Home       SAPA Project Test       Papers

A Critical Review of Eysenck's Theory of Psychoticism and How it Relates to Creativity

Shane K. Porzio
Rochester Institute of Technology

This paper summarizes and examines Eysenck's theory of psychoticism in personality and how it could relate directly to creativity. The main topic covered is Eysenck's view of psychoticism as a dimension of personality and temperament. Eysenck's theory of psychoticism is based on mostly physiological factors. It states that normal subjects who are not diagnosable psychotics can, under certain circumstances, exhibit some qualities commonly found among psychotics. It is also shown how he ascribes this theory to creativity and how the genetic and psychological traits of psychoticism and creativity are found to be greatly overlapping. Lastly, an extrapolation and a counterargument posed to Eysenck by Amabile details what Eysenck overlooked, took for granted, or failed to define and incorporate into his theory.

Hans J. Eysenck developed his model of personality based mostly on a psychophysiological basis. Being a behaviorist, Eysenck considers learned habits of great importance, but more importantly he considered personality differences as growing out of our genetic inheritance (Boeree, 1998). Because he based his theory on the genetic factors associated with psychology, Eysenck was mainly concerned with what is known as temperament.

Eysenck developed his model of personality and temperament with three dimensions: Psychoticism, Extraversion, and Neuroticism. This became known as the "PEN" model of personality. Originally Eysenck only theorized about neuroticism and extraversion, but later he realized that psychoticism was also a contributing factor of personality. He then added psychoticism into his theory as the third factor of his model giving birth to his BIG-Three model of personality.

Psychoticism Defined

His description of psychoticism states that a person will exhibit some qualities commonly found among psychotics, and that they may be more susceptible, given certain environments, to becoming psychotic. Examples of such psychotic tendencies include recklessness, disregard for common sense, and inappropriate emotional expression to name a few (Boeree, 1998). As stated by Heath and Martin (1990), "It is conceptualized as a continuum of liability to psychosis (principally schizophrenia and bipolar affective disorder) with 'psychopathy' (i.e., anti-social behavior) defined as 'a halfway stage towards psychosis'" (p. 111).

It was also found that the psychoticism scale correlates significantly with other hostility and tough-mindedness scales and traits such as non-acceptance of cultural norms, immaturity, and anti-authoritative attitudes. Higher psychoticism scores were also reported amongst psychopaths and criminals (Howarth, 1986). This reinforces the idea Eysenck described as his psychoticism scale.

Psychoticism as it Relates to Creativity

Many proposals have been given linking Eysenck's psychoticism scale to the level of creativity in people. It has been generally found that more creative people generally have higher psychoticism scores than people with a lower creativity. Eysenck (1993) stated, "I argue that intelligence is essentially characterized as a search process in order to discover a noegenetic solution, to use Spearman's (1923, 1927) term, bringing together different ideas from memory to produce new answers to problems" (p. 147). Eysenck argues further with Campbell, Funeaux, and Simonton who say that this search process is blind or random by saying that there is strong evidence that search processes of the kind stated are always guided by explicit or implicit ideas of relevance (Eysenck, 1993).

Eysenck (1993) poses his argument in relation relevance by saying that there are individual differences to the definition. By this he means that some people are over inclusive in their thinking, meaning they have a very broad conception of relevance, as opposed to people who have a much more narrow conception of what is relevant. He continues by stating that unusual responses to word association test could be used as a measure of this hypothetical quality. His suggestion is that individuals with an over-inclusive style of thinking have a larger sample of ideas on which a conclusion can be based. These people he concludes are able to come up with innovative, unusual and creative ideas more easily than people who have a more conventional view of relevance to a problem. Ultimately he believes that this could possibly be the base of creativity.

