• tonijocoppa

Art & Mental Illness

Interpretive Review of Art and Mental Illness Research by Toni Jo Coppa

“Art saved my life”.

I have made this statement countless times throughout my twenty-five-year art career, “art saved my life.” I have also seen many experts in the professional psychology realm during this time period. Recently, I have found myself growing out of my dependency on art making, I now look back and wonder how this worked exactly. Why did I credit my art making and not the people trying to help me? I had my ideas, for sure, and even wrote a thesis about it. My theory was something like this: having the darkness of our psyches displayed outside of our own minds would give one a sense of empowerment. Any disturbing emotions we might carry would no longer quietly eat at our souls like cancer but would be transformed by turning the original negativity into something positive like creativity. I was aware of the history of famous artists and their mental illness, but what could this possibly have to do with me and other professional artists of today? In this research project I found both answers to some of my personal questions and leads on how to proceed with further research about the topic.

When I first began looking for research articles, I used search words as general as “art” and “catharsis” but began to narrow down my interests specifically to “artists”, “mental health” and “transference”. I retrieved my selected articles from the PsychINFO, PsychARTICLES, Google Scholar and the Academic Search Complete databases. I was happy to discover the many books on the subject in the reference sections of the research papers as well. I find it exciting to consider this final assignment a jumping off point rather than a culmination of this course.

Theory, Design and Purpose

Let me also be clear to start, that when I use the term “artist” throughout this review, I am referring to creative people like writers, visual artists, composers, dancers, etc. who use their imagination to address life’s significance, meaning, emotions by employing their own subjective interpretations. There is plenty of research to choose from on art and mental illness, and I found a good mix of studies. In the end I narrowed it down to one qualitative and five quantitative articles. The majority of them supported the idea that there are some positive aspects to being an artist with diagnosed mental health challenges. Two of these positive attributes were that making art gave the creator hope for the future and that their

cognitive styles of divergent thinking were valued in particular professions. There were also a number of examples supporting my own experience. One sentence that particularly resonated with me on contributing to a sense of mental well-being was “An art practice was seen to offer a reliable, containing platform for this reconfiguration, one that many participants declared had not been provided through psychological therapies” (Sagan 2015, p. 76).

The one qualitative study I chose may have answered what it was that was being provided in art making that the participants were not getting from psychological therapies. Sagan (2015) wrote in “Hope Crept In”: A Phenomenological Study of Mentally Ill Artists’ Biographic Narrative”, that it “both offered and nurtured hope” (p. 76). Sagan (2015) used an ethnographic, phenomenological and narrative study design to get an etic perspective on working artists for her research. There has been a long-standing belief, with documentation as early as the 1980s, that producing visual art benefits people who suffer from long-term mental illness. She wanted to find out precisely how this “works” through narrative studies.

The other quantitative studies seemed to flesh out the answers to questions I wasn’t even aware I had, and also answered the ones I did. The following two studies were located in the PsychArticles database. The first, by Greason, Glaser & Mroz (2015), used a nonexperimental group comparison. They wondered if the perception that student artists are in much greater need of mental health care than other students had any truth to it and decided to find out. In my third article, “Method and Madness in the Arts and Sciences”, Ludwig (1998) uses a correlational secondary analysis using inferential statistics and continuous variables. There were two questions being addressed in this article. The first one was what bearing mental instability had on the nature of creative expression in the arts alone and in the sciences alone. The second one was if the findings would line up with fractal geometry theory, which would give us patterns to predict likelihood of mental issues.

I found the next two studies on the psychINFO database. The first; “Heightened Incidence of Depressive Symptoms in Adolescents Involved in the Arts,” Young, Winner & Cordes (2013) was a quantitative group comparison using descriptive statistics’ to find if the association of affective disorders and the arts could be found out as early as adolescents. The second study from Knudsen, Bookheimer & Bilder (2019) sought to find out if psychopathology is elevated in exceptionally creative visual artists and scientists. Their article “Is

Psychopathology Elevated in Big-C Visual Artists and Scientists?” was also quantitative study using a group comparison design. They wanted to see if “Big-C” visual artists and scientists would differ on the laboratory assessments of creativity and used a measure of variability to describe results.

