Integrative Literature Review

Psychology of Aesthetics, Creativity, and the Arts Psychopathology in World-Class Artistic and Scientific Creativity Gregory J. Feist, Daniel Dostal, and Victor Kwan Online First Publication, October 21, 2021. http://dx.doi.org/10.1037/aca0000440

CITATION Feist, G. J., Dostal, D., & Kwan, V. (2021, October 21). Psychopathology in World-Class Artistic and Scientific Creativity. Psychology of Aesthetics, Creativity, and the Arts. Advance online publication. http://dx.doi.org/10.1037/aca0000440

Psychopathology in World-Class Artistic and Scientific Creativity

Gregory J. Feist1, Daniel Dostal2, and Victor Kwan1 1 Department of Psychology, San Jose State University

2 Department of Psychology, Faculty of Arts, Palacký University Olomouc

The role of psychopathology in creative achievement has long been a debated topic in both popular culture and academic discourse. Yet the field is settling on various robust trends that show there is no one answer. Conclusions vary by level and kind of creativity and level and kind of psychopathology. The current study sought to replicate previous findings that linked lifetime rates of psychopathology to world-class levels of crea- tivity. A total of 199 biographies of eminent professionals (creative artists, creative scientists, eminent athletes) were rated by raters who were blind to the identity of the eminent person on 19 mental disorders using a 3- point scale of not present (0), probable (1), and present (2). Athletes served as an eminent but not creative comparison group to discern whether fame, independently of creativity, was associated with psychopathology. Results showed that artists exhibited higher lifetime rates of psychopathology than scientists and athletes in the more inclusive criterion for psychopathology (i.e., it was either probable or present), whereas both artists and athletes exhibited higher rates than scientists in the stricter criterion for psychopathology (i.e., it was present). Apart from anxiety disorder, athletes did not differ from the U.S. population in lifetime rates of psychopathol- ogy, whereas artists differed from the population in terms of alcoholism, anxiety disorder, drug abuse, and depression. These data generally corroborate and replicate previous biographical research on the link between artistic creativity and life-time rates of psychopathology.

Keywords: creativity, mental illness, psychopathyology, artists, scientists

Supplemental materials: https://doi.org/10.1037/aca0000440.supp

The stereotype of the mad genius has been a popular notion for quite some time. Brilliant yet mad artists, such as Vincent van Gogh, innovators such as Howard Hughes, and mathematicians such as Isaac Newton have inspired this view throughout history (Brownstein & Solyom, 1986; Jeste et al., 2000; Perry, 1947). The list of geniuses with mental illness could go on and on. Of course, the list of creative geniuses not afflicted with mental illness would no doubt be at least as long. The range of conclusions on the ques- tions are highlighted in the following four quotes: “There is no

great genius without some mixture of madness” (Aristotle, as para- phrased by Seneca, 2007; cf. Motto & Clark, 1992).

Thus the creative genius may be at once naïve and knowledgeable, being at home equally to primitive symbolism and to rigorous logic. He is both more primitive and more cultured, more destructive and more constructive, occasionally crazier and yet adamantly saner, than the average person. (Barron, 1963, p. 224)

“Psychopathology and creativity are closely related, sharing many traits and antecedents, but they are not identical, and out- right psychopathology is negatively associated with creativity” (Simonton, 2006). “Despite centuries of professional attention, the link between creativity and madness remains more stereotype than science” (Schlesinger, 2017, p. 60).

Anecdotal evidence is just that—anecdotal. But is there truly a legitimate empirical link between psychopathology and creative genius? Over the last 10 to 15 years, the field has begun to settle on various robust trends that show there is no one answer to the question of creativity and mental health. Conclusions vary by level and kind of creativity and level and kind of psychopathology (Baas et al., 2016; Beaussart et al., 2017; Fisher, 2015; Glazer, 2009). As Simonton (2014, 2017a, 2017b, 2019) has recently dem- onstrated, all of these positions may have validity, with both linear and nonlinear relationships. As is true of all entrenched scientific debates, there must be some truth to each side otherwise one side would die off very quickly. The current study sought to replicate and extend previous biographical findings that linked lifetime rates of psychopathology to world-class levels of creativity by examin- ing the moderating effects of the relationship.

Gregory J. Feist https://orcid.org/0000-0002-3123-1069 Daniel Dostal https://orcid.org/0000-0001-5489-7907 This research was in partial fulfillment of the Master’s Thesis for Victor

Kwan. Portions of these findings were presented in 2019 at the Southern Oregon University Creativity Conference, Ashland, Oregon, and in 2016 to the Ecole Polytechnique Federale de Lausanne (EPFL), Lausanne, Switzerland. The authors thank Abiola Awolowo, Brian Barbaro, Kimya Behrouzia,

Catherine Erickson, Evander Eroles, Janet Dai, Adrian Davis, Sheila Greenlaw, Jennifer Kang, Illhame Khabar, Thomas Lu, Caitlyn Ma, Dat Nguyen, Elizabeth Shallal, Kimia Sohrabi, Eldita Tarani, Ryan Willard, and Laura Weber. Data collection was sponsored by a grant from the Research Foundation, San Jose State University. We have no conflicts of interest to disclose. Correspondence concerning this article should be addressed to Gregory

J. Feist, Department of Psychology, San Jose State University, One Washington Square, San Jose, CA 95192-0120, United States. Email: greg [email protected]

1

Psychology of Aesthetics, Creativity, and the Arts

© 2021 American Psychological Association ISSN: 1931-3896 https://doi.org/10.1037/aca0000440

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Creativity

Creativity researchers have most regularly described creativity as consisting of two qualities, namely originality/novelty and mean- ingfulness/usefulness (Amabile, 1996; Feist, 2017; Runco & Jaeger, 2012; Sternberg, 1988). A creative endeavor must not only be dif- ferent from what has been previously performed in a given domain but also useful. In this case, the term “usefulness” can also mean beautiful or provocative for artwork and literature. Some have argued that the term “usefulness” could be replaced by the word “meaningful” (Feist, 2017). With this change in terminology, the need to qualify “useful” as also beautiful or provocative is no longer necessary. Products of both art and science can be meaningful, whereas a piece of artwork would not necessarily be useful. Mean- ingful makes clear that the meaning is in the evaluator, and this can be the general public, experts, peers, or historians, among others. In the last 10 years, however, a few creativity researchers have

argued for three, rather than two, factor definitions of creativity. Simonton (2013, 2016), for example, has proposed a logical quan- titative model whereby a creative idea or solution is a multiplica- tive function its originality, utility, and surprisingness. More formally, and omitting the i subscript for each individual idea, cre- ativity c = (1 – probability p) utility u (1 – prior knowledge of util- ity v) and where p, u, v, and c each range from 0 to 1. The compelling aspect of this formulation is its multiplicative function, whereby any value of 0 for p, u, or v results in an idea that is not at all creative.

Psychopathology and Creativity

Research and theory on psychopathology and creativity are growing and contentious. A recent edited volume entitled Creativ- ity and Mental Illness captures the history, current state of the field, and the wide range of views on the topic (Kaufman, 2017). Although the ancient Greeks (Aristotle in particular) were the first people in the Western world to examine the nature of creativity and its association with “melancholia,” it was not until the Roman- tics in literature in the 1830s that the argument was made for any connection between serious mental affliction and creative genius (Becker, 2017). The modern literature on the topic was jumpstarted in the 1980s and

1990s with the work of Andreasen, Jamison, Richards, and Ludwig. In her early investigations, Andreasen (1978, 1987) reported qualified relationships between creativity, especially literary creativity, and men- tal illness, making use of historical, familial, and genetic studies. Jami- son (1996) reported historical, biographical, and literary evidence for the association between artistic creativity and mood disorders, most specifically bipolar disorder. Jamison et al., (1980) also examined the relationship between creativity and bipolar illness in noncreative peo- ple and found that the hypomanic period led to heightened creativity. Richards and colleagues (1988) also found that it was mild levels of mania (hypomania) and bipolar (cyclothymia) that were most strongly associated with creative thinking. Large-scale biographical examina- tions by Ludwig and Post in the 1990s reported associations between artistic creativity and lifetime rates of mental illness. Ludwig (1992, 1995) examined more than 1,000 eminent professionals, including, but not limited to, artists, writers, scientists, and musicians and revealed that extremely creative individuals, especially in the visual and literary arts, exhibited elevated rates of various lifetime psychopathologies.

Post (1994) also drew a similar conclusion from biographical analysis of more than 200 world-famous creative people. The sample in this study was restricted to deceased subjects of biographies reviewed by the New York Times. These biographies were then examined for signs of psychopathology in each eminent professional and correlated with each domain of expertise. The results showed that people who excelled at creative endeavors such as poetry and fiction writing experienced higher rates of psychopathology than scientists or politicians.

To be sure, partly owing to inconsistent empirical results, not all scholars agree there is a connection between high levels of creativ- ity and psychopathology (Sawyer, 2011; Schlesinger, 2009, 2017; Thys et al., 2014). The most outspoken and harshest critique comes from Schlesinger (2009, 2012, 2017). She essentially dis- misses the entire field and literature on creativity and “madness” as based on nothing more than poorly conducted and flawed research (e.g., biographies are dismissed as little more than “gos- sip”), even going so far as to call it a “hoax” (Schlesinger, 2012). She takes particular issue with three of the key figures in the field, Andreasen, Jamison, and Ludwig. Andreasen (1987) had too few participants, relied on personal relationships with participants, and overgeneralized results. Jamison’s (1996) “autopsy subjects” of 166 deceased artists, writers, and musicians is challenged. Schle- singer critiques three of the 166 subjects (Michelangelo, Emerson, and Cole Porter) and yet fails to provide a more general and struc- tured critique of the participant selection. Her methodological criticisms have some validity because no research is without limi- tations and flaws, but Schlesinger does her arguments a disservice when she makes many absolutist statements that no one really claims, such as the supposedly common belief that “no one receives the gift of genius without the curse of depression” (Schle- singer, 2017, p. 60) or that much of the research gets propagated without people reading the original articles.

Moderating Effects in the Relationship Between Creativity and Psychopathology

The apparent contradictory set of results gains clarity when we begin to look more closely at the reasons for the mixed, moderat- ing, and complex findings. One study, for example, reported that gender moderates the relationship between creativity and psycho- pathology, with positive results only holding for men (Martín-Bru- fau & Corbalán, 2016). More general findings highlight four common moderating effects in the relationship between creativity and psychopathology (Baas et al., 2016; Beaussart et al., 2017; Feist, 2012; Fisher, 2015; Glazer, 2009; Silvia & Kaufman, 2010; Simonton, 2017a, 2019):

• degree of creativity • domain of creativity (art v. science) • degree of psychopathology • domain of psychopathology

Degree of Creativity

Creative achievement exists on a wide spectrum, from minor to major contribution. For decades, researchers who study creativity have realized there is a need for distinguishing levels of creativity. To label someone creative is nearly meaningless if it does not dif- ferentiate historical genius global level creativity from minor, indi- vidual, everyday creativity. As Kaufman and Beghetto (2009)

2 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

argued, there are at least four distinct levels of creativity that are more or less developmental. First, there is mini-c creativity, that is, personal level of creativity, where someone is creating ideas or behaviors that are novel and meaningful to him or her only. If that person goes on to produce novel products, ideas, or behaviors that involve solving everyday problems creatively, we would call that little-c creativity. This could involve activities such as making a new meal, creatively making use of materials around the house to build a play structure, or writing an essay for a class. Next, we move to people who make a living doing creative work, such as writers, musicians, scientists, architects, actors, and painters, but whose acclaim and impact are regional rather than national or international. This is Pro-C creativity, for “professional.” In other words, the circle of people who find it meaningful or useful is rela- tively narrow. It has not yet changed the course of the discipline or started a new branch of the discipline. Lastly, there is Big-C crea- tivity, or genius level or historically significant creativity. These are people who change the course of their fields, may have biogra- phies written about them, and often earn award and recognition from their peers for doing the best work in their field, and some- times are studied by historians. In short, these four levels of the creative spectrum move from smallest to largest social/cultural circle: self, family, region, and nation/world. As we discussed at the outset, there are different degrees of cre-

ativity, different kinds of creativity, and how one defines and measures it matters (Reiter-Palmon & Schoenbeck, 2020). One reason for conflicting results in the literature on creativity and psy- chopathology is the fact that degrees of creativity are conflated. The size of the effect seems to be directly related to the degree or level of creativity, with the effect being largest in Big-C creative samples and smallest in little-c creative people. Moreover, method of research is related to these two levels, with most little-c samples being studied psychometrically or experimentally and Big-C or Pro-C samples being studied historiometrically, biographically, or epidemologically (Johnson et al., 2012; Paek et al., 2016; Rich- ards, 1990; Silvia & Kimbrel, 2010; Taylor, 2017). For example, in a large Swedish national-population sample that included tens of thousands professional (Pro-C) creative artists and scientists, Kyaga and colleagues (2011) reported significant associations between creative professions and being treated for bipolar disorder or having a sibling or parent treated for schizophrenia. Moreover, in a study of an undergraduate (little-c creativity) sample, Silvia and Kimbrel (2010) found that anxiety and depression could only explain 3% of the variance in creative thinking. Finally, Johnson and colleagues (2012) conducted an extensive qualitative review of the literature on bipolar disorder and creativity and found stron- ger effects with Big-C than little-c creativity.