On the psychophysiological side, Eysenck (1993) finds that the over inclusive thinker is more likely to have a personality which relates to that of a schizophrenic or general psychotic disorder. The difference is that although the psychoticism score of the creative person is high just as that of someone diagnosed with a psychotic disorder, the creative person not necessarily diagnosable as a psychotic. He attempts to support his proposition by using evidence that there is a true relationship between insanity and genius, that truly creative people have a high psychoticism score on personality tests, and that unusual responses on a word association test are a good measure of psychosis, psychoticism, and creativity.

Eysenck (1993) gave some fairly distinct qualifications of what he is and is not attempting to prove with this theory. He stated, "...this theory explains (a) the observed differences between creativity as a personality trait (originality) and creativity as demonstrated by scientific or artistic achievement and (b) the difference between creativity and intelligence" (p. 148). He the qualifies this by stating that he is not suggesting that the theory he is attempting to develop is going to solve the "riddle of creativity" in a definite way, but rather will aid in making "testable predictions", bring together many existing theories of creativity and ultimately lead to improvements and new discoveries which will help make the theory more accurate and precise in the different parts that go into it.

To further describe the interrelation of creativity and psychoticism, Eysenck (1993) continued by saying, "Creativity from the earliest times has been thought to be related to psychosis or "madness" (Hyslop, 1925, Lange-Eichbaum, 1931; Lombroso, 1895; Nisbet, 1900), with some writers (Ellis, 1926; Juda, 1949) actually bringing forward some evidence from controlled studies to supplement these mostly anecdotal accounts (Hasenfus & Magaro, 1976)" (p. 156). Further evidence is presented stating that high levels of creativity have been found in descendents of psychotic parents. This was observed by Eysenck (1983) who stated that it was not psychosis but rather psychoticism, defined as a predisposition or susceptibility towards psychotic behavior, which was related to creativity. Eysenck (1993) is very clear in stating that his theory does not claim that psychosis produces creativity, or people who are highly creative people are therefore psychotic. He says instead that it is possible that a high psychoticism score is necessary for creativity, and that people with a high psychoticism score may develop a psychosis or suffer some psychotic episodes during their lives. This would be expected considering a high psychoticism score indicates a higher risk and susceptibility to psychotic episodes and behavior.

Corroboration of Eysenck's Psychoticism-Creativity Theory

Eysenck (1993) went into lengthy detail about the different ways to test the correlation of psychoticism and creativity, and ultimately finds five different ways of supporting the psychoticism-creativity model. First, he stated that persons genetically related to diagnosed psychotics are unusually creative. He cited genetic studies that corroborate this view as well as a previously made hypothesis which contends "that there is a common genetic basis for great potential in creativity and for psychological deviation."

The second supporting criterion is that psychoticism is related to tested creativity (also referred to as originality). In support of this contention, Eysenck cites a study performed by Woody and Claridge (1977). This study was performed at Oxford University with 100 graduate and undergraduate students as the test subjects. The students were administered the EPQ and Wallach-Kogan Creativity Tests (which were somewhat modified for the experiment). The Wallach-Kogan Creativity Tests measured two important things, number of responses which indicated fluency of the person, and number of unique responses produced by the subject which measured the originality of the person. It was found that psychoticism scores had a very high direct correlation to that of number of responses on the creativity tests. An even higher direct correlation was found between psychoticism scores and uniqueness scores in the subjects. This test was very good for this because it basically eliminated the factor of general intelligence based on the subjects selected for the study. Eysenck (1993) infers that the students used had IQs above 120 because creativity was significantly related to IQ up to a score of 120 at which point the two become independent. This test was partially replicated later by another group and the findings were similar to the previous study.

The third supporting study relates psychoticism to creative achievement. In citing a study performed by K. O. Götz and K. Götz (1979) it was found that both male and female artists had significantly higher psychoticism scores than their male and female non-artists counterparts. This is significant due to the fact that it uses actual artistic achievement as a criterion for the measurement of creativity and originality (Eysenck, 1993). This also serves to support the findings of Woody and Claridge as stated previously.

