The Idea of the Creative Personality

The following post was originally written and submitted in April 1994 as a term paper for the course ‘Artificial Intelligence and Creativity’, taught by Prof. Margaret Boden as an optional module of Sussex University’s Knowledge-Based Systems MSc programme in academic year 1993–94. This version incorporates several very minor changes, which don’t affect the substance of what is said. I’ve also added several notes to draw the reader’s attention to particular points.

“What, if anything, is there about the psychological make-up of some individuals that enables them to produce great creative works?” (Weisberg 1993, p.6)

This is one of the key questions we must ask and attempt to answer once we reject the notion that outstanding creative accomplishment is the result of divine inspiration, or is, more generally, materially undetermined. However, perhaps it is too general. A more specific question, which emphasizes the fact that creative acts occur at particular times and places, within a particular cultural and social milieu, is: can one plausibly suppose that anyone of moderate cognitive endowment and devoid of mental deficit would, if they found themselves occupying the same sociocultural circumstance which led a historically eminent individual to make some significant discovery, indeed make that discovery? The short answer, we suspect, is ‘probably not; it all depends …’ — but on what? If we call the original discover O and the hypothetical person who, we suggest, might make the same discovery H, then how much of what O knows does H need to know? To what extent must H have experienced the same past as O? Pressing the point, to what extent must H’s mental structures be isomorphic with those of O? Before long we find confronting knotty problems to do with cognitive ability, styles of thought, motivation, personality …

Consider, for example, the discovery of X-rays by Roentgen, described by Kuhn (1962) (and also by Price (1975)). One day Roentgen happened to notice that a barium platino-cyanide screen some way from a shielded cathode-ray discharge apparatus glowed during discharge. Intrigued, Roentgen performed a number of experiments to investigate the glow phenomenon. These established amongst other things that the cause of the glow propagated in straight lines from the cathode-ray tube and that it was unperturbed by a magnetic field. Kuhn notes that “At least one other investigator had seen the glow and, to his subsequent chagrin, discovered nothing at all” (p.58).

This story seems to exemplify the reverse of the phenomenon described by Boden (1991), whereby “Historians of science and art are constantly discovering cases where other people, even in other periods, have had an idea popularly attributed to some national or international hero” (p.3). The way our intuitions lie is evident from the surprise we feel on learning of parallel or independent discovery like that, a surprise which is entirely absent when we hear of the X-ray case. Notwithstanding Weisberg’s derision of the ‘myth of genius’ (Weisberg 1993), we have a deep-seated belief in the idea of the creative personality.

In the 1950s and 1960s much research was carried out which sought to isolate the psychological characteristics of creative individuals. Some of the studies can be criticized on the grounds of flawed subject selection. For example, it was often assumed that performance on tests of ‘divergent thinking’ is a measure of creativity, and so such tests were used for selecting subjects. If, however, performance on such tests does not correlate especially strongly with ‘real-life’ creative accomplishment (as seems to be the case) then one is simply studying the psychological characteristics of divergent thinkers.

The 64 scientists investigated by Roe (1952) were selected by a panel of experts; collectively the scientists had “received a staggering number of honorary degrees, prizes and other awards”. Roe applied a battery of tests and conducted exhaustive interviews of her subjects, and, notwithstanding wide individual variation, a ‘more than chance’ profile of the eminent scientist emerged:

“He was the first-born child of a middle-class family, the son of a professional man. He is likely to have been a sickly child or to have lost a parent at an early age. He has a very high IQ and in boyhood began to do a great deal of reading. He tended to feel lonely and ‘different’ and to be shy and aloof from his classmates … What decided him (almost invariably ) (to become a scientist) was a college project in which he had occasion to do some independent research — to find things out for himself … He works hard and devotedly in his laboratory, often seven days a week … Better than any other interest or activity, scientific research seems to meet the inner need of his nature.”