Domain of Creativity (Art Versus Science)

One of the main empirical findings from the literature on creativ- ity and psychopathology is the stronger effect size between artistic than scientific creativity and psychopathology (Damian & Simon- ton, 2015; Ludwig, 1995, 1998; Post, 1994). Trauma, stress, mood disorders, and fear of death each seem to motivate artistic creativity in a way they do not motivate scientific creativity or innovation in business. For instance, Ludwig’s later analyses (Ludwig, 1998) made clear that lifetime rates of psychopathology are mostly ele- vated in the arts compared with the sciences, and in the expressive

arts in particular (writing (fiction and nonfiction), poetry, and visual arts; see Figure 1). The performing artists (musicians and dancers) had moderately elevated rates of psychopathology, whereas the for- mal artists (architects) were not different from the general popula- tion. He argued and reported, therefore, that it is the more expressive, intuitive, and subjective creative professions where psy- chopathology and creativity should be and was most likely to be associated. The more formal, logical, precise, and objective profes- sions should be and were less likely to see psychopathology. In short, in the expressive arts, personal meaning, subjectivity, and emotion play a motivational role in ways not common in the more formal creative professions. Further analysis within the artistic forms revealed consistently higher rates of psychopathology in the emotive/expressionistic styles than in the formal/realism styles. To be sure, these are correlational findings, so whether those with men- tal and mood disorders are drawn to artistic careers or the other way around has yet to be established.

Other scholars have reported similar patterns whereby scientists suffer relatively low lifetime rates of psychopathology, whereas the other professions, especially the arts, had elevated rates of mental illness compared with base-rates in the general population (Damian & Simonton, 2015; Post, 1994; Simonton, 2014). A bio- graphical replication of the mental health status of 40 jazz musi- cians from the 1940s and 1950s replicated Ludwig’s basic finding (Wills, 2003). Wills found elevated rates on chemical dependency, mood disorders, and anxiety disorders in jazz musicians. Heroin addiction was also elevated in Will’s jazz sample, with 52% sam- ple having heroin problems at some point during their lives.

The one exception to this general finding with scientists may be autism-spectrum disorder (ASD), predominantly in the high-func- tioning range (Baron-Cohen et al., 2007; Billington et al., 2007; Focquaert et al., 2007; Thomson et al., 2015; Wei et al., 2017). We should point out, however, that most of this research on ASD and science and technology is with interest and careers in STEM (sci- ence, technology, engineering, and math) and not necessarily highly creative scientists and technologists. We are interested in examining whether this relationship holds at high levels of scientific achieve- ment and creativity.

Although Ludwig (1992) argued that psychopathology explained very little variance in terms of scientific achievement, there are cer- tain circumstances where psychopathology may exist in scientific creativity. A good demonstration of this is the analysis by Ko and Kim (2008) of 76 scientific geniuses from Simonton’s sample. Ko and Kim predicted and found that the relationship between scien- tific creativity and psychopathology would be moderated by the kind of contribution the scientist made, namely whether it preserved or rejected paradigms. Specifically, scientists without pathology were more creative when they made paradigm-preserving than par- adigm-rejecting contributions. Paradigm-preserving is a contribu- tion that advances but does to change a field’s direction. Paradigm- rejecting contributions do in fact change a field’s direction. Those with psychopathology, especially psychotic disorders, were more creative (based on biographical/encyclopedia index ratings) when they made paradigm-rejecting rather than paradigm-preserving contributions. In addition, this moderator analysis more than doubled the variance explained (18% vs 8%) by psychopathol- ogy in scientific creativity compared with Ludwig’s (1992) sam- ple. An implication of these results is that psychopathology may

PSYCHOPATHOLOGY AND CREATIVITY 3

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

be detrimental to their creative contribution only to scientists who make paradigm preserving contributions. Taylor (2017) discussed important ways how art and science

may be different: art requires less formal education and has more flexible work schedules (Simonton, 2010; Simonton & Song, 2009) and hence does not exclude people with mood disorder; poets, writers, and so forth more likely to experience trauma, men- tal health problems in childhood; Ludwig (1998) suggests “occu- pational drift” owing to emotional expressiveness required for different professions. In addition, we argue for internal versus external experiences and meaning and understanding. That is, art —especially the expressive arts of literature and visual arts—is of- ten based in internal emotional/traumatic/stressful personal experi- ences that lead to a need/motive to understand, give meaning to, or to express these experiences in artistic form, whether it be music, dance, poetry, visual art, or fiction writing (Akinola & Mendes, 2008; Gardner, 1973; Forgeard et al., 2017; Thomson, 2017). It may not always be traumatic experiences, but at the least involves a need or desire to express one’s perceptions of the world and their place in it (Forgeard et al., 2017). Science, on the other hand, is of- ten driven by a need to understand and figure out the external world, especially in the physical sciences. Scientists most often ask: “What is that? Why? How can we explain that?” That is a

search for meaning and understanding, to make sense of one’s external rather than world.

Degree of Psychopathology

It has become clear with the accumulation of research that high levels of mental illness are generally at odds with high levels of creativity in any field, including art. Even when creativity and ill- ness go together in certain people, it is mostly during periods of relative calm and milder dysfunction that creative behavior may coexist with pathology. In short, there is a nonlinear relationship between illness and creativity, with mild to moderate levels of pa- thology being most associated with creative achievement (Abra- ham, 2017; Acar et al., 2018; Feist, 2012; Kinney & Richards, 2017; Simonton, 2017a; Swain & Swain, 2017; Wuthrich & Bates, 2001). Jamison’s well-known book Touched With Fire (Jamison, 1996), for instance, presented historical evidence for an associa- tion between bipolar disorder and creativity, especially in literature (i.e., Big-C creative people). However, she also made clear that it was those with milder forms (cyclothymia) that were most crea- tive. Other researchers have come to the same conclusion, namely that the relationship between creativity and bipolar disorder, schiz- ophrenia, and schizotypy is mostly curvilinear (Acar et al., 2018; Cox & Leon, 1999; Gostoli et al., 2017; LeBoutillier et al., 2014;

Figure 1 Categories of the Arts and Sciences and Prevalence of Mental Illness

Note. Dark Gray: . 70%. Light Gray: . 60% , 70% White: , 60%. Adapted from Ludwig, 1998. See the online article for the color version of this figure.

4 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Ruiter & Johnson, 2015). For instance, Kinney and Richards (2017) reported support for the nonlinear inverted-U hypothesis in which creative thought and behavior were maximum at mild levels of bipolar disorder (cyclothymia) and were relatively low at the low and high ends of the disorder. Moreover, it was first degree relatives of those suffering from bipolar who tended to exhibit highest levels of creativity. After reviewing literature on affective traits and creativity, Feist

proposed a quadratic model of mental health and creativity, that makes clear the complex relationship (cf. Feist, 2012; see Figure 2). The peaks of creativity tend to be with little and moderate levels of psychopathology, with valleys of creativity coming when psycho- pathology is low-medium and again high (cf. Feist, 2012).

Domain of Psychopathology

Not only degree but also domain of pathology matters. In gen- eral, it is clear from research that particular forms of psychopathol- ogy are more associated with high levels of creative achievement than other forms. The milder forms of mood disorders, including depression and bipolar disorder as well as milder forms of psy- chotic disorders (schizotypy), appear to be among the more robust correlates of creative achievement.

Mood Disorders

Of all psychological disorders, perhaps none is more often empiri- cally connected to creativity than mood disorders, especially bipolar depression (and its less severe offshoot, cyclothymia). The general finding is there is an elevated rate of bipolar disorder exists among creative people compared with general population (Andreasen, 1987; Andreasen & Glick, 1988; Fodor & Laird, 2004; Furnham, Batey, Anand, & Manfield, 2008; Gostoli et al., 2017; Jamison, 1996; Jami- son et al., 1980; Johnson et al., 2012; Johnson, Murray, et al., 2015; Johnson, Tharp, et al., 2015; Nowakowska et al., 2005; Ramey & Weisberg, 2004; Richards, 1994; Taylor, 2017). Taylor’s (2017)

meta-analysis of studies published between 1987 and 2014 that examined mood disorders (bipolar, cyclothymia, major depression) in creative samples reported a Hedges g = .64 (95% CI [.45, .82]), meaning creative people are nearly two thirds of a standard deviation higher in mood disorder than noncreative people. When examining simple correlational studies on creativity scales and mood disorder in students and adults, however, Taylor (2017) reported a very small effect (g = .09; 95% CI [.01, .17]). In short, the effect size was mod- erated by level of creativity. Flaherty (2005) reviewed a large range of neuroscientific evidence suggesting that frontal-temporal-limbic brain activity as well as dopaminergic activation are implicated in the relationship between creative drive and mood disorders. More specif- ically, Flaherty proposed a two-dimensional model with frontal-tem- poral activity being on the x axis and dopaminergic activity being on the y axis. As abnormal temporal lobe activity and dopaminergic ac- tivity both increase, mania, psychosis, and creative drive increase. As abnormal frontal activity increases and dopaminergic activity decreases, creative blocks become more likely.

The relationship holds in the other direction too. Other studies have reported higher rates of creativity among bipolar patients (Richards, 1994; Richards et al., 1988; Santosa et al., 2007; Simeo- nova et al., 2005). For example, when compared with healthy con- trols, patients with bipolar disorder scored higher on the Barron- Welsh Art Scale (BWAS) measure of creativity (Santosa et al., 2007; Simeonova et al., 2005). Of note, however, is Taylor’s (2017) finding that people with mood disorder are not necessarily more creative than those without mood disorder (g = .08; 95% CI [�.00, .16]). Kaufmann and Kaufmann (2017) reviewed research on the complex association between mood, mood disorders, and creative thought and behavior. Both positive and negative affect and mood can be associated with creativity.