The fourth study Eysenck uses to support his claim deals with the idea that creative people often suffer from various psychopathologies. In backing this he simply points readers to a "much more detailed review...given by Prentky (1980), who also came to the conclusion that creative achievement is linked with psychopathology, although not with actual psychosis." The fifth line of support for this theory Eysenck (1993) describes as being "exceptionally important". This describes how identical cognitive styles are characteristic of psychotics, persons who score high on the psychoticism scale, and creative achievers. Eysenck notes countless examples of studies which support his supposition and he gives a great deal of detail relating to the importance of these studies, thus they will not be detailed again here. A list of the tests which were administered included Word Association Tests, EPQ tests, Word Halo Tests (which test the selection of words which the subject considers to have nearly the same meaning as presented "stimulus words"), Word Sorting Tests (which require subjects to group together words of similar meaning), and an Object Sorting test (which measures "allusive thinking") among others. Eysenck's research showed the results of these tests and studies to further support his theory.

Flaws in Eysenck's Theory: A Counterpoint

Eysenck (1993) recognizes that theories in psychology are far from perfect and realizes that by basing this theory off of some "fuzzy" theories, he is making the matter somewhat worse. However, he also states his goal is not to create a perfect theory here, but to "throw some light on a fundamental problem in science, art and possibly other fields as well."

Amabile (1993) makes an attempt to expand on Eysenck's ideas and actually describe what is needed for a theory of creativity. The first point she notes is that Eysenck fails to give an explicit operational definition of creativity as unique achievement. She argues that since the theory is designed to test for such creative productivity that it cannot adequately tested for without defining exactly how creative achievement is to be measured. Her next contention with Eysenck's theory proposal comes from Eysenck's use of word association tests and high psychoticism scores on the EPQ-R (Revised Eysenck Personality Questionnaire). Amabile feels that these data sources are "weak at best" and cites that there may be other "noncreative individuals (such as schizophrenics) which have these same personality traits."

Amabile (1993) further points out that within Eysenck's article, he never really gives a clear description of what his theory actually is. The section entitled "A Theory of Creativity" describes a notion that there is a definite link between psychoticism and creativity. She states that this section never leads to a formal theory with specific postulations, but it did appear to be the central theoretical theme to the theory that Eysenck was trying to describe. One improvement suggested by Amabile is that "formal statements of the ways in which the P-creative theory incorporates or intersects with the cognitive model of original thinking and the grand model of creativity development" (Amabile, 1993).

Many sections of the "theory" posed by Eysenck could use clarification or elaboration. Amabile's (1993) first example is that even though both highly creative people and highly psychotic people have shown as having high psychoticism scores, what differentiates these two groups is never clearly specified. Although Eysenck touches on this a few times by suggesting that one difference is that creative people simply have the absence of a psychosis. He later says that creative people have other personality traits which differentiate them from psychotic individuals including traits such as ego-strength. Amabile argues that since this is such a central issue in the theory, that it requires much more rigorous clarification. Another point which Amabile (1993) says requires clarification is the relationship between the personality variables and the cognitive variables. Questions such as "How does personality give rise to cognitive style?" and "How specifically, do both give rise to clarity?" require answering before the theory can be solidified.

One of the largest contentions that Amabile brings forth is the lack of breadth being as Eysenck's theory is almost entirely focused on psychoticism. Eysenck mentions in his article that there are several other relevant factors involved in creativity, but fails to identify them and give them descriptive coverage. Amabile (1993) goes on to suggest that Eysenck integrate some broader theories of creative achievement into his framework to allow for a more complete representation of creativity. Amabile concludes her arguments by saying that Eysenck's statements would be much stronger if, in addition to describing how his theory is testable, he were to elaborate and specify precisely what would constitute these crucial tests.

