Genetic risk for schizophrenia or bipolar predicts likelihood of being a creative professional

Genetic risk scores for schizophrenia or bipolar disorder, calculated by looking at variations in a large number of genes could predict both psychiatric disorders and membership of artistic societies in a sample of 86,292 Icelanders, according to a study in Nature Neuroscience. The same scores identified creative professionals in two further populations from Sweden and the Netherlands, leading the authors to suggest that genetic variants that increase the risk for schizophrenia and bipolar disorder also influence creativity.


Dr. David Cutler, Assistant Professor, Department of Human Genetics, Emory University (webpage):

Expertise: population genetics applications to human disease studies, analyzing whole genome data sets to discover genetic variants associated with disease.

“To the extent that we understand the genetic basis of either schizophrenia or bipolar disorder it appears that many, many genes – perhaps even most of the genome – are involved. You can think of the model as something akin to everything in the genome contributing a tiny bit, with some genes contributing a little more than others; at every variant in those genes, one allele (genetic variant) slightly increases the risk and the other allele slightly decreases the risk. The risk scores the authors are modeling are therefore akin to the sum of the effects of everything in genome, more or less, and they show that these risk scores do in fact predict a small but significant fraction of the variation in schizophrenia or bipolar. All of this is fairly well known.

“If more or less every gene contributes to both schizophrenia and bipolar, these genes must do lots and lots and lots of things other than simply contribute to those diseases. Thus, the variants that in combination ever so slightly increase your risk of schizophrenia and bipolar must also ever so slightly increase or decrease other things too, and in the paper the authors show that the variants also slightly increase the chance of you being a professional artist.

The authors show that the variants in the genome that when combined explain approximately 6% of of all schizophrenia also explain 1/4 of 1 percent of the variation in artistic ability and the variants that when combined explain about 1% of all bipolar explain about 1/4 of 1 percent of the variation in artistic ability. To put the number “1/4 of 1 percent” into a metaphorical context, if the distance between me (the least artistic person you are going to meet) and an actual artist is 1 mile, these variants appear to collectively explain 13 feet of that distance. Most of the distance between the artist and me is therefore due to other genetic variants and / or environmental factors.

“The findings are likely robust, but they are also very, very small, and importantly must be interpreted as I have laid out above. The effects are tiny and combined across hundreds or thousands of genes to even be detectable. Nonetheless, they are probably real and we can conclude that genes involved in schizophrenia and bipolar are probably broadly involved in all sorts of neurological and cognitive function including, but surely not limited to, cognitive functions related to artistic endeavors.”


Dr. Mark A. Runco, Distinguished Research Fellow of the American Institute for Behavioral Research and Therapy; Editor, Creativity Research Journal  (webpage):

Expertise: creativity research focusing on idea generation and divergent thinking. Dr Runco has devised a comprehensive battery of tests for the assessment of creativity.

“There are indeed indications that various forms of psychopathology share genetic roots with some indicators of creative talent. This new research by Stefansson et al. thus supports earlier reports, and such replication is useful in the sciences. It is especially useful because psychopathology is so varied.

“Two points should be underscored: one is the focus on creative professions and creativity defined in terms of memberships in societies and professional efforts. This is a very limited view. For one thing, not all creative people join or are invited to societies, and not all members of society are unambigously creative. Additionally, some domains of creativity differ from the arts. Twenty years ago data showed that psychopathology is related to artistic creative but not to all forms of creativity. Most important is that professional creativity reflects more than creative talent (e.g., self-promotion, to quote one earlier study in this area) and does not say much at all about creative potentials that are widely distributed in the population.

“The work of Stefansson et al. is based on impressively large samples and their description is appropriately cautious: the genetic contributions lead to propensities which might in turn lead to psychopathology, or to creative achievement. One notable concern is that genetics are not “either/or.” It makes no sense to ask if the association between psychopathology and creativity is a result of shared genetics or environment. Genes supply a range of reactions and interact with the environment to varying degrees. The nature/nurture dichotomy is false, especially when discussing things are complex as creativity and psychopathology. A next step in this line of work might include consideration of the research showing DRD2 and DRD4 (dopamine receptor genes) to be related to indicators of creative potential.”


Dr Roel Ophoff, Professor of Psychiatry and Human Genetics, David Geffen School of Medicine, University of California, Los Angeles (webpage)

Expertise: identification of genetic susceptibility of complex traits, in particular neuropsychiatric illnesses such as schizophrenia and bipolar disorder.

“This study deals with fascinating phenotype correlations in which known genetic risk measures for schizophrenia and bipolar disorder are correlated with measures of creativity. The overall outline of the study is appropriate in that it is a population-based discovery sample from Iceland and independent replication cohorts are from the Netherlands and Sweden. It is worth noting that even though the p-values are very significant that the reported correlations are tiny. This means that the predictive power of the finding is limited.

“The polygenic risk score used in the study is a collective risk of many individual genetic variants in the genome for either schizophrenia or bipolar disorder. Given the way these risk scores are differently established for the two disorders, I am surprised that the findings with regard to creativity measures are yet so similar.

“It is also worth noting that no polygenic risk profiles of any of the other neuropsychiatric traits, which are readily available, were used. The correlations between creativity and the other neuropsychiatric traits were previously studied by Kyaga et al in a Swedish registry (references 8 and 9 provided in the study). Our understanding of the correlation between creativity and schizophrenia and bipolar disorder risk scores would improve enormously if the risk profiles of additional neuropsychiatric traits were included as well as those of non-neuronal/brain disorders.

“What makes me cautious about the overall conclusions, however, is that the replication finding, the key aspect of the study, is mainly driven by only one of the four cohorts (i.e. the Netherlands Twin Registry, NTR) and not the other three cohorts of roughly the same size. There are differences between these replication samples that may affect the results and conclusions. For example, while the latter three cohorts have similar percentages of artists, the number of artists in the NTR cohort is substantially higher. It is clear that the polygenic risk scores of schizophrenia and bipolar disorder do not predict creativity equally across the different cohorts, as one would expect. Let’s hope that the genetic correlations between creativity and neuropsychiatric disorders will be replicated in independent studies, so that we can start think about the underlying biology.”


Dr.Vishwajit L. Nimgaonkar, Professor of Psychiatry and Human Genetics, Director, Program for Genetics and Psychosis, University of Pittsburgh (webpage)

Expertise: identifying genetic variants and environmental factors conferring susceptibility to severe psychiatric disorders to enable rational therapeutics, genetic counseling and prevention.

“The paper is authored by a highly regarded group. The genetic methodology is robust. They suggest that a genetic predisposition to schizophrenia or bipolar disorder increase chances of membership in Icelanding artistic societies. The nub is whether membership in an artistic society indicates creativity, as the authors acknowledge. I question this assumption and it is an important caveat.

“By analyzing a national database, the study raises the credibility of prior, smaller published studies. Creativity is such a broad and difficult to define term that I personally would hesitate to investigate its genetic underpinnings.”


Polygenic risk scores for schizophrenia and bipolar disorder predict creativity‘ by Power et al, published in Nature Neuroscience on Monday 8 June, 2015.

Declared interests (see GENeS register of interests policy):

No interests declared

Please feel free to leave your comments below, but be aware that by doing so you agree to our Terms & Conditions.