An important qualification to the bipolar-creativity connection is that it seems to be more mania than depression that is associated with creative thought and behavior (Andreasen & Glick, 1988; Jamison, 1996; Jamison et al., 1980). Given the quickness and

Figure 2 Nonlinear Model of Degree of Creativity and Psychopathology

PSYCHOPATHOLOGY AND CREATIVITY 5

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

fluency of ideas that occur during mania, its relationship with crea- tivity is understandable (Jamison, 1996; Richards, 1994). Moreover, numerous studies support the idea that milder hypomanic states are most clearly correlated to creative thinking and achievement (Furn- ham et al., 2008; Ruiter & Johnson, 2015; Schuldberg, 1990). The relationship between unipolar depression and creativity is

less robust than bipolar. Paek and colleagues (2016) conducted a meta-analysis that included 27 studies that reported results on depression and creativity. These 27 studies consisted of 103 effect sizes on over 14,000 participants. The mean effect size was essen- tially zero (r = .04; 95% CI [�.08, .16]). Silvia and Kimbrel (2010) reported the same very small effects between depression and various forms of creativity in a college student sample. Some research that broke nonclinical depression down into components of rumination, self-reflective pondering, and brooding found that rumination and self-reflective pondering but not brooding were associated with creativity (Verhaeghen et al., 2005, 2014). Note, however, that all of these studies were conducted with students and were little-c creative samples. Similar small effects between trait anxiety and creativity have

been reported (Silvia & Kimbrel, 2010). For example, Paek and col- leagues (2016) also included in their meta-analysis research 32 studies that reported results on anxiety and creativity. These 32 studies consisted of 60 effect sizes on more than 15,000 partici- pants. As was true with depression, the mean effect size between anxiety and creativity was not significantly different from zero (r = �.05; 95% CI [�.16, .06]). Psychotic Disorders

The psychotic disorders—schizophrenia, schizotypy, schizoaf- fective disorder, among others—also have a complex and not easy to summarize relationship with creativity. With anecdotal excep- tions such as John Nash, full blown schizophrenia is seldom linked to creativity (cf. Nasar, 2011; Rothenberg, 1990). Kyaga and col- leagues (2011) reported that people who had first degree relatives suffering from schizophrenia and bipolar disorder were overrepre- sented in creative professions. Moreover, Eysenck (1993, 1995) proposed and found some support for the idea that the nonclinical personality trait and psychoticism is associated with creative thought and behavior. Psychoticism is a nonpathological rather than clinical personality trait consisting of consistent social isola- tion, aloofness, hostility, and unusual thoughts and behaviors. Feist (1998) found support for this idea in a meta-analysis on the personality correlates of creativity. As numerous scholars have pointed out, however, psychoticism

is too broad and diverse a construct to consistently be related to cre- ative thought and behavior (Batey & Furnham, 2008; Carson et al., 2003; Mason et al., 1995; Nettle, 2006). They argue that psychoti- cism’s specific and somewhat more clinical cousin, schizotypy, is more robustly related to creativity. Schizotypy is a personality dis- order in which subclinical symptoms of psychosis are exhibited, such as unusual experiences, magical thought, eccentric behavior, and cognitive disorganization (Claridge et al., 1996). In little-c crea- tive and Big-C creative samples, schizotypy is associated with crea- tive thought (Acar & Runco, 2012; Acar & Sen, 2013; Baas et al., 2016; Batey & Furnham, 2008; Burch et al., 2006; LeBoutillier et al., 2014; Nettle, 2006; Schuldberg, 1990). Baas and colleagues (2016), for instance, argued for a moderation effect by type of

pathology in the association between creativity and mental illness. More specifically, they proposed and found meta-analytic evidence that approach-based pathologies (positive schizotypy and risk of bipolar) were more strongly and positively associated with high lev- els of creativity. Positive schizotypy consists of unusual experiences and impulsive nonconformity, whereas negative schizotypy consists of cognitive disorganization and withdrawn schizoid traits. In addi- tion, avoidance-based pathologies (e.g., anxiety, negative schizo- typy, and depressive mood) were associated with lower levels of creativity. Similarly, Acar and Sen (2013) in a meta-analysis found small negative effect sizes between creativity and negative schizo- typy (r = �.09; 95% CI [�.12, �.06]; k = 76) and a small positive association with positive schizotypy and creativity (r = .14; 95% CI [.12, .17]; k = 121).

The Current Study

The primary purpose of the current study is to update and attempt to replicate the results of Ludwig (1992, 1995, 1998) and to test a more complex model of creativity and psychopathology. Not only is the Ludwig sample itself over 25 years old, but the subjects examined were required to be deceased, further distancing them from their contemporaries. Therefore, an update and exten- sion of the study is now in order. Additionally, the professional categories proposed in Ludwig (1992, 1995, 1998) required reworking in the current study. For instance, several of the profes- sions listed under social sciences, such as historian and philoso- pher, are not actually sciences at all and are frequently grouped with humanities. The current study also improves on the previous methodology, which was vulnerable to researcher bias owing to the investigator’s awareness of the hypotheses (Ludwig, 1992). Another goal of the current study is to see whether Ludwig’s find- ings from 20 years ago and with a different sample still hold and replicate in a more current sample. More importantly, Ludwig did not compare eminent famous creative people against eminent fa- mous noncreative people, and simply analyzed his data with chi- squares for inequalities between groups. We decided, therefore, to hold fame constant in our comparison group of famous athletes (with published biographies) to determine whether fame more than creativity could explain the presence or absence of mental illness in our sample. In short, by having a control group that was eminent and famous (i.e., biography-worthy) and yet not creative, we could rule out pure fame and eminence as a confounding explanation in any relationship between creativity and psychopathology.

Hypotheses

1. World-class creative artists will have elevated rates of any lifetime mental illness relative to creative scientists and famous athletes (controls).

2. The mental health difference between artists and scientists will be most pronounced on mood/affective disorders (anxiety, depression, bipolar) and chemical dependency, with artists expected to have higher rates than scientists.

3. Creative scientists should show elevated rates of being on the high functioning end of the autism spectrum relative to athlete controls.

6 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Method

Subjects and Materials

The list of potential creative and eminent scientists, artists, and athletes for inclusion in the study came from rankings in diction- aries, encyclopedias, and best-of lists. The original list, after merg- ing 163 source lists and removing duplicates, contained 17,689 distinct names. Each “best-of” source list was ranked by a member of the research team (D.D.) on a 3-point scale for trustworthiness, with a 1 being of questionable validity, 2 being more subjective, and a 3 being very trustworthy. The primary criterion for a code of 3 was whether the list was created by experts in the field and/or was of international award such as the Nobel Prize. A code of 2 was awarded if the list were created by professionals in the field (e.g., a poll among more than a hundred contemporary leading physicists conducted by Physics World magazine, Dunani & Rodgers, 1999), whereas a code of 1 was awarded for lists made by amateurs or based on unclear methodology. One such example of an unclear methodology is the “Greatest Mathematicians of All Time” list published on server thetoptens.com.1

Next, an index of eminence was calculated for each potential subject within each domain as a sum of trustworthiness scores of all lists in the domain containing the subject’s name. To prevent overlap with Ludwig’s (1992) sample, subjects must have either died after 1950 or been born before 1980, if they were still alive. The 45 most eminent professionals in each domain were selected as potential subjects in the sample. Individuals who tied for the 45th most eminent position were included in the sample. This pro- cedure led to a total of 766 potential subjects. Professionals in multiple domains were sorted in the category in which they ranked in a higher position. After the list of potential subjects was obtained, the next step

was to determine whether or not a viable and relevant biography was written about that person. When available, e-versions (Kindle) of biographies were purchased. If no e-version was available, hard copies were purchased, had their bindings removed, and were digi- tally scanned via optical character recognition (OCR) conversion. To be selected for study, biographies had to be written for an adult audience and include information on the creator’s personal life and were not solely intellectual or work biographies. Moreover, autobiographies, biographies written by close relatives, biographi- cal chapters, letters, and memoires were excluded. Of the 766 potential subjects, 391 did not have appropriate biographies writ- ten about them, leaving a potential sample of 375. If there were more than one biography written about a person, we chose the one that had the most life-history information. Owing to time and resource constraints, 199 of the 375 biographies were purchased for coding (18% female).2 Analyses revealed that the 199 biogra- phies were representative of the larger 375 sample on proportion of artists, scientists, and athletes as well as proportion of deceased subjects. The proportion of women however increased from 13% to 17.5% in the final sample of 199, v2(1) = 9.45, p = .002. Never- theless, the proportion of women in each subgroup did not change, v2(2) = 1.34, p = .510. In the final sample, there was a higher per- centage of women in the arts than sciences or sports, 28%, 7% and 4% respectively, v2(2) = 16.11, p , .01. The overall sample was 83% White-Caucasian, 13% Black/African American, 2% Latinx,

and 1.5% Asian-Pacific Islander. In 2016 when the data were col- lected, the U.S. demographics were 60% White-Caucasian, 18.5% Latinx, 13% Black/African American, 6% Asian American (U.S. Census Bureau, 2019).

Each subject was placed into either a scientific, artistic, or ath- letic domain. Scientific domains were defined as technology/ invention, mathematics, physics, chemistry, biology/medicine, psychology, and social sciences (anthropology and sociology). Earth scientists (e.g., geologists, oceanographers, climatologists) were excluded because of a lack of biographies. Artistic domains were defined as visual arts, fiction writing, poetry, acting, musical performance, and musical composition. Using these career group- ings, the current sample consisted of 104 artists, 68 scientists, and 27 athletes (see Table 1). The entire sample had a median year of birth of 1919 with an average birth year of 1921. The range of birth years was 1873 to 1979. Of the 199 subjects in the final list, 46 (23%) were alive at end of data collection in 2016. For the 153 participants who had died, the average age of death was 72 (me- dian = 75; mode = 81; range 25 to 98). For the 46 participants who were alive, the average age was 71.70 (median = 73; mode = 75; range 41 to 89). Ninety-three percent of the sample was married at least once (mean age of first marriage = 27.13). There was no ca- reer domain difference in mean age of first marriage. Artists (59%) were more likely to have divorced than scientists (38%).

Procedure

Biography Selection and Preparation

After the subject-pool was narrowed down to subjects who had usable biographies written about them, we purchased each biogra- phy either in digital or bound format. If the book was in bound for- mat, we then detached its binding and scanned the entire body of the text (excluding front- and rear-matter) into readable ORC/digital format. Next, we cleaned the digital books by removing all images, headings, footers, foot notes and most tables and equations.

Pathology Selection

Before ratings of pathologies could be made, the research team discussed and decided which specific diagnostic illnesses would be coded. For this process we mostly followed Ludwig by obtain- ing the original variable list. Ludwig’s team coded mental health status of immediate family members as well as the creative person. We coded only the creative person. Moreover, we also added a few illness categories that we believed Ludwig missed and might be relevant as exploratory analyses, such as Asperger’s syndrome (high functioning autism) and synesthesia. The final list consisted of 19 diagnostic illnesses listed in the Diagnostic and Statistical Manual of Mental Disorders, 5th edition (DSM–5; American Psy- chiatric Association, 2013; see Table 2).

Paragraph Selection

After each of the 199 biographies was scanned and cleaned, a linguistic analysis program was created by the second author to

1 For complete Best of Lists, Awards, and Rankings, see https://doi.org/

10.17605/OSF.IO/TFYDK. 2 Raw data are posted on Open Science Framework at https://doi.org/10

.17605/OSF.IO/TFYDK.

PSYCHOPATHOLOGY AND CREATIVITY 7

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

automatically locate and highlight any of the 175 relevant key- words related to 19 mental illness categories (see Appendix A in the online supplemental materials). The initial list was based on words used by Ludwig (1995) but was expanded through a discus- sion between the investigators after a review of the DSM–5 and Stein and colleagues (2010). From either the biographical or pub- lic websites we obtained the following demographic variables: profession/career, date of birth, date of death (if deceased), year of mother’s death, year of father’s death, birth order, race/ethnicity, gender, year of marriage (first), year of marriage (second), country of birth. Two trained graduate students further narrowed the biographical

texts to include only paragraphs relevant for assessment of mental health of subject in question. For example, “depression” may have been tagged by the program, but if it referred to the economic period of the 1930s then that paragraph would be excluded from further rat- ing. Similarly, if key terms were tagged but referred to someone other than the target creator, those paragraphs were also de-selected for further rating.

Pathology Ratings

In the fifth and final step, seven raters were selected and trained to identify possible psychopathologies in each biography excerpt. Raters could only begin once they met the .80 interrater reliability with training data from Ludwig’s (1992, 1995). Poten- tial raters were given paragraphs selected from a biography then asked to code the given reading material for the psychopatholo- gies described above. Ratings were compared against the origi- nal coding data from Ludwig’s study. Interrater reliability was measured using Gwet’s agreement coefficient AC1 (Gwet, 2008), which was preferred over Cohen's kappa coefficient as it pro- vides unbiased estimate even in case of strongly uneven occur- rences of categories (Gwet, 2002; Wongpakaran, Wongpakaran, Wedding, & Gwet, 2013). To keep raters blind and free of any previous bias, the name of the subject in question was replaced with the word “Creator” in all biography excerpts. Subjects were coded for lifetime prevalence of any of 19 psychopathologies.