Eysenck's basis for a theory of creativity seems to have some very promising possibilities. Given the diversity of people's view of relevance to certain situations or types of media, the model could be used to describe creativity strength or weakness in certain specific fields of creativity. Unfortunately as Amabile pointed out, Eysenck's theory still has a way to go if it is going to be a truly unifying theory of creativity as it relates to personality and other psychophysiological characteristics. It is however a stepping stone to that end and will indefinitely lead to more debate over the topic.

I personally believe that Eysenck's theory has merit. Although Amabile pointed out some points which needed clarification, Eysenck did seem to defend his argument with valid data as well as various other studies that backed his findings. The psychoticism scale of personality appears to have a definite correlation to the presence of creativity in people. Hopefully in the future this will be the basis of the one unifying theory of creativity that Eysenck had described.

Peer Commentary

Though Important, Eysenck's Is Not the Only Theory

Jonathan S. Byrd
Rochester Institute of Technology

In Porzio's paper, he discussed at length Eysenck's theory of psychoticism and creativity. After presenting a concise definition of psychoticism, Porzio discussed how Eysenck related psychoticism to creativity. The paper then moved on to the theory itself, breaking down corroborating evidence from five studies. Lastly the paper discussed flaws in Eysenck's theory and counterpoints presented by Amabile. One of the problems in Eysenck's theory, as pointed out in the paper, is that not enough research has been conducted in the field. In fact, creativity itself is a vastly understudied area.

Eysenck's is one of the most recognized theories on the notion of creativity, but this does not mean that other theories can be dismissed. To critically analyze Eysenck's theory, a number of theories should be brought to light and discussed. The vast majority of the paper merely discussed and summarized Eysenck's theory. Although this was useful for giving the reader an idea of where the author was coming from, the overall effect of the paper was that of an introduction without a body or conclusion. Though the paper did well in bringing to light Amabile's objections to Eysenck's theory, the paper did not discuss how Eysenck's theory related to any other theory, nor did the author himself make many conjectures as to the benefits or flaws of Eysenck's theory. Other theories do not necessarily have to refute or corroborate Eysenck's theory, but a comparison among them would go far in the author's attempt to critically analyze Eysenck.

No corroboration was given for Eysenck's theory other than from Eysenck himself. This goes back to the original point that more needs to be looked at than just Eysenck's theory. Numerous corroborating theories fit in with Eysenck's model, but none was discussed or even brought up.

The paper did a very good job of introducing the reader to Eysenck's theory of creativity. It explained his theory and then gave corroborating evidence to show that the theory was sound. It was a very good introduction to the paper, and if only a little more were added in terms of the critical analysis, then it would have an excellent ending too. It was written very well, but the author needed to be a little more involved with dissecting Eysenck's theory and analyzing it.

It is clear that Eysenck has made enormous contributions to the field of psychoses and creativity and how they are linked. This cannot be disputed. The problem comes from the fact that Eysenck is not the only researcher who has made strides in this field, nor is he the only researcher who should be looked at when attempting to discuss such a broad concept as creativity.

Peer Commentary

Review Lacking Thought and Creativity

Brian A. McManus
Rochester Institute of Technology

Porzio did an adequate job of examining and summarizing Eysenck's (1993) theory. He then offered a counterpoint given by Amabile (1993) to cast doubt on the theory. His summation of the topic, "I personally believe that Eysenck's theory has merit," is hardly adequate, however. A well-presented plan for further substantiating Eysenck's theory should be formed. What specific research must be conducted?

The Oxford University study (Woody & Claridge, 1977) offered as support for Eysenck's theory was by no means a large enough study to be used as concrete evidence. A study of only 100 college students at a specific college was conducted. Porzio stated that another study was conducted partially replicating this study with similar findings. This study must be presented and discussed as further evidence.

Porzio discussed Eysenck's own lack of confidence in his theory. He also states that Eysenck's goal was not to create a perfect theory, but to throw some light on a fundamental problem in science, art, and possibly other fields as well. Porzio needs to discuss on which “fuzzy” theories Eysenck based his theory. Perhaps these theories are not so fuzzy anymore and could further validate Eysenck's theory.