The remark about parental loss refers to the finding that some 15% of the scientists had lost a parent before the age of 10, rising to 26% by adulthood. Albert investigated this phenomenon specifically (Albert 1971), and found comparable figures for parental loss amongst the historical geniuses identified by Cox.[1] As they included individuals eminent in the arts, humanities and the military, in addition to scientists, the possibility arises that there may be common factors underlying creative accomplishment in general. Albert speculates that one consequence of losing a parent in childhood is the early development of emotional and intellectual independence. The mechanism by which this comes about is not elaborated explicitly, although he sees the role played by an extended family or by ‘interested others’ who interact with the child in a way that is “adult-like, reasonable, and open to disagreement” as an important one (p.24). A possibility not mentioned is that the death of one parent alters the nature of the relationship of the remaining parent with the child; it seems reasonable to suppose that they might often be brought closer by the loss. Storr, citing Winnicott and Bowlby, argues that the formation of a ‘secure attachment’ in early life fosters the inner security needed if the ‘capacity to be alone’ is to flourish (Storr 1989, pp.18–21). He contends that this capacity is, if not a sine qua non of creative achievement, at least a common prerequisite, for reflective thought and imaginative play surely occur most readily and stably during periods of anxiety-free solitude, as Schopenhauer recognized:

“The imagination is, consequently, the more active the fewer perceptions from without are transmitted to us by the senses. Protracted solitude, in prison or in a sick bed, silence, twilight, darkness are conducive to it: under their influence it comes into play without being summoned. On the other hand, when a great deal of real material is provided from without for us to perceive … then the imagination takes a holiday and refuses to become active even when summoned: it sees that this is not its season.” (Hollingdale 1970, p.177)

Weisberg’s interpretation of the parental loss data is rather different, with personal intellectual activities becoming emphasized after a withdrawal from social contacts (Weisberg 1993, p.259).

MacKinnon (1962), in his study of the personalities of successful architects, selected subjects by peer nomination. He chose to investigate this professional group on the assumption “that they might as a group reveal that which is most generally characteristic of creativity and the creative person”, for the practise of architecture demands both artistic-aesthetic and technological modes of thought. A panel of five professors of architecture was asked to nominate the 40 most creative architects and place them on a ranked scale of creativity; 86 names were given. Each panel member was then asked to rank the names provided by the other panel members. The rankings were combined to obtain an overall ranked list of the most creative architects; this group was termed Architects I. Then the 1955 Directory of Architects was surveyed to obtain two further groups. Architects II consisted of architects who had worked for at least two years with a member of Architects I, while the members of Architects III had never worked with any of the nominated creative architects. Indices based on appearance in the architectural literature bore out expectations that Architects I were the most prominent group and Architects III the least.

The three groups were subjected to a number of personality and other psychometric tests. The Study of Values test measures the relative importance placed by an individual on the theoretical, economic, aesthetic, social, political and religious. All three groups of architects showed a clear preference for the aesthetic and theoretical, and placed low importance on the economic and social. However, this pattern was exhibited most strongly by Architects I and most weakly by Architects III. A similar ranking of the groups occurred on the Barron — Welsh Art Scale of the Welsh Figure Preference Test. This consists of a series of more-or-less abstract drawings, ranging from the simple and symmetric to the complex and highly asymmetric. The test was originally standardized on a group of painters, who showed marked preference for the complex, asymmetric figures; non-artists preferred the simpler, more symmetric figures. The architects showed similar preferences to the artists, with Architects I being the most marked in this respect, Architects III the least. At the very least, this result suggests that it is possible to discern common aesthetic values in several groups which serve to distinguish them from the general population.

MacKinnon also reported that, besides their high femininity rating on a measure of masculinity — femininity, the creative architects were notable for their desire for solitude: on one test “the creative architects revealed less desire to be included in group activities than any other group we have studied”. Also pronounced was the independence of judgement they exhibited at college, such that “in work and courses which caught their interest they could turn in an A performance, but in courses that failed to strike their imagination, they were quite willing to do no work at all”.