Psychopathologies were rated on a 3-point scale of not present (0), probable (1), and present (2) if they occurred at any point in the creator’s lifetime. Rating present was used in cases where DSM–5 criteria were clearly met or where the subject was diag- nosed professionally during their lifetime. If not enough infor- mation was given in the biography to provide a clear diagnosis from DSM–5 criteria and yet there was some evidence that a dis- order was suspected, then raters gave that a probable rating. In short, a probable rating was provided whenever there was some but not overwhelming evidence of a disorder. Present was pro- vided when a professional diagnosis was made during the per- son’s lifetime or when the biographical evidence was very clear. During training on the Ludwig sample, if raters initially fell below the .80 reliability criterion, research meetings were held with other raters and the lead researcher (G.J.F.) to discuss dis- crepancies and to reach consensus. For final nontraining ratings, two independent and randomly assigned raters coded each biog- raphy. Any disagreement was adjudicated by a third rater (G.J. F.) to establish the final rating.

Results

Previous and Current Lifetime Rates of Disorders

For sake of comparison, in Table 3 we present the population estimates of lifetime rates of psychological disorders published in the literature. Two studies have reported large-scale national popu- lation estimates of lifetime rates of any disorder. Kessler, Berglund, and colleagues (2005) reported a rate of 46.4% and Lev-Ran and colleagues (2013) reported a rate of 33.7% for any mood, anxiety, personality or psychotic disorder. In this context, the lifetime rate in our sample for creative artists and scientists was 49.7% (85 of 171) and for athletes was 48.1% (13 of 27), v2(2) = 2.39, ns. For artists only, the percentage of “present” cases was 61/103 (59.2%), and for scientists it was 24/68 (35.3%). Over the course of their lifetime, artists were more likely to have at least one form of psychopathol- ogy than scientists, v22(1) = 9.38, p = .002. Artists were not more likely to have a lifetime bout of psychopathology compared with athletes (59% versus 48%, respectively) , v2(1) = 1.07, ns.

When the less exclusive “probable” cases were also included, the frequency of psychopathology increased to 126 of 171 (73.7%) in the creative groups compared with 16 of 27 (59.3%) athletes, v2(1) = 2.39, ns. The observed frequencies on “probable” lifetime psychopa- thology for artists (83%), scientists (59%), and athletes (59%) were

Table 1 Specific Domains and group Sizes

Domain % White N of Men N of Women Total N

Artists 75 29 104 Visual arts 100.0 8 1 9 Fiction writing 85.7 26 9 35 Poetry writing 90.9 9 4 11 Acting 95.0 11 9 20 Music performance 46.2 18 8 26 Music composition 100.0 3 0 3

STEM 63 5 68 Technology/Invention 100.0 9 1 10 Mathematics 100.0 7 0 7 Physics 100.0 19 0 19 Biology/Medicine 100.0 6 1 7 Chemistry 100.0 4 1 5 Psychology 100.0 11 0 11 Social Sciences 100.0 7 2 9

Comparison group 26 1 27 Athletes 55.6 26 1 27

Total 164 35 199

Table 2 List of Rated Psychopathologies

Rated psychopathologies

Adjustment disorder Obsessive-compulsive disorder Alcoholism Paraphilia Anxiety disorder Personality disorder (of any kind) Autism spectrum disorder Posttraumatic stress disorder Conduct disorder Schizophrenia/Psychotic disorders Depression/Depressive disorder Sleep disorders Drug use/dependency Somatic disorder Eating disorder Suicide/Suicide attempt Gambling disorder Synesthesia Kleptomania

8 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

different from chance, v2(2) = 14.68, p = .001. This effects stems from artists having a higher lifetime rate than both scientists, v2(1) = 12.86, p , .001, and athletes, v2(1) = 7.45, p = .006. Even though the overall frequency of any psychopathology was

extraordinarily high, relative frequencies of individual disorders rarely exceeded a few percent. Occurrences higher than 5% were only observed in depression/depressive disorders (26%), alcoholism (16%), drug use/dependency (12%), and anxiety disorder (11%). These compare with the rates of 28.8% for anxiety, 20.8% for mood, and 14.6% for substance abuse in the general population (Kessler et al., 2007; Kessler, Berglund, et al., 2005). For a more sophisticated analyses of these trends, for each listed

psychopathology, a null hypothesis about the uniform distribution of its occurrences in compared groups was tested. In each comparison, a chi-squared test with a Monte Carlo simulated p value was used because it has no assumptions about minimal expected frequencies (estimates were done with 106 replicates; Hope, 1968). We have included both present and probable occurrences of psychopathology without differentiating between them into analysis. Observed frequen- cies for each category and resulting p values are detailed in Table 4. The results suggest that there are unequal frequencies between the do- main of eminent individual and the occurrence of Alcoholism, Drug use/dependency, Gambling disorder, Suicide/suicide attempt and also Depression/depressive disorders, Anxiety disorder, Sleep disorder and Autism spectrum disorder. Note that there were 21 significance tests performed, which results in substantial increase in familywise first type error rate. To keep familywise error rate under 5% Bonferroni corrected p-values can be used. This correction suggests that the fre- quencies in Autism spectrum disorder (pBonf. = .271), Suicide/suicide attempt (pBonf. = . 282), and Gambling disorder (pBonf. = .169) are not different from expected, so we should not consider those results as

conclusive. Finally, 52% of the artists and 24% scientists experienced the loss of a parent in childhood, v2(1) = 15.04, p , .001.

Planned Analyses

Recall, the three main predictions were that compared with athletes and scientists, artists as a whole would have elevated lifetime rates of psychopathology, especially in the mood disorders and substance abuse categories. We also predicted that scientists would have ele- vated rates of high functioning autism compared with athlete controls.

To test these hypotheses, a Bayesian logistic regression3 was used to model the relationship between mental disorder presence and creativity domain. The dependent variable was an occurrence of particular mental disorder. We have used two separate models with differently defined dependent variables for each disorder. In the first one, the present cases only were coded as 1. In the second one, both present and probable cases were coded as 1. The values of the dependent variable were predicted with categorical factor

Table 3 Published Population Estimates of Lifetime Rates of Psychological Disorders

Author(s), Date Disorder Percent of population

Baca-Garcia et al. (2010) Suicide attempted: Male 1.75% Suicide attempted: Female 2.95% Suicide ideation: Male 6.00% Suicide ideation: Female 7.60%

Hudson et al. (2007) Eating disorder: Male 2.80% Eating disorder: Female 5.90%

Lev-Ran et al. (2013) Any mood, anxiety, psychotic, personality disorder 33.7% Kessler, Berglaund, et al. (2005) Anxiety disorder 28.80%

Mood disorder 20.80% Impulse-control disorder 24.80% Substance abuse 14.60% Any disorder 46.40%

Merikangas et al. (2007) Bipolar I 1.00% Bipolar II 1.10% Subthreshold bipolar 2.40%

Nock and Kessler (2006) Suicide (ideation/attempt) 2.70% Perälä et al. (2007) Psychotic disorder (any kind) 2.29% Robins et al. (1984) Any disorder 28.8% to 38.0%

Anxiety disorder 10.4% to 25.1% Substance abuse disorder 15.0% to 18.1% Affect-mood disorder 6.1% to 9.5% Psychotic disorder 1.1% to 2.0% Personality disorder 2.1% to 3.3% Eating disorder 0.0% to 0.1%

Note. Robins et al. (1984), consisted of three samples from New Haven, CT, Baltimore, MD, and St. Louis, MO, and interviews were conducted between 1980 and 1982. Bolded text highlights the overall “any disorder” category.

3 Computations were performed in statistical program R with the

rstanarm package (Stan Development Team, 2016). The student t- distribution with 7 degrees of freedom and a scale parameter 2.5 was chosen as a prior for all regression weights. The student distribution was preferred from the normal distribution because its heavy-tailedness enables substantial differences from expected value. The location was set to zero in all parameters with the exception of the intercepts in which case logits of population prevalences according to Kessler, Berglund, and colleagues (2005) and Nock and Kessler (2006) were used. The parameter estimations were computed with NUTS sampling method (Hoffman & Gelman, 2014) on 8 MCMC chains each performing 6,000 iterations (2,000 in burning phase). The e-values in favor of null hypothesis stating that odds ratio equals one were computed using fbst package (Kelter, 2020). Above mentioned priors were used as the reference functions. For details see Pereira and Stern (2020).

PSYCHOPATHOLOGY AND CREATIVITY 9

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

domain with seven levels listed in Table 1. Dummy variable cod- ing was used with the group of athletes used as a reference group. Besides the categorical independent variable, the gender and two quantitative covariates were included in the model to hold them constant: year of birth of each subject and the length of their life. If any biography was published during the life of the subject, the age of the subject in the year of publication was used instead of age of their death, as the biography author did not have informa- tion about the subjects’ lives from that point. Both quantitative covariates were centered and scaled to the z-score format. The hypotheses about model parameters were tested with Full

Bayesian Significance test (FBST; Pereira & Stern, 1999). This pro- cedure uses e value (evidence value) as a measure of statistical signif- icance (Pereira & Stern, 2020). In its simplest form, e-value in favor of the null hypothesis stating parameter equals zero is close to the idea proposed by Thulin (2014): what is the maximal value of a resulting in 1 � a posterior credible interval not containing zero. Keeping in mind this analogy between p values/confidence intervals and e values/credible intervals, we also used value .05 for rejection of the null hypothesis. Figure 3 shows the prevalence interval estimates (95% highest

density regions) of the selected present or probable psychopatholo- gies. Each creative domain was also compared with the control group of athletes. The odds ratios of psychopathology occurrence in each group compared with the control are represented in the figures as text labels. The asterisk notation indicates e values lower than .05, .01, and .001, respectively. Dot indicates nonsignificant result with e value lower than .1. Figure 3 reports that artists were more than 2.71 times more likely to experience any mental illness over the course of their life compared with eminent athletes, whereas

scientists had approximately the same odds as athletes. Artists were most at risk for depression/bipolar (ORs = 6.28 and 9.19 for present and present/probable, respectively) compared with controls. The only disorder for which scientists were at elevated risk was the more inclusive present/probable rate for depression/bipolar (OR = 4.53). For all other disorders, scientists were either equally likely or less likely to experience them compared with athlete controls. For exam- ple, creative scientists were significantly less likely (ORs = .21 - .23) to be afflicted with substance related and addictive disorders than athletes.

In Figure 4 we present results broken down by subgroups. As is evident in Figure 4, the elevated risk of mental illness in the artist group primarily is a result of writers and visual artists. Looking at only “present” rates of specific disorders, writers and visual artists were 4.52 times more likely than athletes to suffer from any pres- ent mental illness over the course of their lifetime. In fact, no other subgroup was more or less at risk compared with athlete controls. In addition, it was primarily the writers and visual artists who were at increased risk of depression/bipolar (OR = 8.11), anxiety (OR = 7.53), and suicide/attempt (OR = 15.79).

When we expand the analysis to include “probable” or “pres- ent” rates, results were much the same except for depression and substance abuse. Musicians, actors, writers, visual artists, poets, mathematicians, and technologists were all more likely to be sus- pected of having depression and/or bipolar compared with fa- mous athletes (ORs = 3.82 – 10.86). Visual artists, writers and poets were more likely to suffer anxiety disorder (OR = 3.58) and be suicidal or attempt suicide (OR = 6.18) than athletes. Moreover, physicists and chemists were less likely than athletes to suffer substance related and addictive disorders (OR = .04).