The last article I selected, which I found on the Google Scholar database was “Schizotypy and Mental Health Amongst Poets, Visual Artists, and Mathematicians,” Nettle (2006). This was a group comparison as well, with multiple regressions predicting the trajectory that creative people will score high on the unusual experiences’ measurements and low on the introvertive anhedonia O-LIFE measurements. The second purpose was to investigate the relationship between creativity and psychopathology concerns and the differences between creative domains. All of the articles I chose were exploratory studies rather than experimental, including the qualitative one I started with by Olivia Sagan.

I want to briefly discuss the appropriateness of the literature chosen to support my articles. Four of the six articles that I am reviewing were written within the last eight years and the two others were done closer to the turn of the century. 65% of the supporting literature for the more recent studies were from the last nineteen years. I would say almost all of the research referenced was actually quite current, but the numbers were skewed a little bit by quotes from books like Aristotle. (1936). Problems: Books XXII-XXXVIII and/or Nietzsche, F. (1954) The birth of tragedy.

I found that A. Ludwig provided the least amount by way of literature review. Ten of the eighteen sited references were books rather than journals of research, and some of those were only used as sources for sample data. This third article choice stood out to me on every level of our research review breakdown, and the supporting literature chosen was not any different. All the others were pretty rich on supporting literature, with many of the same studies used in multiple articles. I have to wonder though, could some of these references just be added at the end to suggest rich supporting data? What are the criteria for included references? I am quite sure reference selection is not a neutral act.


All of the samples, in all of the articles I chose, had at least one group that was “artistic”. This group was always chosen in a non-random fashion to make up a convenient sample. The first example of this was Sagan’s (2015) group in “‘Hope Crept In’: A Phenomenological Study of Mentally Ill Artists’ Biographic Narrative”, which consisted of artists chosen from

mental health networks meeting the criteria that they were actively involved in the arts.

Being a qualitative study, I did not find the final small number of 40 participants surprising. The mean age was 40.6 of this homogenous group, which I respected because it suggested a commitment to the arts and a supposed level of maturity.

Although the next articles discussed are quantitative, they also all follow the non-random method. “Thinking Outside the Box: Psychological Needs of Art Students Compared with Traditional Students” artistic sample was chosen from three art universities; two private, one public. “Method and Madness in the Arts and Sciences” gathered their non-random sample from a secondary analysis of a prior study done by the author. Ludwig also used a book by Lucie-Smith (1986)., “Lives of the great twentieth century artists.” New York: Rizzoli, which when combined with the earlier study, produced 137 well-known artists that met his criteria. I think this is about as non-random as you can get. I do not like the art history book sample recruitment method one bit. It is so convenient, it’s hard to respect it. How much was the information edited? Was Lucie-Smith biased in choosing which artists were included in the book? This all boils down to a questionable sample selection. In “Is Psychopathology Elevated in Big-C Visual Artists and Scientists?”, their artistic sample was pulled from the candidates for prestigious awards given to artists, like the MacArthur or

the Guggenheim Fellowship. I wished Ludwig had gathered his sample more along the lines of this method. I’ve been witness to the intense process of awarding highly competitive grants (for the State of Illinois) and the procedure eliminates most chances for favoritism, unlike book publishing.

According to Beaudry & Miller (2016) having so many non-random samples is expected in group comparisons. It says in Research Literacy; A Primer for Understanding and Using Research, “The sample for a nonexperimental group comparison study always includes subjects who are part of an existing membership group that is based on a shared, immutable characteristic “(p. 204). All of my selections support this as true. I think it is more relevant to have random sampling when one has an experimental design.

Data Collection

With the exception of my first article, there were a range of tests and measures throughout. Addressing the oddball in the group of studies I chose, “’Hope Crept In’: A Phenomenological Study of Mentally Ill Artists’ Biographic Narrative”, used only interviews, which is a common practice in qualitative research. The contributors to the data were told to reflect on their life, mental illness and their art, which was then transcribed verbatim.