Getzels and Csikszentmihalyi (1976) studied the personalities of a group of student artists, following them through the start of their careers. By using a range of tests it was possible to identify elements of personality which distinguish art students from the general student population. The profile in many ways resembles that of MacKinnon’s creative architects:

“Young artists, while still students, already tend to be reserved, amoral, introspective, imaginative, radical and self-sufficient, and tend to possess attitudes usually associated with the opposite sex. They hold aesthetic values in high regard, and neglect economic and social values — a pattern that contradicts the ethos of the culture in which they live. They do not differ substantially from college students in intelligence as measured in conventional tests, but are far superior in spatial and aesthetic perception” (quoted in Perkins 1981, p.264)

More than this, these qualities were manifested more strongly by fine arts majors than by the commercial (advertising and design) and education majors also tested, and most strongly of all by the most successful of the fine artists. Perkins (1981) concludes that “Getzels & Czikszentmihalyi may have caught the essence of the creative aesthetic personality, at least as it occurs in our culture”. (p.264)

In the follow-up study, conducted seven years later, Getzels & Czikszentmihalyi looked at the relationship between early career success and the original personality measures. It was found that none of the measures correlated well with career success except one: the artistically successful individuals had measured lower on a scale of conformity and need for social approval. However, Weisberg cautions that “given the large number of measures involved … a significant difference on a single factor could very well be due to chance” (Weisberg 1993, p.78). (Unfortunately, Perkins does not mention the follow-up study.[2])

It is interesting to compare the conclusions Perkins and Weisberg reach after consideration of these and similar studies. For Perkins, they affirm the proposition that “creative artists are amoral bohemians” while scientists are “cold assertive dwellers in ivory towers” (Perkins 1981, p.268). His verdict is that

“The studies do reveal some interesting differences from field to field. They also demonstrate that there is such a thing as a creative personality. It might have turned out that creativity involved only inventiveness in a profession, rather than qualities pervading the whole person. Perhaps of most interest, the efforts to relate creativity to personality have been more successful than the efforts to relate it to abilities. This affirms the importance of matters other than ability in creativity.” (p.269)

Weisberg takes a rather different line. His aim is to debunk the idea of the creator as hero and to show that “making sense of the development of creative works requires no elaborate construction, because creativity is firmly rooted in past experience and has its source in the same thought processes that we all use every day” (Weisberg 1993, p.3). As a prelude to this he sets out in detail what the idea entails:

“In the genius view, that some individuals produce great works while others do not is seen to also be based on a personality structure that allows such special thinking to flourish.” (p.8)

Yet elsewhere he appears to grant the importance of what might reasonably be thought of as aspects of personality structure:

“Comparisons of creative and non-creative individuals have yielded a small number of characteristics shared by all creative individuals. Among these characteristics are broad interests, independence of judgment, self-confidence, intuition, and a firm sense of the self as ‘creative’.” (pp.72–3)

Quite what Weisberg takes personality to be about is never made clear. At times he appears to exclude motivational factors as being important ingredients:

“… examination of the results (of studies of the creative personality) reveals that many of the isolated ‘personality’ characteristics of genius have no content; they are basically motivational in nature.” (p.76)

Elsewhere, cognitive attributes are separated out (“… the result of cognitive and personality characteristics …” (p.84)). These exclusions not only render Weisberg’s definition of personality unclear; they distinguish his use of the term from usages more commonly encountered. Boden, for example, discusses multiple personality disorder in terms of “different cognitive — motivational streams of behaviour and experience” (Boden 1993, abstract) and refers to the “cognitive — motivational integration criterial of ‘personality’”.

Before going any further it would seem sensible to explicate the notion of personality. Somewhat curiously, there is no entry for the word in (Gregory 1987), yet, as the problems of meaning already encountered make plain, it is not as if one can take the term for granted. What might be termed the ‘classical’ view of personality is couched in terms of habits, traits and dimensions, these labels serving to divide up what is conceived as a complex hierarchical structure. Habits are consistent, situation-specific behaviours, whilst a trait is a group of related habits. Zuckerman’s (1991) example is that hand-washing is a habit but cleanliness is a trait. Traits range from the narrow and highly differentiated, subsuming relatively few habits or situations, to the broad, which cover behaviour over large classes of situations. Related traits are grouped together as a supertrait, and at the top of the personality structure hierarchy are a few supertraits that meet a number of criteria; these are so-called basic dimensions of personality. The exact criteria which must be fulfilled is a matter of debate, but generally they include evidence of the trait’s heritability and the existence of an analogous trait in (at least one) other, preferably social, species. Increasingly it is held that the trade must be accompanied by biological markers (e.g. correlated hormone levels).