Table 4 Sample Frequencies of Lifetime Rates of Psychopathology Across Professional Domains

Disorder Athletes (n = 26)

Biologists (n = 7)

Math. and Technol. (n = 17)

Musicians and Actors (n = 50)

Physicists and Chemists

(n = 24)

Psych. & Social Science (n = 21)

Visual Artists, Writers, and

Poets (n = 54) p

Synesthesia 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.2) 0.0 (0.0) 0.0 (0.0) 18.5 (18.5) 1.00 Adjustment disorder 0.0 (0.0) 0.0 (0.0) 0.0 (17.6) 0.0 (0.4) 0.0 (4.2) 0.0 (0.0) 0.0 (7.4) .152 Alcoholism 7.7 (11.5) 0.0 (0.0) 11.8 (11.8) 20.0 (14) 0.0 (4.2) 9.5 (9.5) 27.8 (38.9) .002 Drug use/dependency 7.7 (7.7) 0.0 (0.0) 5.9 (11.8) 26.0 (16) 0.0 (0.0) 4.8 (4.8) 11.1 (11.1) .002 Depressive disorders 7.7 (7.7) 14.3 (28.6) 23.5 (41.2) 20.0 (19) 20.8 (41.7) 23.8 (42.9) 48.1 (64.8) ,.001 Bipolar disorder 7.7 (7.7) 0.0 (0.0) 5.9 (11.8) 0.0 (0.4) 4.2 (4.2) 0.0 (0.0) 18.5 (5.6) .770 Anxiety disorder 7.7 (11.5) 0.0 (0.0) 11.8 (17.6) 0.8 (24.0) 8.3 (20.8) 0.0 (14.3) 13 (38.9) .046* Obsessive compulsive 0.0 (0.0) 0.0 (0.0) 0.0 (11.8) 0.2 (0.6) 0.0 (4.2) 0.0 (0.0) 0.0 (5.6) .526 Schizophrenia 7.7 (7.7) 0.0 (0.0) 5.9 (5.9) 0.0 (0.2) 0.0 (0.0) 0.0 (4.8) 18.5 (18.5) .909 Somatic disorder 7.7 (11.5) 0.0 (0.0) 0.0 (5.9) 0.0 (0.4) 0.0 (0.0) 0.0 (4.8) 18.5 (3.7) .587 Autism spectrum disorder 0.0 (0.0) 0.0 (0.0) 0.0 (5.9) 0.0 (0.0) 0.0 (12.5) 0.0 (0.0) 0.0 (1.8) .038* Suicide/suicide attempt 0.0 (0.0) 0.0 (0.0) 0.0 (5.9) 0.0 (0.4) 4.2 (12.5) 0.0 (0.0) 7.4 (11.1) .227 Sleep disorder 7.7 (15.4) 0.0 (14.3) 0.0 (5.9) 0.4 (1.2) 0.0 (0.0) 0.0 (0.0) 3.7 (16.7) .185 Eating disorder 0.0 (0.0) 0.0 (0.0) 0.0 (5.9) 0.0 (0.6) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) .242 Personality disorder 7.7 (7.7) 0.0 (0.0) 5.9 (11.8) 0.6 (7) 8.3 (16.7) 0.0 (4.8) 0.0 (3.7) .261 Gambling disorder 15.4 (19.2) 0.0 (0.0) 0.0 (0.0) 0.2 (0.2) 0.0 (0.0) 0.0 (0.0) 3.7 (3.7) .008* Conduct disorder 7.7 (7.7) 0.0 (0.0) 0.0 (0.0) 0.2 (0.4) 4.2 (16.7) 0.0 (0.0) 3.7 (3.7) .106 Kleptomania 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (18.5) 1.00 Posttraumatic stress 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.2) 0.0 (0.0) 0.0 (0.0) 18.5 (3.7) .852 Paraphilia 0.0 (0.0) 0.0 (0.0) 0.0 (0.0) 0.0 (0.2) 0.0 (0.0) 0.0 (0.0) 3.7 (3.7) .852 Any psychopathology 46.1 (57.7) 14.3 (28.6) 52.9 (70.6) 52.0 (78.0) 33.3 (62.5) 33.3 (57.1) 66.7 (87.0) .003*

Note. Percentages outside brackets involve present cases of psychopathology occurrence. Percentages inside the brackets include both present and proba- ble cases. Column p contains Monte Carlo simulated p values testing null hypothesis that relative frequencies of present or probable psychopathology occurrences are uniformly distributed across compared domains. *pBonf . .05.

10 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

There were no differences between the creative groups and ath- letes in sleep disorders.

Discussion

The intention of the current study was to examine the more com- plicated and moderated relationships between creativity and psycho- pathology by updating and replicating Ludwig’s (1992) biographical analysis of world-class creative artists and scientists. Our expectation was that artistic creative professions in general would possess higher levels of psychopathology than creative scientists. We also predicted that scientists would not differ from the base rates of psychopathol- ogy found in the U.S. population, whereas artists would. The current study controlled for researcher bias by removing

the biographical material of its subjects’ identities. The previous study conducted by Ludwig (1992) was executed with the researcher knowing the identity of each subject, and may have been biased by previous working knowledge of each professional. Certain professions that were given new classifications as the older categorizations, as designated in Ludwig (1992, 1995), may have been incorrectly assigned. For example, historians and philoso- phers were considered scientists by Ludwig. Although history and

philosophy are scholarly subjects, they are not typically considered sciences.

The current study also sought to streamline the process of finding relevant information in books by digitizing each biography into a searchable digital media. This would allow for the researchers to oper- ate at an increased pace by eliminating irrelevant text very quickly. Transforming each book into a digital format also made it possible to censor the names of each creator to limit any previous knowledge that could bias the rating group.

The results of the current study generally corroborated the findings reported in Ludwig’s and other studies, lending further support to previously established hypotheses. Despite using an entirely new set of subjects, not included in Ludwig’s (1992) sample, artists still pos- sessed higher rates of psychopathological traits than scientists, ath- letes, and the U.S. population in general. Scientists were consistently rated lower on symptoms of psychopathology than artists, despite equal eminence. These results held true in both inclusive and exclu- sive requirements for classification into the mentally ill group. How- ever, the difference between artists and athletes was not significant in the more exclusive interpretation of the data. Rates of drug abuse and anxiety also differed between artists, scientists, and athletes depend- ing on whether inclusive or exclusive criterion were used to define what constituted psychopathology. In both cases, fewer subjects

Figure 3 Prevalence of Selected Mental Disorders (Present and/or Probable) in Creative Artists and Scientists Compared With Eminent Athletes

PSYCHOPATHOLOGY AND CREATIVITY 11

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Figure 4 Prevalence of Selected Mental Disorders (Present and/or Probable) in Creative Domains and Athletes

Note. Text labels indicate odds ratios for psychopathology occurrence in creative domains compared to the athletes. The aster- isks indicate whether the 1-a centered highest density region for posterior density of regression weight contains zero (no effect) when a equals 5% (*), 1% (**), or 0.1% (***).

12 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

qualified for inclusion into the mentally ill group when exclusive cri- teria were used. However, the differences between groups grew larger in the case of anxiety and smaller in the case of drug use, thus moderating the results. Artists also showed greater rates of alcoholism, drug abuse,

depression, and OCD than those found in the U.S. population. Again, statistical significance changed for a few of psychopathologies depending on the strictness of criterion for inclusion. Anxiety among artists was considered lower in the stricter assessment but still gained significance due to the high rate of anxiety reported in the U.S. popu- lation. Rates of OCD also fell for artists and scientists under stricter criterion and detectable differences were no longer found. Because we had a small comparison sample of famous but not

professionally creative athletes, we could also address the question of whether fame itself—isolated from creativity—is a contributing factor to psychopathology. Because athletes were generally less likely than artists and even scientists to develop psychopathology at some point during their lives, we can tentatively conclude that fame per SE is not the driving force behind psychopathology. There was one exception to this, namely anxiety. Athletes exhib- ited higher lifetime rates of anxiety disorders than the general pop- ulation. An interesting question therefore becomes “Do higher rates of anxiety precede or follow athletic fame?” That is, is anxi- ety a cause or effect of athletic eminence? Assuming the relationship between some forms of creativity

and some forms of pathology are robust and real, then the ques- tion becomes why might these two traits covary? Recent litera- ture from biological and evolutionary approaches have suggested biological bases and even potentially adaptive func- tions of the relationship between creative thought and behavior and psychopathology. For instance, research has reported a pol- ymorophism of a particular gene involved in psychosis that is associated with high levels of creativity and high IQ (Kéri, 2009). The gene in question is neuregulin 1, which is a candi- date gene for psychosis and affects neuroplasticity, glial func- tion, and neuronal development in general. One form of the gene, the T/T genotype, was related to both high creativity and risk for psychosis. Based on this and other evidence, Kozbalt and colleagues (2017) argued that one possible reason why a maladaptive trait may still exist in humans is its shared genetic linkage with creative behavior (cf. Akiskal & Akiskal, 2007; Greenwood, 2020; Nettle, 2001, 2006; Power et al., 2015). Sim- ilarly, other empirical and theoretical evidence supports the idea that the milder levels of mental illness, for example cyclothy- mia, confer advantages such as increase fluency of ideas that make creative thought more likely (Carson, 2011; 2014; Green- wood, 2020). Carson (2011, 2014), in fact, proposed that the “shared vulnerability” traits of openness, impulsivity, schizo- typy, cognitive disinhibition, hypomania, and cyclothymia are the traits that connect psychopathology (risk factors) and crea- tive (protective factors) behavior. In sum, there are various neu- rological, evolutionary, and adaptive factors that may undergird the associations between some forms and degrees of creativity and some forms and degrees of psychopathology.

Caveats and Limitations

A number of confounding variables limit the results of this study. One such limitation is sample bias. In the case of the current

study, writers and publishers may be more inclined to pursue biog- raphies for particularly interesting people to tell more compelling stories. Because someone with a history of psychopathology may serve as a more desirable subject for a biography than someone who is not, healthier professionals may have fewer books written about them. Indeed, the study also contained a much smaller num- ber of scientists than artists, which may be due to writers and pub- lishers favoring more artists rather than scientists since the latter may be perceived as less interesting or hold less recognition in the general public.

The level of fame could not be held constant through all three groups. Although some scientists such as Stephen Hawking and Richard Feynman are particularly well known, not all eminent sci- entists are easily recognizable to the public (e.g., Alfred Tarski, Grigori Perelman, George Beadle). Most of the actors (e.g., Mar- lon Brando, Robert DeNiro, Sophia Loren) and musicians (e.g., Ella Fitzgerald, Prince, Diana Ross, Johnny Cash) were well known to the general public. Athletes, although more recognizable than scientists, tend to dwindle in fame after retirement. Because the careers of most athletes are particularly short, their highest point of fame tends to come earlier in their lives rather than later. This is incongruent with scientists as fame for their achievements tend to come later, after their work has been recognized. Both ath- letes and scientists may also possess lower levels of fame than per- formance artists such as musicians and actors.

Another limitation of the current study is the gender imbalance in the creative sample. The biographies of men in the sample out- numbered women 164 to 35 (18% female). There are historically fewer biographies written about women than men, especially the sciences. Moreover, women have been less likely to reach the highest levels of their professions, whether they have biographies written about them or not—the famous “glass ceiling.” For exam- ple, women have historically been seriously underrepresented in mathematics and sciences. Only 8.8% 15.8% of tenure-track posi- tions among top universities are held by women in math-centric domains (Ceci & Williams, 2011), and only 20% of physics PhDs were awarded to women as recently as 2017 (Porter & Ivie, 2019). During the time that many eminent people in this sample were most active professionally—the 1950s to 1980s—the percentage of women earning PhDs in physics was between 3% to 7% (Porter & Ivie, 2019). Yet not all professions are so imbalanced. For example, our sample had five women of 68 (7%) in the STEM dis- ciplines but 29 of 104 (28%) women in the arts. In acting, the cur- rent sample had a ratio of nine of 20 (45%), and in musical performance it was eight of 26 (31%).

This finding begs the question of why are women underrepre- sented in certain fields more than others and in particular at the top of their fields? There is an extensive scientific literature on this question that goes well-beyond the scope of this article (Cheung & Halpern, 2010). Suffice it to say that social and cultural biases about marriage, child rearing, and performance play a very large role in “glass ceiling” effect (Cheung & Halpern, 2010). For example, in a nationwide study, biology, chemistry, and physics professors were found to consider men as both more hirable and competent (Moss-Racusin et al., 2012).