This was then analyzed using the method of Phenomenological Interpretative Analysis (IPA). I have found out that this method of analysis is quite popular in qualitative research. The general consensus seems to be that the IPA provides a rich and nuanced insight into the experiences of research participants.

Greason, et al. (2015) used the survey called The American College Health Association National College Health Assessment II (ACHA-ANCHA II) to collect their data. They employed the chi-square as the statistical tool to determine whether the IVs (art vs. traditional students) showed a significant difference in their study for the article “Thinking Outside the Box: Psychological Needs of Art Students Compared with Traditional Students”. In addition to the survey, they asked forty-five questions about the students’ mental health (e.g., “Have you ever felt things were hopeless”), and they responded with their symptoms and experiences. These questions were then analyzed by the researchers without the use of the IPA or any other official evaluation tool. I believe that it was because the other data collected beforehand gave them a solid foundation to build on. Because of this foundation, the credibility strength of these supporting questions and insight provided did not have to be as critically compelling.

Another quantitative study that used the survey method was the Schizotypy article. Nettle gathered his data through a comprehensive questionnaire that was given to grouped participants using the O-LIFE schizotypy scales. Like the Greason study above, he also utilized additional questioning. The questions in this study were less about mental health and more about demographics and general life history though. There was a moderate correlation (r=.25 to .42) between the four factors that make up schizotypy: cognitive disorganization, unusual experiences, introvertive anhedonia and impulsive nonconformity. Unusual experiences are also correlated with impulsive nonconformity (r=.46).

I rate the reliability of the data collection methods was medium to high, with the exception of the Method and Madness article because of the lack of actual data. The only information about how Ludwig gathered his data was in a note on the bottom of page ninety-five. He writes, “A detailed discussion of methodological, statistical, and conceptual issues can be found in my book The Price of Greatness”, Ludwig (2015, p. 95.)

Data Analysis and Results

So, my original question was how making art or being actively creative might help artists in a way that therapy could not. The article on hope found that their results suggested movement from the static past to feelings of a more fluid future that is facilitated by having an art practice. They discovered this by coding the verbatim transcripts, identifying themes within and then those were given to another researcher for analysis to increase the

trustworthiness and confidence in the results. None of the other articles answered my question directly, but like I wrote earlier, some of them answered questions I didn’t even know I had.

An example of this is the analysis and results from the Method and Madness article. Ludwig’s data analysis showed a significant difference between natural and social scientists (X2=5.3, p<.01). The question I didn’t even know I had was how this would transfer across professions. Ludwig (1998) found that “the lifetime prevalence for any form of mental disturbance not only varies among the different professions but appears to increase progressively as we move from more structured, precise professions, such as architecture and design, to more subjective, imprecise ones such as the visual arts and writing” P. 95). He found that this same pattern even applies as we break down artistic styles. For example, the more formal or technical a process, the less likely there would be for the presence of mental illness, the more emotive a process, the higher the prevalence of mental illness. I find this absolutely fascinating.

This idea of a correlating prevalence of how much mental illness is in a profession based on how mathematical, formal and objective the modes used is also supported by the

Schizotypy article. The schizotypy scale scores for the Greason, et al (2015) study that were derived from the O-LIFE questionnaire looked at the group differences using an analysis of

variance. This was done by taking the scores and using them as DVs for the multivariate analysis. An ANOVA was used to report all significant results. They found that the mathematicians had a significantly lower frequency of psychopathology compared to the non-mathematicians (X2=5.35, p<.05, OR=.38). This suggests a correlation of a U-shaped model, meaning that creative professionals would score high on the schizotypy scales, but no where near as high as schizophrenia patients. Put another way, creativity increased with increasing schizotypy but then decreased as psychopathology begins to take its toll.

If what these studies show is true, then it would absolutely make sense that Van Gogh, the father of expressionism, was plagued by psychiatric illness his whole life; eventually ending by suicide. Pollock, a purely emotive painter, was posthumously diagnosed as bipolar and is widely known to have suffered from alcoholism. On the other end of the spectrum with a more mathematical style, the formal abstract artist Mondrian has no mention of mental illness in any of his biographies. Like I said in the beginning of this paper, this final assignment in not really a conclusion, but rather springboard.

