Cover Story

Question: "Why don't sharks bite lawyers?" Punch line: "Professional courtesy." That joke may be a groaner, but for Vinod Goel, PhD, it's also a useful research tool.

Goel, a psychology professor at York University in Toronto, is interested in figuring out how our sense of humor is hardwired. In his lab, participants read lawyer jokes, silly puns and sexist cartoons--all in the name of science.

Goel is one of a handful of researchers who have, over the past decade or so, begun to use brain imaging to explore people's responses to humor. Mainly using functional magnetic resonance imaging (fMRI), the researchers have been finding that some common neural circuitry--particularly the brain's reward circuit, the set of structures that underlie our reactions to all sorts of pleasurable experiences, from eating to sex--may also underlie our response to humor.

At the same time, different sorts of humor, from puns to visual jokes, may involve separate areas of the brain that do the cognitive work of figuring out what's so funny.

"If I tell you a joke, you could respond in several ways," Goel says. "You might not get it, in which case that's the end of it. Or, you might get it, but maybe someone told you the same joke yesterday--so it doesn't seem funny. Or you might really find it funny….In our study we wanted to find the neural distinctions between these things."

Where is funny?

Researchers come to the study of humor from many different directions. "It's not like there's such a thing as 'humorology,'" says California Institute of Technology humor researcher John Allman, PhD. "Most of us became interested in humor because of what it could tell us about other subjects."

For Goel, that other subject was reasoning and problem solving. Goel sought to explore "set shifting"--the mental leaps people make to solve problems. But most laboratory tests of set shifting, such as logic puzzles that involve moving matchsticks into various configurations, are, Goel says, boring.

"We wanted a different paradigm where you have to make a quick set shift, and humor is ideal," he explains. To understand a joke, you must quickly readjust your frame of reference when the joke takes an unexpected turn--as in the lawyer joke above, which veers from the literal, aquatic meaning of the word "shark" to its use as a disparaging nickname for lawyers.

Goel's 2001 study, published in Nature Neuroscience (Vol. 4, No. 3, pages 237-238), was the first to use fMRI to explore humor. He imaged the brains of 14 college-student participants as they listened to two types of jokes. One type, puns, involved plays on the literal meanings of words: Why did the golfer wear two sets of pants? He got a hole in one. The other type, semantic jokes, relied on more abstract concepts: What do engineers use for birth control? Their personalities.

Goel found that when the participants heard the semantic jokes, blood flow increased in their bilateral posterior temporal lobes--a region involved in semantic language processing. When the participants listened to the puns, the jokes activated their left inferior prefrontal cortex--an area involved in phonological processing.

But Goel also wanted to find out whether there were any universal humor regions of the brain. To do so, he asked the participants whether they found the jokes funny. Then, he compared the jokes they found funny with the ones that fell flat. The funny jokes, compared with the unfunny ones, all activated a part of the brain's reward system called the medial ventral prefrontal cortex, regardless of whether the joke was semantic or phonetic.

Converging evidence

Other researchers have followed up on Goel's initial study and found complementary results.

In a study published in November in the Proceedings of the National Academy of Sciences(Vol. 102, No. 45, pages 16,496-16,501), for example, Stanford University psychiatrist Allan Reiss, MD, and his colleagues scanned participants' brains as they read cartoons. The researchers found that the cartoons the participants rated as "funny" activated both the prefrontal cortex, which is used in language processing, and a part of the reward system called the nucleus accumbens.

And in a 2004 study published in Neuroimage (Vol. 21, No. 3, pages 1,055-1,060), Dartmouth College psychologist William Kelley, PhD, and graduate student Joseph Moran imaged the brains of 25 young-adult participants--11 men and 14 women--as the participants watched episodes of the sitcoms "Seinfeld" and "The Simpsons." By matching up the fMRI readings with the laugh tracks of the sitcoms, the researchers could explore what was happening in their participants' brains at the funniest moments in the shows and in the moments right before that, when presumably the participants were working to "get" the jokes.

The researchers found that at the moments the participants were "getting" the jokes--the two seconds before the laugh track laughs--their posterior temporal lobes, an area of the brain that helps resolve incongruities, lit up. At the moments that they were presumably enjoying the joke, during the laughs, the amygdala was more active--agreeing perfectly with Goel's results.

Finally, a just-published study by John Allman and graduate student Karli Watson in the online journal Cerebral Cortex (Vol. 16) examined people's brains as they looked at language-based cartoons ("The Far Side" cartoons that relied on captions for their humor) and purely visual sight gags (cartoons that were funny even without a caption). Allman found that the sight gags activated high-level visual processing areas, and language-dependent cartoons activated language-processing areas in the left temporal lobe. Both types of cartoons, though, activated emotional areas in the amygdala and midbrain.

Overall, Moran says, imaging research is beginning to point to the parts of the brain that we use to understand and appreciate humor. But, he says, the results don't yet constitute a body of knowledge as definite and extensive as researchers have about, say, language processing.

"These studies are still very exploratory," he explains. "So while the results broadly agree with each other, there's no standard set of regions that we know are involved in humor, the way we know about Broca's area and language."