Similarly, the sample was skewed racially, with being of White-European ancestry (83% of our sample, compared with 60% in the U.S. population). Indeed, 100% of the Science-Tech- nology group was White. Part of this bias comes from the bias in

PSYCHOPATHOLOGY AND CREATIVITY 13

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

published biographies, especially during the time frame of our study. Moreover, basing research on biographies will inherently require older samples given the delay between creative accom- plishments and publication of a biography. Future research will need to continue to determine whether these trends hold with more ethnically diverse samples. Additionally, determining how to interpret historical and bio-

graphical texts is a challenge for psychological study (Citlak, 2016; Czechowski et al., 2016). Biographies still require interpre- tation as historians of certain types of professions may differ from others in what report. Some professions may encourage exagger- ated stories, particularly of drug use, to sell their fame (Lucijani�c et al., 2010). Musicians such as rappers and rock stars may benefit from rumors of psychopathology as increased notoriety would increase exposure and thus raise the likelihood of album sales. Sci- entists do not typically benefit from fame in the same way artists do, as they typically work to discover new knowledge rather than sell products or develop a fan-base, thus there is less incentive to exaggerate claims of illness or drug use. Furthermore, we must acknowledge the fact that different disor-

ders are easier to rate from biographies than others and are more likely to show up in biographies than others. The former consists of more behaviorally expressed disorders such as drug or alcohol addiction, violence, suicide, and depression, whereas the latter con- sists of more internalized or private disorders, such as PTSD, sleep disorders or even more moderate degrees of anxiety disorders. We also need to make clear that our raters were not licensed

clinical psychologists but rather trained undergraduate research assistants. To be sure, they had to go through a reliability training process that involved learning the DSM–5 criteria for the 19 disor- ders and they could not begin rating until they obtained the .80 interrater reliability threshold. Finally, the raters were blind to the subject of the biography, and we had two independent raters code each biography. Nevertheless, these are not assessments by li- censed clinical psychologists.

Future Directions

We make little claim that this investigation settles the “debate” over the “mad-genius.” At best, it confirms one aspect of it, namely the higher rate of pathology and the different pathologies in the creative arts than other creative domains. Many questions remain. For example, as we mentioned above, the biggest question left unresolved is the gender question. Are these patterns that we found in a heavily male-dominated sample the same in famous creative women? Our dataset does not allow this question to be satisfactorily answered. Moreover, because of the restricted sam- ple size, certain analyses were not possible among smaller groups and specific professions. Additionally, no comparisons could be made for fiction writers against nonfiction writers, limiting the conclusions that could be made. Thus, more specific examinations of individual professions can be made as the dataset grows larger. Additional demographic variables that may influence professional vocation and creative output will also be collected. These variables include birth order, religious affiliation, ethnicity, and marital sta- tus of parents. Owing to time constraints, the collection of these data lay beyond the scope of the current study. In conclusion, the results of this study provide support and repli-

cation for the findings of previous biographical investigations of

highly creative people. The use of digital resources allowed for the researchers to limit bias through the use of censors to hide the identity of each creator. The classification and grouping of each profession were also reworked for further accuracy. As is true of all research, however, for each question answered, others arise and await further attention from future investigators. The topic of psy- chopathology and world-class creative achievement is a rich and complex topic and will provide material for researchers for years to come.

References

Abraham, A. (2017). Neurocognitive mechanisms underlying creative thinking: Indications from studies of mental illness. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 79–101). Cambridge University Press.

Acar, S., Chen, X., & Cayirdag, N. (2018). Schizophrenia and creativity: A meta-analytic review. Schizophrenia Research, 195, 23–31. https://doi .org/10.1016/j.schres.2017.08.036

Acar, S., & Runco, M. A. (2012). Psychoticism and creativity: A meta-ana- lytic review. Psychology of Aesthetics, Creativity, and the Arts, 6(4), 341–350. https://doi.org/10.1037/a0027497

Acar, S., & Sen, S. (2013). A multilevel meta-analysis of the relationship between creativity and schizotypy. Psychology of Aesthetics, Creativity, and the Arts, 7(3), 214–228. https://doi.org/10.1037/a0031975

Akinola, M., & Mendes, W. B. (2008). The dark side of creativity: Biolog- ical vulnerability and negative emotions lead to greater artistic creativ- ity. Personality and Social Psychology Bulletin, 34(12), 1677–1686. https://doi.org/10.1177/0146167208323933

Akiskal, H. S., & Akiskal, K. K. (2007). In search of Aristotle: Temperament, human nature, melancholia, creativity and eminence. Journal of Affective Disorders, 100(1–3), 1–6. https://doi.org/10.1016/j.jad.2007.04.013

Amabile, T. M. (1996). Creativity in context: Update to the social psychol- ogy of creativity. Westview Press.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.).

Andreasen, N. C. (1987). Creativity and mental illness: Prevalence rates in writers and their first-degree relatives. The American Journal of Psychi- atry, 144(10), 1288–1292. https://doi.org/10.1176/ajp.144.10.1288

Andreasen, N. C. (1978). Creativity and psychiatric illness. Psychiatric Annals, 8(3), 23–45. https://doi.org/10.3928/0048-5713-19780301-05

Andreasen, N. C., & Glick, I. D. (1988). Bipolar affective disorder and cre- ativity: Implications and clinical management. Comprehensive Psychia- try, 29(3), 207–217. https://doi.org/10.1016/0010-440X(88)90044-2

Baas, M., Nijstad, B. A., Boot, N. C., & De Dreu, C. K. W. (2016). Mad genius revisited: Vulnerability to psychopathology, biobehavioral approach-avoidance, and creativity. Psychological Bulletin, 142(6), 668–692. https://doi.org/10.1037/bul0000049

Baca-Garcia, E., Perez-Rodriguez, M. M., Keyes, K. M., Oquendo, M. A., Hasin, D. S., Grant, B. F., & Blanco, C. (2010). Suicidal ideation and suicide attempts in the United States: 1991-1992 and 2001-2002. Molec- ular Psychiatry, 15(3), 250–259. https://doi.org/10.1038/mp.2008.98

Baron-Cohen, S., Wheelwright, S., Burtenshaw, A., & Hobson, E. (2007). Mathematical talent is linked to autism. Human Nature, 18(2), 125–131. https://doi.org/10.1007/s12110-007-9014-0

Barron, F. (1963). Creativity and psychological health. D. Van Nostrand. Batey, M., & Furnham, A. (2008). The relationship between measures of

creativity and schizotypy. Personality and Individual Differences, 45(8), 816–821. https://doi.org/10.1016/j.paid.2008.08.014

Beaussart, M. L., White, A. E., Pullman, A., & Kaufman, J. C. (2017). Reviewing recent empirical findings on creativity and mental illness. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 42–59). Cam- bridge University Press.

14 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Becker, G. (2017). A socio-historical overview of the creativity-pathology connection: From antiquity to contemporary times. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 3–24). Cambridge University Press.

Billington, J., Baron-Cohen, S., & Wheelwright, S. (2007). Cognitive style predicts entry into physical sciences and humanities: Questionnaire and performance tests of empathy and systemizing. Learning and Individual Differences, 17(3), 260–268. https://www.doi.org/10.1016/j.lindif.2007.02 .004

Brownstein, M., & Solyom, L. (1986). The dilemma of Howard Hughes: Paradoxical behavior in compulsive disorders. Canadian Journal of Psy- chiatry, 31(3), 238–240. https://doi.org/10.1177/070674378603100311

Burch, G. S. J., Pavelis, C., Hemsley, D. R., & Corr, P. J. (2006). Schizo- typy and creativity in visual artists. British Journal of Psychology, 97(Pt 2), 177–190. https://doi.org/10.1348/000712605X60030

Carson, S. (2014). Leveraging the “mad genius” debate: Why we need a neuroscience of creativity and psychopathology. Frontiers in Human Neuroscience, 8, 771. https://doi.org/10.3389/fnhum.2014.00771

Carson, S. H. (2011). Creativity and psychopathology: A shared vulner- ability model. Canadian Journal of Psychiatry, 56(3), 144–153. https:// doi.org/10.1177/070674371105600304

Carson, S. H., Peterson, J. B., & Higgins, D. M. (2003). Decreased latent inhibition is associated with increased creative achievement in high- functioning individuals. Journal of Personality and Social Psychology, 85(3), 499–506. https://doi.org/10.1037/0022-3514.85.3.499

Ceci, S. J., & Williams, W. M. (2011). Understanding current causes of women’s underrepresentation in science. Proceedings of the National Academy of Sciences of the United States of America, 108(8), 3157–3162. https://doi.org/10.1073/pnas.1014871108

Cheung, F. M., & Halpern, D. F. (2010). Women at the top: Powerful lead- ers define success as work þ family in a culture of gender. American Psychologist, 65(3), 182–193. https://doi.org/10.1037/a0017309

Citlak, A. (2016). The Lvov-Warsaw School: The forgotten tradition of historical psychology. History of Psychology, 19(2), 105–124. https:// doi.org/10.1037/hop0000029

Claridge, G., McCreery, C., Mason, O., Bentall, R., Boyle, G., Slade, P., & Popplewell, D. (1996). The factor structure of "schizotypal’ traits: A large replication study. British Journal of Clinical Psychology, 35(1), 103–115. https://doi.org/10.1111/j.2044-8260.1996.tb01166.x

Cox, A. J., & Leon, J. L. (1999). Negative schizotypal traits in the relation of creativity to psychopathology. Creativity Research Journal, 12(1), 25–36. https://doi.org/10.1207/s15326934crj1201_4

Czechowski, K., Miranda, D., & Sylvestre, J. (2016). Like a rolling stone: A mixed- methods approach to linguistic analysis of Bob Dylan’s lyrics. Psychology of Aesthetics, Creativity, and the Arts, 10(1), 99–113. https://doi.org/10.1037/aca0000045

Damian, R. I., & Simonton, D. K. (2015). Psychopathology, adversity, and creativity: Diversifying experiences in the development of eminent Afri- can Americans. Journal of Personality and Social Psychology, 108(4), 623–636. https://doi.org/10.1037/pspi0000011

Dunani, M., & Rodgers, P. (1999). Physics: Past, present, future. Physics World, 12(12), 7–14. https://doi.org/10.1088/2058-7058/12/12/2

Eysenck, H. (1993). Creativity and personality: Suggestions for a theory. Psychological Inquiry, 4(3), 147–178. https://doi.org/10.1207/s15327 965pli0403_1

Eysenck, H. J. (1995). Genius: The natural history of creativity (Vol. 12). Cambridge University Press. https://doi.org/10.1017/CBO9780511752247

Feist, G. J. (1998). A meta-analysis of personality in scientific and artistic creativity. Personality and Social Psychology Review, 2(4), 290–309. https://doi.org/10.1207/s15327957pspr0204_5

Feist, G. J. (2012). Affective states and affective traits in creativity: Evi- dence for non-linear relationships. In M. A. Runco (Ed.), The creativity research handbook (Vol. 3, pp. 61–102). Hampton Press.

Feist, G. J. (2017). The creative personality: Current understandings and debates. In J. Plucker (Ed.), Creativity and innovation: Current under- standings and debates (pp. 181–198). Prufrock Press.

Fisher, J. E. (2015). Challenges in determining whether creativity and men- tal illness are associated. Frontiers in Psychology, 6, 163. https://doi .org/10.3389/fpsyg.2015.00163

Flaherty, A. W. (2005). Frontotemporal and dopaminergic control of idea generation and creative drive. The Journal of Comparative Neurology, 493(1), 147–153. https://doi.org/10.1002/cne.20768

Focquaert, F., Steven, M. S., Wolford, G. L., Colden, A., & Gazzaniga, M. S. (2007). Empathizing and systemizing cognitive traits in the sciences and humanities. Personality and Individual Differences, 43(3), 619–625.

Fodor, E. M., & Laird, B. A. (2004). Therapeutic intervention, bipolar in- clination, and literary creativity. Creativity Research Journal, 16(2–3), 149–161. https://doi.org/10.1080/10400419.2004.9651449

Forgeard, M. J. C., Mecklenburg, A. C., Lacasse, J. J., & Jayawickreme, E. (2017). Bringing the whole universe to order: Creativity, healing, and posttraumatic growth. In J. C. Kaufman (Ed.), Creativity and mental ill- ness (pp. 321–342). Cambridge University Press.