Quality of Studies

The quality of all the research I used was quite high, with the exception of Ludwig’s sample

selection. I do also wish he provided more information about his data collection process in

this article from the original study he did. It would have been very helpful in judging the strength or weakness of his data collection methods.

Every article I chose started their studies with supportive background information about artists and mental illness, except the study about the Big-C visual artists. The very first sentence in the abstract for the Schizotypy article by Nettle (2006) reads, “Many researchers have found evidence of an association between creativity and the predisposition to mental illness” (p. 866). The study on adolescents in the arts and mental health follows suit. The first sentence in this article reads, “Studies have shown a higher than average incidence of mental illness in adult artists” (p. 197). Even Ludwig shares, “We also know that several recent studies, including my own, have documented exceptionally high rate of mental illness among creative artists” (p.93). If all this evidence abounds, why does the Knudsen, et al (2019) article begin stating the opposite? She writes, “The idea that psychopathology is associated with exceptional creativity has persisted despite a paucity of corroborating or disconfirming evidence” (p. 273). She also writes, “Seventy years of scientific research on creativity have produced no clear consensus about the often-claimed

association of creativity with psychiatric illness” (p. 273). Upon looking at the supporting

research on the statements in the other articles in comparison to hers, her statements are

followed up with not a single reference to support her statements. Beginning a study in conflict with the research of others of the same subject makes me wary of her entire study. If you are going to challenge something, back it up. Luckily, she redeems herself during the research process with her data collections methods and analysis, and eventually touches on something of great interest for me moving forward. Knudsen, et al (2019) writes in her conclusion:

“It appears that people benefit from psychopathological traits, of uniqueness, only up to a certain extent. If uniqueness is divergent enough to cause dysfunction, this dysfunction probably also affects creativity as well. This is consistent with studies showing people with diagnosable psychopathology do not perform as well on tests of cognition. Overall, these data are persuasive that full expressions of mental illness most likely interfere with creative achievement” (p. 281).

Sagan (2015) writes:

“Studies investigating art activity as distinguished from art therapy point to such benefit as cognitive distraction, “derailing” of negative thinking patterns, increased social capital, fostering a sense of belonging and reduced symptoms of stress, depression and anxiety” (p 73).

Turns out my own theory was somewhat supported by my research. There was no specific mention of ‘transference’ onto objects, but the word ‘derailing’ works for me in a similar manner. It also makes a distinction between art therapy and art practice, and all of these studies were based on an active art practice. In addition, it has substantiated my belief that making art does help in a way that therapy doesn’t, and I am not talking about art therapy. Knudsen says there are some positives to artists with mental illness up to a point, then it becomes a liability. Sagan shows that being an active artist is beneficial for people with mental health struggles. Is there a quantifiable median point for this? I wonder.


Beaudry, J. & Miller, L. (2016). Research Literacy: A Primer for Understanding and Using Research. New York, NY: The Guilford Press.

Greason, D., Glaser, T., & Mroz, K. (2015). Thinking Outside the Box: Psychological Needs of Art Students Compared with Traditional Students. Journal of College Student Psychotherapy, 29, 53-71.

Knudsen, K., Bookheimer, S., & Bilder, R. (2019). Is Psychopathology Elevated in Big-C Visual Artists and Scientists? Journal of Abnormal Psychology, 128(4), 273-283

Ludwig, A. (1998). Method and Madness in the Arts and Sciences. Creativity Research Journal, 11(2), 93-101.

Ludwig, A. (1995). The Price of Greatness. New York, NY: The Guilford Press.

Nettle, D. (2006). Schizotypy and Mental Health Amongst Poets, Visual Artists, and Mathematicians. Journal of Research in Personality, 40, 876-890.

Sagan, O. (2015). “Hope Crept In”: A Phenomenological Study of Mentally Ill Artists’ Biographic Narrative. Journal of Mental Health, 24(2), 73-77.

Young, L., Winner, E., & Cordes, S. (2013). Heightened Incidence of Depressive Symptoms in Adolescents Involved in the Arts. Psychology of Aesthetics, Creativity, and the Arts, 7(23), 197-201.

16 views0 comments

Recent Posts

See All