and study scientifically.

For instance, TV producers have long known that adding pre-recorded “laugh tracks” makes people find you funnier than when they don’t hear that laughter (I’m looking at you, 90s sitcoms). But neuroscientist Robert Provine found that there didn’t even need to be any gags whatsoever. In 2013, Provine did a study where he simply played the laugh track on its own, and strangely found that was enough to trigger laughter in almost all of the study’s participants.

Provine went on to revolutionise the way we understand laughter, by taking it out of the lab and observing people having a laugh in public spaces. He was able to demonstrate empirically that people are vastly more likely to laugh in the presence of other people rather than when they are on their own – 30 times more likely, in fact. And what’s more, the reason people laugh is rarely because they find something funny: more often we’re using it as a form of social communication, a non-verbal way of identifying something as a positive, non-threatening interaction.

And laughter isn’t unique to humans, you’ll find it throughout the animal kingdom. Apes love being tickled and vocalise throughout; kea parrots, well known for their playfulness, have a distinct warble to call others to fun; and rats have an ultrasonic giggle they make when playing or being tickled on the tummy. Actually, if the rats have a handler who regularly tickles them, they’ll start laughing even as the person enters the room, such is their delight at the anticipation of being played with.

But humour? Humour is much harder to nail down. Why are some things funnier than others? And how do you define what it means to be funny if it changes depending on who you ask?

Perhaps the most persuasive description I’ve heard is that humour is the build up and release of tension. That certainly does a good job at describing some famous comedic moments – Del Boy falling over on the bar, Basil Fawlty hitting the car with a tree branch, or Mark ​​Simmons’ gag at this year’s Edinburgh fringe: “I was going to sail around the globe in the world’s smallest ship, but I bottled it” – however, it’s a definition that falls short of being a usable formula for being funny. Especially when you aren’t a human yourself.

The newest batch of generative AI has been trained by reading all of the internet (minus some of the nastier parts) and is exceptionally good at recreating styles of humour that have gone before. That includes auto-completing jokes, if it’s given a clear framework. For instance, I asked an AI chatbot to find a punchline to the joke: “I was going to sail around the globe in the world’s smallest ship, but …” and it came up with “it gave me a sinking feeling”. Not bad, right?

But finding the framework itself – finding an original way to narrow in on the weird, quirky ideas that somehow communicate a shared human experience, that is much more difficult for an algorithm that has no real experience of the world at all, except for what it’s read online. And for now, at least, those are the surprising ideas that connect with an audience.