Furnham, A., Batey, M., Anand, K., & Manfield, J. (2008). Personality, hypomania, intelligence and creativity. Personality and Individual Dif- ferences, 44(5), 1060–1069. https://doi.org/10.1016/j.paid.2007.10.035

Gardner, J. (1973). The arts and human development. Wiley. Glazer, E. (2009). Rephrasing the madness and creativity debate: What is

the nature of the creativity construct? Personality and Individual Differ- ences, 46(8), 755–764. https://doi.org/10.1016/j.paid.2009.01.021

Gostoli, S., Cerini, V., Piolanti, A., & Rafanelli, C. (2017). Creativity, bipolar disorder vulnerability and psychological well-being: A prelimi- nary study. Creativity Research Journal, 29(1), 63–70. https://doi.org/10 .1080/10400419.2017.1263511

Greenwood, T. A. (2020). Creativity and bipolar disorder: A shared genetic vulnerability. Annual Review of Clinical Psychology, 16, 239–264. https://doi.org/10.1146/annurev-clinpsy-050718-095449

Gwet, K. (2002). Inter-rater reliability: Dependency on trait prevalence and marginal homogeneity. Statistical Methods for Inter-Rater Reliabil- ity Assessment Series, 2(1), 9.

Gwet, K. L. (2008). Computing inter-rater reliability and its variance in the presence of high agreement. British Journal of Mathematical & Statistical Psychology, 61(Pt 1), 29–48. https://doi.org/10.1348/000711006X126600

Hoffman, M. D., & Gelman, A. (2014). The No-U-turn sampler: Adap- tively setting path lengths in Hamiltonian Monte Carlo. Journal of Machine Learning Research, 15(1), 1593–1623.

Hope, A. C. A. (1968). A simplified Monte Carlo significance test proce- dure. Journal of the Royal Statistical Society: Series B, Methodological, 30, 582–598. https://doi.org/10.1111/j.2517-6161.1968.tb00759.x

Hudson, J. I., Hiripi, E., Pope, H. G., Jr., & Kessler, R. C. (2007). The prevalence and correlates of eating disorders in the National Comorbid- ity Survey Replication. Biological Psychiatry, 61(3), 348–358. https:// doi.org/10.1016/j.biopsych.2006.03.040

Jamison, K. R. (1996). Touched with fire. Simon & Schuster. Jamison, K. R., Gerner, R. H., Hammen, C., & Padesky, C. (1980). Clouds

and silver linings: Positive experiences associated with primary affective disorders. The American Journal of Psychiatry, 137(2), 198–202. https://doi.org/10.1176/ajp.137.2.198

Jeste, D. V., Harless, K. A., & Palmer, B. W. (2000). Chronic late-onset schizophrenia-like psychosis that remitted: Revisiting Newton’s psycho- sis? The American Journal of Psychiatry, 157(3), 444–449. https://doi .org/10.1176/appi.ajp.157.3.444

Johnson, S. L., Murray, G., Fredrickson, B., Youngstrom, E. A., Hinshaw, S., Bass, J. M., Deckersbach, T., Schooler, J., & Salloum, I. (2012). Cre- ativity and bipolar disorder: Touched by fire or burning with questions? Clinical Psychology Review, 32(1), 1–12. https://doi.org/10.1016/j.cpr .2011.10.001

PSYCHOPATHOLOGY AND CREATIVITY 15

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Johnson, S. L., Murray, G., Hou, S., Staudenmaier, P. J., Freeman, M. A., & Michalak, E. E., & the CREST.BD. (2015). Creativity is linked to ambition across the bipolar spectrum. Journal of Affective Disorders, 178, 160–164. https://doi.org/10.1016/j.jad.2015.02.021

Johnson, S. L., Tharp, J. A., & Holmes, M. K. (2015). Understanding crea- tivity in bipolar I disorder. Psychology of Aesthetics, Creativity, and the Arts, 9(3), 319–327. https://doi.org/10.1037/a0038852

Kaufman, J. C. (2017). Creativity and mental illness. Cambridge Univer- sity Press.

Kaufman, J. C., & Beghetto, R. A. (2009). Beyond big and little: The four c model of creativity. Review of General Psychology, 13(1), 1–12. https://doi.org/10.1037/a0013688

Kaufmann, G., & Kaufmann, A. (2017). When good is bad and bad is good: Mood, bipolarity, and creativity. In J. C. Kaufman (Ed.), Creativ- ity and mental illness (pp. 205–235). Cambridge University Press.

Kelter, R. (2020). fbst: The Full Bayesian Significance Test and the e-Value (R package version 1.0). https://CRAN.R-project.org/package=fbst

Kéri, S. (2009). Genes for psychosis and creativity: A promoter polymor- phism of the neuregulin 1 gene is related to creativity in people with high intellectual achievement. Psychological Science, 20(9), 1070–1073. https://doi.org/10.1111/j.1467-9280.2009.02398.x

Kessler, R. C., Angermeyer, M., Anthony, J. C., DE Graaf, R., Demyttenaere, K., Gasquet, I., DE Girolamo, G., Gluzman, S., Gureje, O., Haro, J. M., Kawakami, N., Karam, A., Levinson, D., Medina Mora, M. E., Oakley Browne, M. A., Posada-Villa, J., Stein, D. J., Adley Tsang, C. H., Aguilar-Gaxiola, S., . . . Ustün, T. B. (2007). Lifetime prevalence and age-of-onset distributions of mental disorders in the world health organization’s world mental health survey initiative. World Psychiatry, 6(3), 168–176.

Kessler, R. C., Berglund, P., Demler, O., Jin, R., Merikangas, K. R., & Walters, E. E. (2005). Lifetime prevalence and age-of-onset distribu- tions of DSM–IV disorders in the National Comorbidity Survey Replica- tion. Archives of General Psychiatry, 62(6), 593–602. https://doi.org/10 .1001/archpsyc.62.6.593

Kinney, D., & Richards, R. (2017). Creativity as “compensatory advant- age”: bipolar and schizophrenic liability, the inverted-U hypothesis, and practical implications. In J. C. Kaufman (Ed.), Creativity and mental ill- ness (pp. 295–317). Cambridge University Press.

Ko, Y., & Kim, J. (2008). Scientific geniuses’ psychopathology as a mod- erator in the relation between creative contribution types and eminence. Creativity Research Journal, 20(3), 251–261. https://doi.org/10.1080/ 10400410802278677

Kozbalt, A., Kaufman, S. B., Walder, D. J., Ospina, L. H., & Kim, J. U. (2017). The evolutionary genetics of the creativity-psychosis connec- tion. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 102–132). Cambridge University Press.

Kyaga, S., Lichtenstein, P., Boman, M., Hultman, C., Långström, N., & Landén, M. (2011). Creativity and mental disorder: Family study of 300,000 people with severe mental disorder. The British Journal of Psy- chiatry, 199(5), 373–379. https://doi.org/10.1192/bjp.bp.110.085316

LeBoutillier, N., Barry, R., & Westley, D. (2014). The role of schizotypy in predicting performance on figural and verbal imagery-based measures of creativity. Creativity Research Journal, 26(4), 461–467. https://doi .org/10.1080/10400419.2014.961778

Lev-Ran, S., Imtiaz, S., Rehm, J., & Le Foll, B. (2013). Exploring the asso- ciation between lifetime prevalence of mental illness and transition from substance use to substance use disorders: Results from the National Epi- demiologic Survey of Alcohol and Related Conditions (NESARC). The American Journal on Addictions, 22(2), 93–98. https://doi.org/10.1111/j .1521-0391.2013.00304.x

Lucijani�c, M., Breitenfeld, D., Mileti�c, J., Buljan, D., Ozimec-Vulinec, Š., & Akrap, A. (2010). Rock musicians’ Club 27. Alcoholism: Journal on Alcoholism and Related Addictions, 46(2), 109–113.

Ludwig, A. (1995). The price of greatness: Resolving the creativity and madness controversy. Guilford Press.

Ludwig, A. (1998). Method and madness in the arts and sciences. Creativity Research Journal, 11(2), 93–101. https://doi.org/10.1207/s15326934crj1102_1

Ludwig, A. M. (1992). Creative achievement and psychopathology: Com- parison among professions. American Journal of Psychotherapy, 46(3), 330–356. https://doi.org/10.1176/appi.psychotherapy.1992.46.3.330

Martín-Brufau, R., & Corbalán, J. (2016). Creativity and psychopathology: Sex matters. Creativity Research Journal, 28(2), 222–228. https://doi .org/10.1080/10400419.2016.1165531

Mason, O., Claridge, G., & Jackson, M. (1995). New scales for the assess- ment of schizotypy. Personality and Individual Differences, 18(1), 7–13. https://doi.org/10.1016/0191-8869(94)00132-C

Merikangas, K. R., Akiskal, H. S., Angst, J., Greenberg, P. E., Hirschfeld, R. M., Petukhova, M., & Kessler, R. C. (2007). Lifetime and 12-month prevalence of bipolar spectrum disorder in the National Comorbidity Survey replication. Archives of General Psychiatry, 64(5), 543–552. https://doi.org/10.1001/archpsyc.64.5.543

Moss-Racusin, C. A., Dovidio, J. F., Brescoll, V. L., Graham, M. J., & Handelsman, J. (2012). Science faculty’s subtle gender biases favor male students. Proceedings of the National Academy of Sciences of the United States of America, 109(41), 16474–16479. https://doi.org/10 .1073/pnas.1211286109

Motto, A. L., & Clark, J. R. (1992). The paradox of genius and madness: Seneca and his influence. Cuadernos de filología clásica. Estudios Lat- inos, (2), 189–200.

Nasar, S. (2011). A beautiful mind. Simon & Schuster. Nettle, D. (2001). Strong imagination: Madness, creativity and human na-

ture. Oxford University Press. Nettle, D. (2006). Schizotypy and mental health amongst poets, visual

artists, and mathematicians. Journal of Research in Personality, 40(6), 876–890. https://doi.org/10.1016/j.jrp.2005.09.004

Nock, M. K., & Kessler, R. C. (2006). Prevalence of and risk factors for suicide attempts versus suicide gestures: Analysis of the National Comorbidity Sur- vey. Journal of Abnormal Psychology, 115(3), 616–623. https://doi.org/10 .1037/0021-843X.115.3.616

Nowakowska, C., Strong, C. M., Santosa, C. M., Wang, P. W., & Ketter, T. A. (2005). Temperamental commonalities and differences in euthy- mic mood disorder patients, creative controls, and healthy controls. Journal of Affective Disorders, 85(1-2), 207–215. https://doi.org/10 .1016/j.jad.2003.11.012

Paek, S. H., Abdulla, A. M., & Cramond, B. (2016). A meta-analysis of the relationship between three common psychopathologies—ADHD, anxiety, and depression—and indicators of little-c creativity. Gifted Child Quar- terly, 60(2), 117–133. https://doi.org/10.1177/0016986216630600

Perälä, J., Suvisaari, J., Saarni, S. I., Kuoppasalmi, K., Isometsä, E., Pirkola, S., Partonen, T., Tuulio-Henriksson, A., Hintikka, J., Kieseppä, T., Härkänen, T., Koskinen, S., & Lönnqvist, J. (2007). Lifetime preva- lence of psychotic and bipolar I disorders in a general population. Ar- chives of General Psychiatry, 64(1), 19–28. https://doi.org/10.1001/ archpsyc.64.1.19

Pereira, C. A. B., & Stern, J. M. (1999). Evidence and credibility: Full Bayesian significance test for precise hypotheses. Entropy, 1(4), 99–110. https://doi.org/10.3390/e1040099

Pereira, C. A. B., & Stern, J. M. (2020). The e-value: a fully Bayesian sig- nificance measure for precise statistical hypotheses and its research pro- gram. São Paulo Journal of Mathematical Sciences. Advance online publication. https://doi.org/10.1007/s40863-020-00171-7

Perry, I. H. (1947). Vincent van Gogh’s illness: — case record. Bulletin of the History of Medicine, 21, 146–172.