Kline and Cooper (1986) challenged Eysenck's theory in a study using a different set of creativity measures. Their findings, along with others that refute Eysenck’s theory, should be presented to provide a clear understanding of both sides of the issue. One counterpoint was not enough. In general, Porzio needed to clarify which points in Eysenck's theory were weakest, what strong evidence corroborated his theory, and what further research needed to be undertaken.

Peer Commentary

A (Somewhat More) Critical Review of Eysenck's Theory of Psychoticism and Creativity

Matthew P. Rick
Rochester Institute of Technology

We humans have a tendency to revere those who come before us. This is not necessarily a bad thing, but it can be taken to the extreme, and that is where the whole practice falls down. So it is with Eysenck's theory of psychoticism.

That being said, it is fair to point out that the theory has much to recommend it. It separates conditions into discrete groups for ease of consideration, it pioneered the idea of spectrum disorders, and these are only a few of its good points. Sadly, there are two major problems with the theory. The first of these problems is the lack of definitional clarity in Eysenck's theory, and the second is the heterogeneity of the groups that Eysneck ends up defining through his model.

Porzio went into a great deal of detail covering the definitional problems in Eysenck's theory; one point that he did not mention, however, bears some discussion, and that is the definition of creativity that Eysenck's model allegedly measures. According to proponents of Eysenck's model, the inappropriate responses and the inability to properly filter information that individuals with high psychoticism posses cause them to perform creative tasks better than their more mundane counterparts. By testing just the "creativity" of the solution proposed, one quickly runs into the problem that a schizophrenic who spouts word salad in response to a word association would be determined to be possibly more creative than an author or painter who was psychologically normal. A better definition of creativity was proposed by Sternberg (1999), that creativity is the capacity to create a solution that is both novel and appropriate.

Although admittedly most of this problem stems from the poor definition of creativity, one must also consider the problem of heterogeneity in the groups that are created by the psychoticism measure. Heath and Martin (1990) stated that psychoticism "is conceptualized as a continuum of liability to psychosis (principally schizophrenia and bipolar affective disorder) with 'psychopathy' (i.e., anti-social behavior) defined as 'a halfway stage to psychosis'" (p. #). Although at first glance this may seem to bulwark the theory, there are two main problems with the line of reasoning. The first is the fact that psychoticism is defined as a vulnerability to conditions that seem to have very little to do with each other or, at the very least, are certainly very unlikely to share a common cause. It therefore should be easy to differentiate between these drastically different disorders. Clearly psychoticism measures some other variable but for the most part; in Porzio's paper, psychoticism is defined, in short, as possessing symptoms similar to those of antisocial personality disorder and schizophrenia. Furthermore, to say that one has found a relation between depressed people and depression is to make a non-statement, so one is forced to wonder how Porzio's statement, "higher psychoticism scores were reported amongst psychopaths and criminals," says any more.

Sternberg (1999) further stated that creativity seemed to be affected by both individual and developmental differences, as opposed to intelligence, which is a heritable trait. Because it is known that schizophrenia, at least, is largely biological, it seems supremely likely that creativity is in fact unrelated to psychoses and that psychoticism measures something very different from creativity, but something that perhaps shares traits with creativity. If this is a case for all the conditions that psychoticism attempts to measure, then it would then be fair to say that creativity is not a product of psychoses but is merely difficult to differentiate from them.

Author Response

Eysenck as Contributor, Not Founder

Shane K. Porzio
Rochester Institute of Technology

I felt that many of the commentators on my paper viewed it as describing Eysenck's theory as being put above others. This was not my intention. The paper was intended solely to give insight into Eysenck's theory, because Eysenck was a very large contributor to forwarding the study of creativity, and his theory of a creativity-psychoticism relation was a good starting point for a lot of researchers. I am by no means trying to paint Eysenck as being the founder of the field or the only theory that should be considered.

