Skip to main content

Critics say that a new study linking creativity and mental illness is lacking

Critics say that a new study linking creativity and mental illness is lacking


Can a link exist without a proper definition for creativity?

Share this story

Genetic variations that can collectively increase a person’s risk of being diagnosed with schizophrenia and bipolar disorder can also be used to predict creativity, according to a new study published in Nature Neuroscience. The study's researchers claim that this is the strongest argument yet for shared roots linking psychosis and creativity. But the correlation found in this particular study isn't very strong, some researchers argue. And without a proper definition for creativity, arguing that such a link exists could do more harm than good.

"Schizophrenia and creativity share biology."

In the study, researchers at the biopharmaceutical company Decode Genetics analyzed medical and genetic data from 86,000 people in Iceland. They found that certain genetic variants, when combined, can be used to predict bipolar disorder and schizophrenia. Then, the researchers looked at these variants in over 1,000 people who work in "creative occupations" in Iceland — meaning people who belong to national societies of visual artists, dancers, musicians, writers, and actors.

The analysis showed that people who belong to these artistic societies were more likely to carry variants that increase a person’s risk of developing either bipolar disorder or schizophrenia than people in other occupations. And when the researchers replicated the study using data from four studies carried out in Sweden and the Netherlands, they found similar results. These particular results couldn’t be explained by IQ, education, or how closely a study participant was related to a person who had been diagnosed with either mental illness, the researchers note. "What we have shown is basically is that schizophrenia and creativity share biology," says neurologist Kari Stefánsson, CEO of Decode Genetics and a co-author of the study.

The link is "astonishingly weak."

But here’s the thing: the increase in risk can be kind of misleading. Together, the variants used the study only explain about 6 percent of schizophrenia and 1 percent of bipolar disorder, according to a graph in the study. And these same variants only explain about one-quarter of 1 percent of artistic ability. That means that the same genetic variants that explain "about 1/20th of schizophrenia also explain about 1/250th of artists' ability," according to David Cutler, a population geneticist at Emory University who didn’t take part in the study. So, if you think of artistic ability as a 1-mile long road where someone with low artistic ability stands at one end and someone with high artist ability stands at the other, Cutler says, these particular variants collectively explain about 13 feet of that distance.

"Now 13 feet is not zero, but if I want to jump across a 1-mile distance, 13 feet doesn’t get me very far," Cutler says. "There is a link, and it is astonishingly weak in the sense that most people would care about." Roel Ophoff, a human geneticist at the University of California-Los Angeles, agrees. "The reported correlations are tiny," he told Genetic Expert News Service. "This means that the predictive power of the finding is limited."

There's another issue to consider. For this 13-foot-long association to matter, you have to have a pretty good definition of creativity — something that psychologist Judith Schlesinger, author of The Insanity Hoax: Exposing the Myth of the Mad Genius, argues is missing from the study.

"Creativity is simply defined by occupation," Schlesinger says, and this presumes two things. The first is "that there are no creative accountants or lawyers who play music, paint, or dance wonderfully on the weekends, preferring to actually earn a living during the week." The second is that "everyone who self-defines as a writer, artist, and joins an artistic association, automatically qualifies as unusually creative" — meaning that they "don’t suck." Overall, she says, this idea is "scientifically hollow, but convenient for their purposes."

"There is no way you can define it any differently."

Stefánsson acknowledges that there are some issues with his definition for creativity. "There's absolutely no guarantee that...members of these associations are particularly creative," he says. It’s possible that what they found is that genetic variants, when combined, are more predictive of people who would like to be creative, rather than the ones who really are. But "there is no way you can define it any differently. This is the best you can come up with in society," he says.

But using an occupation-based definition to link creativity to mental illness is harmful, Schlesinger argues. It inadvertently encourages people who are artists and who have these illnesses to avoid treatment because they believe that latter causes the former. The same goes for the families of people who suffer from certain mental illnesses; some people may fail to address behaviors displayed by their creative kin because they’re considered beneficial among artists, she says. Finally, the findings may instill fear in people who are artistic.

"Society tells them that they're headed for it," Schlesinger says. "These academics are up there in their towers, and they have no idea the damage they are doing. It's absolutely unnecessary."

"These academics are up there in their towers..."

Yet Erik Thys, a psychiatrist at KU Leuven in Belgium who didn’t work on the study, thinks that the association could provide what he calls a "positive stigma" — one that could change the "negative ideas about these people that prevail today, and are still very present in the media." Shelley Carson, a psychologist at Harvard University, also thinks that these kinds of studies can lead to positive change. "Both schizophrenia and bipolar disorder are devastating illnesses," she says. "However, work like the present study may help identify certain genetic aspects of these diseases that can confer benefits to humankind by increasing creative thinking."

This isn't the first study to make a link between creativity and psychosis. In 2013, Swedish researchers found a similar association. "The findings are really interesting and support the idea of a real association between creativity and psychosis," says Simon Kyaga, one of the researchers behind the 2013 study and an epidemiologist at the Karolinska Institutet, in Sweden. But as with this current paper, Kyaga and his colleagues relied strongly on a person’s occupation to measure their creativity.

"People who are very gifted...often have inadequacies."

There's something tantalizing about being able to link creativity and mental illness. The association is both tragic and romantic, fulfilling humanity's need to explain away other people’s talents. As Stefánsson explains, the data "shows that people who are very gifted — people who have special abilities — often have inadequacies; that is the price they pay for the genius."

We have used this argument many times in the past. How many people have heard others link classical composer Robert Schumann’s "madness" to the gorgeous music he wrote? There may be a link, but it doesn’t have to be one caused by a genetic "inadequacy." Many historians suggest that the composer suffered from an advanced form of syphilis — a disease that brings on symptoms that can be mistaken for mental illness.

The association is both tragic and romantic

There are tons of creative people out there who don't suffer from mental illnesses, and many less creative individuals do. Studies that perpetuate this idea without coming up with a more robust measure for creativity may be reaching for something that doesn’t exist. "My overall impression [of this study] is of a pot filled with various types of pasta, with pieces taken out and thrown against the wall to see what sticks," Schlesinger says.

The data demonstrating that profession is a good indicator of overall creativity are lacking. Some of the most creative people I know are coders, and they certainly aren’t being named among the creative class. If there is a link between creativity and psychosis — it’s entirely possible that one exists — we probably won’t be able to find it by looking at a person’s job. And, as far as the genetics go, we may want to find means of predicting schizophrenia and bipolar disorder that can account for more than 13 feet in a creative mile.