Porter, A. M., & Ivie, R. (2019). Women in physics and astronomy, 2019. https://www.aip.org/statistics/reports/women-physics-and-astronomy-2019 #:�:text=Highlights,and%2040%25%20of%20astronomy%20doctorates

16 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Post, F. (1994). Creativity and psychopathology. A study of 291 world-fa- mous men. The British Journal of Psychiatry, 165(1), 22–34. https://doi .org/10.1192/bjp.165.1.22

Power, R. A., Steinberg, S., Bjornsdottir, G., Rietveld, C. A., Abdellaoui, A., Nivard, M. M., Johannesson, M., Galesloot, T. E., Hottenga, J. J., Willemsen, G., Cesarini, D., Benjamin, D. J., Magnusson, P. K. E., Ullén, F., Tiemeier, H., Hofman, A., van Rooij, F. J. A., Walters, G. B., Sigurdsson, E., . . . Stefansson, K. (2015). Polygenic risk scores for schizophrenia and bipolar disorder predict creativity. Nature Neuro- science, 18(7), 953–955. https://doi.org/10.1038/nn.4040

Ramey, C. H., & Weisberg, R. W. (2004). The” poetical activity” of Emily Dickinson: A further test of the hypothesis that affective disorders foster creativity. Creativity Research Journal, 16(2-3), 173–185. https://doi .org/10.1080/10400419.2004.9651451

Reiter-Palmon, R., & Schoenbeck, M. (2020). Creativity equals creativity- –or does it? How creativity is measured influences our understanding of creativity. In V. Dorfler & M. Stierand (Eds.), Handbook of research methods on creativity (pp. 290–300). Edward Elgar Publishing. https:// doi.org/10.4337/9781786439659.00031

Richards, R. (1990). Everyday creativity, eminent creativity, and health: ‘Afterview’ for CRJ issues on creativity and health. Creativity Research Journal, 3(4), 300–326. https://doi.org/10.1080/10400419009534363

Richards, R. (1994). Creativity and bipolar mood swings: Why the associa- tion? In M. P. Shaw & M. A. Runco (Eds.), Creativity and affect (pp. 44–72). Ablex Publishing.

Richards, R., Kinney, D. K., Lunde, I., Benet, M., & Merzel, A. P. (1988). Creativity in manic-depressives, cyclothymes, their normal relatives, and control subjects. Journal of Abnormal Psychology, 97(3), 281–288. https://doi.org/10.1037/0021-843X.97.3.281

Robins, L. N., Helzer, J. E., Weissman, M. M., Orvaschel, H., Gruenberg, E., Burke, J. D., Jr., & Regier, D. A. (1984). Lifetime prevalence of specific psychiatric disorders in three sites. Archives of General Psychiatry, 41(10), 949–958. https://doi.org/10.1001/archpsyc.1984.01790210031005

Rothenberg, A. (1990). Creativity and madness: New findings and old ster- eotypes. Johns Hopkins University Press.

Ruiter, M., & Johnson, S. L. (2015). Mania risk and creativity: A multi- method study of the role of motivation. Journal of Affective Disorders, 170(1), 52–58. https://doi.org/10.1016/j.jad.2014.08.049

Runco, M., & Jaeger, G. J. (2012). The standard definition of creativity. Creativity Research Journal, 24(1), 92–96. https://doi.org/10.1080/ 10400419.2012.650092

Santosa, C. M., Strong, C. M., Nowakowska, C., Wang, P. W., Rennicke, C. M., & Ketter, T. A. (2007). Enhanced creativity in bipolar disorder patients: A controlled study. Journal of Affective Disorders, 100(1–3), 31–39. https://doi.org/10.1016/j.jad.2006.10.013

Sawyer, R. K. (2011). Explaining creativity: The science of human innova- tion. Oxford University Press.

Schlesinger, J. (2009). Creative myth conceptions: A closer look at the evi- dence for the “mad genius” hypothesis. Psychology of Aesthetics, Crea- tivity, and the Arts, 3(2), 62–72. https://doi.org/10.1037/a0013975

Schlesinger, J. (2012). The insanity hoax: Exposing the myth of the mad genius. Shrinktunes Media.

Schlesinger, J. (2017). Building connections on sand: The cautionary chap- ter. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 60–75). Cambridge University Press.

Schuldberg, D. (1990). Schizotypal and hypomanic traits, creativity, and psychological health. Creativity Research Journal, 3(3), 218–230. https://doi.org/10.1080/10400419009534354

Seneca. (2007). Dialogues and essays: Tranquility of mind (J. Davie, Trans.). Oxford.

Silvia, P. J., & Kaufman, J. C. (2010). Creativity and mental illness. In J. C. Kaufman & R. J. Sternberg (Eds.), The Cambridge handbook of creativity (pp. 381–394). Cambridge University Press. https://doi.org/10 .1017/CBO9780511763205.024

Silvia, P. J., & Kimbrel, N. A. (2010). A dimensional analysis of creativity and mental illness: Do anxiety and depression symptoms predict creative cognition, creative accomplishments, and creative self-concepts? Psy- chology of Aesthetics, Creativity, and the Arts, 4(1), 2–10. https://doi .org/10.1037/a0016494

Simeonova, D. I., Chang, K. D., Strong, C., & Ketter, T. A. (2005). Crea- tivity in familial bipolar disorder. Journal of Psychiatric Research, 39(6), 623–631. https://doi.org/10.1016/j.jpsychires.2005.01.005

Simonton, D. K. (2006). Creativity and madness. Talk presented to Psy- chology Forum, The Commonwealth Club, San Francisco, CA.

Simonton, D. K. (2010). So you want to become a creative genius? You must be crazy. In D. H. Cropley, A. J. Cropley, J. C. Kaufman, & M. A. Runco (Eds.), The dark side of creativity (pp. 218–234). Cambridge Uni- versity Press. https://doi.org/10.1017/CBO9780511761225.012

Simonton, D. K. (2013). Creative thought as blind variation and selective retention: Why creativity is inversely related to sightedness. Journal of Theoretical and Philosophical Psychology, 33(4), 253–266. https://doi .org/10.1037/a0030705

Simonton, D. K. (2014). More method in the mad-genius controversy: A historiometric study of 204 historic creators. Psychology of Aesthetics, Creativity, and the Arts, 8(1), 53–61. https://doi.org/10.1037/a0035367

Simonton, D. K. (2016). Defining creativity: Don't we also need to define what is not creative? Journal of Creative Behavior, 52, 80–90. https:// doi.org/10.1002/jocb.137

Simonton, D. K. (2017a). Creative genius and psychopathology: Creativity as positive and negative personality. In G. J. Feist, R. Reiter-Palmon, & J. C. Kaufman (Eds.), The Cambridge handbook of creativity and per- sonality research (pp. 235–250). Cambridge University Press. https:// doi.org/10.1017/9781316228036.013

Simonton, D. K. (2017b). The mad (creative) genius: What do we know af- ter a century of historiometric research? In J. C. Kaufman (Ed.), Creativ- ity and mental illness (pp. 25–41). Cambridge University Press.

Simonton, D. K. (2019). Creativity and psychopathology: The tenacious mad-genius controversy updated. Current Opinion in Behavioral Scien- ces, 27, 17–21. https://doi.org/10.1016/j.cobeha.2018.07.006

Simonton, D. K., & Song, A. V. (2009). Eminence, IQ, physical and men- tal health, and achievement domain: Cox’s 282 Geniuses revisited. Psy- chological Science, 20(4), 429–434. https://doi.org/10.1111/j.1467-9280 .2009.02313.x

Stan Development Team. (2016). rstanarm: Bayesian applied regression modeling via Stan. R package version 2.13.1. http://mc-stan.org/

Stein, D. J., Phillips, K. A., Bolton, D., Fulford, K. W. M., Sadler, J. Z., & Kendler, K. S. (2010). What is a mental/psychiatric disorder? From DSM–IV to DSM-V. Psychological Medicine, 40(11), 1759–1765. https://doi.org/10.1017/S0033291709992261

Sternberg, R. J. (1988). A three-facet model of creativity. In R. J. Sternberg & R. J. Sternberg (Eds.), The nature of creativity: Contemporary psycho- logical perspectives (pp. 125–147). Cambridge University Press.

Swain, J. E., & Swain, J. D. (2017). Non-linearity in creativity and mental illness: The mixed blessings of chaos, catastrophe, and noise in brain and behavior. In J. C. Kaufman (Ed.), Creativity and mental illness (pp. 133–144). Cambridge University Press.

Taylor, C. L. (2017). Creativity and mood disorder: A systematic review and meta-analysis. Perspectives on Psychological Science, 12(6), 1040–1076. https://doi.org/10.1177/1745691617699653

Thomson, P. (2017). Trauma, attachment, and creativity. In T. Marks- Tarlow, M. Solomon, & D. J. Siegel (Eds.), Play and creativity in psy- chotherapy (pp. 167–190). W.W. Norton.

Thomson, N. D., Wurtzburg, S. J., & Centifanti, L. C. (2015). Empathy or science? Empathy explains physical science enrollment for men and women. Learning and Individual Differences, 40(2), 115–120.

Thulin, M. (2014). Decision-theoretic justifications for Bayesian hypothe- sis testing using credible sets. Journal of Statistical Planning and Infer- ence, 146(3), 133–138. https://doi.org/10.1016/j.jspi.2013.09.014

PSYCHOPATHOLOGY AND CREATIVITY 17

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.

Thys, E., Sabbe, B., & De Hert, M. (2014). Creativity and psychopathol- ogy: A systematic review. Psychopathology, 47(3), 141–147. https://doi .org/10.1159/000357822

U.S. Census Bureau. (2019). Quick facts: United States. Race and Hispanic Origin. https://www.census.gov/quickfacts/fact/table/U.S ./PST045219

Verhaeghen, P., Joorman, J., & Khan, R. (2005). Why we sing the blues: The relation between self-reflective rumination, mood, and creativity. Emotion, 5(2), 226–232. https://doi.org/10.1037/1528-3542.5.2.226

Verhaeghen, P., Joormann, J., & Aikman, S. N. (2014). Creativity, mood, and the examined life: Self-reflective rumination boosts creativity, brooding breeds dysphoria. Psychology of Aesthetics, Creativity, and the Arts, 8(2), 211–218. https://doi.org/10.1037/a0035594

Wei, X., Yu, J. W., Shattuck, P., & Blackorby, J. (2017). High school math and science preparation and postsecondary STEM participation for students with an autism spectrum disorder. Focus on Autism and Other Developmental Disabilities, 32(2), 83–92. https://doi.org/10.1177/1088357615588489

Wills, G. I. (2003). Forty lives in the bebop business: Mental health in a group of eminent jazz musicians. The British Journal of Psychiatry, 183(3), 255–259. https://doi.org/10.1192/bjp.183.3.255

Wongpakaran, N., Wongpakaran, T., Wedding, D., & Gwet, K. L. (2013). A comparison of Cohen’s Kappa and Gwet’s AC1 when calculating inter-rater reliability coefficients: A study conducted with personality disorder samples. BioMed Central Medical Research Methodology, 13, 61. https://doi.org/10.1186/1471-2288-13-61

Wuthrich, V., & Bates, T. C. (2001). Schizotypy and latent inhibition: Non-linear linkage between psychometric and cognitive markers. Per- sonality and Individual Differences, 30(5), 783–798. https://doi.org/10 .1016/S0191-8869(00)00071-4

Received December 10, 2020 Revision received July 22, 2021

Accepted August 13, 2021 n

18 FEIST, DOSTAL, AND KWAN

T hi s do cu m en t is co py ri gh te d by

th e A m er ic an

P sy ch ol og ic al A ss oc ia ti on

or on e of

it s al li ed

pu bl is he rs .

T hi s ar ti cl e is in te nd ed

so le ly

fo r th e pe rs on al us e of

th e in di vi du al us er

an d is no t to

be di ss em

in at ed

br oa dl y.