One major issue discussed was the fact that only one counterpoint was given to Eysenck's argument. To address Byrd, I did not feel that I needed to give a large number of counterpoints to Eysenck's theory in a review of this theory, and I felt that by giving one, I would encourage interested investigators to conduct more research on the topic form their own views. I did not feel that adding counterpoints by other authors, of which there were many, would add to the presentation of this topic. Because this theory was not my own and was presented as material to help stimulate further interest in the topic, I feel that the content I provided was enough to accomplish this goal.

In response to Rick's criticism of relating the high psychoticism scores of psychopaths and criminals to depressed people having depression, I must counter by clarifying that depressed people suffer from the disorder of depression, whereas psychoticism is not at all a disorder in any sense of the word. Psychoticism as it was defined by Eysenck and in this paper is solely a dimension of personality, just as neuroticism and extraversion are. Therefore, correlating scores from different groups of people, such as criminals and psychopaths, is essential in showing that the trait researchers are trying to measure actually measures what they expect it to. This is no less important than saying that people who produce high scores on neuroticism on a personality test are more likely to be neurotic. If this were not demonstrated, then one could not apply the results of the test to any other situation.

I agree with Rick, however, in his assessment that Eysenck's definition of creativity is at the very least sub-optimal. Sternberg (1999) did a much better job by including the fact that for something to be creative, it must not only be novel but must also be appropriate to the situation. This could have made a very good and very long addition to this paper, but I also felt that it did not fit into the overall purpose, which was to present Eysenck's theory. In the end, I felt that the paper accomplished what it was intended to accomplish, which was to create interest in the topic and make researchers conduct more investigations and form their own opinions, which all the commentators seem to have done.

In response to McManus, countless other studies need to be conducted to substantiate Eysenck's findings. It is very difficult to develop a sound and irrefutable theory on a topic as broad and as understudied as creativity. It is also difficult to conduct studies on something such as creativity, because whether or not something is creative is fairly subjective. As far as the study that was conducted that mimicked the Oxford University study (Woody & Claridge, 1977), this was described in Eysenck's (1993) article, which should be consulted for more information regarding this study.

Lastly, I would like to thank all the peer commentators for their honest opinions and feedback. This practice is what drives us all to become better authors and causes us to reevaluate our works as time passes, making things more beneficial for everyone.


Amabile, T. M. (1993). What does a theory of creativity require? Psychological Inquiry, 4, 179-181.

Boeree, C. G. (1998). Hans Eysenck (1916 - 1997) (and other temperament theorists). Retrieved October 10, 2003 from http://www.ship.edu/~cgboeree/eysenck.html

Eysenck, H. J. (1983). The roots of creativity: Cognitive ability or personality trait? Roeper Review, 5, 10-12.

Eysenck, H. J. (1993). Creativity and personality: Suggestions for a theory. Psychological Inquiry, 4, 147-178.

Götz, K. O., & Götz, K. (1979). Personality characteristics of professional artists. Perceptual & Motor Skills, 49, 327-334.

Heath, A. C., & Martin, N. G. (1990). Psychoticism as a dimension of personality: A multivariate genetic test of Eysenck and Eysenck's psychoticism construct. Journal of Personality and Social Psychology, 58, 111-121.

Howarth, E. (1986). What does Eysenck's psychoticism scale really measure? British Journal of Psychology, 77, 223-227.

Kline, P., & Cooper, C. (1986). Psychoticism and creativity. Journal of Genetic Psychology, 147, 183-189.

Sternberg, R. J. (1999). The theory of successful intelligence. Review of General Psychology, 3, 292-316.

Woody, E., & Claridge, G. (1977). Psychoticism and thinking. British Journal of Social and Clinical Psychology, 16, 241-248.

Last modified November 2003
Visited times since November 2003

Home to Personality Papers

Home to Great Ideas in Personality