A number of trends characterize the ascent of the behavioural hierarchy of traits. At the higher levels of behavioural differentiation, environmental effects are more influential and past experience is an important determinant; there is thus a degree of behavioural lability over time. By contrast, basic dimensions and very broad traits are more likely to be inherited and are less susceptible to temporal change. Such broad, stable traits are often referred to as aspects of temperament, in distinction to the less stable, more plastic traits which constitute personality. The difference is sometimes stated in terms of style of behaviour, a matter of temperament, and content or direction of behaviour, which is specified by personality, although this distinction can be criticized on a number of grounds (Zuckerman 1991, p.6). In general, personality becomes increasingly stable with age.

Interesting research on the physiology of temperament has been conducted by Jerome Kagan and colleagues (Kagan et al., 1987). On the basis of response to unfamiliar situations, two extremal groups of children were selected at 21 or 31 months of age. One was characterized as being extremely cautious and shy, the other as fearless and outgoing. Kagan proposes that the contrasting behavioural responses of the two groups has its origins in neurological differences, and he adduces physiological data in support of this idea. Specifically, the inhibited group show a lower threshold of reactivity in parts of the limbic system, especially the amygdala and hypothalamus, resulting in the heightened activity of the pituitary — adrenal, reticular activating, and sympathetic nervous systems. This manifests itself in elevated cortisol production by the adrenal cortex, contraction of the muscles of the larynx and vocal system (with consequent modulations of voice pitch and stability), elevated heart rate and blood pressure, and a number of other effects. The lowered limbic activation threshold might in turn be explained by the excessive production of norepinephrine by the locus coeruleus.

The patterns of behaviour that distinguished the two groups of children, and the associated physiological responses, were found in many cases to persist through to the sixth year of life. However, some of those who were inhibited at 21 months were no longer inhibited at six years. It is suggested that a negative parental and societal reaction to excessively fearful behaviour prompts the self-aware inhibited child to seek consciously to moderate his response to novelty. It seems reasonable to suppose that the neurological factors that underlie the nature of the behavioural response to unfamiliar situations are governed by early developmental processes, and may even have a genetic basis. There would then exist a parallel with a number of other traits including handedness and dyslexia (Geschwind 1984).

Mischel has argued, from the fact that the cross-situational consistency of measured traits is often quite low, that the regularities we see in the behaviour of others are little more than the product of our own ability to categorize objects and events according to complex property patterns (Zuckerman, p.70). Put another way, personality is more to do with the observer than the observed. But this is surely a very puzzling idea. For, we tend to say, either there are behavioural patterns and regularities or there aren’t; and if there are regularities, then what causes them? This debate find strong echoes in the philosophy of mind, where the status of folk psychology, “the perspective which invokes the family of mentalistic concepts, such as belief, desire, knowledge, fear, pain, expectation, intention, understanding, dreaming, imagination, self-consciousness, and so on” (Dennett 1987) is a current concern. In using everyday psychological terms to account for the behaviour of others we make certain assumptions:

“… explanations of actions citing beliefs and desires not only describe the provenance of the actions, but at the same time defend them as reasonable under the circumstances. They are reason-giving explanations, which make an ineliminable allusion to the rationality of the agent.” (Dennett 1987, p.48)

Making this rationality assumption constitutes taking the ‘intentional stance’ (Dennett 1987, p.15). Doing so allows us to predict behaviour as well as explain it post hoc: given that agent A believes b and desires d, and assuming that A is rational, then we are justified in expecting A in situation S to select a behaviour B from a set of acts rational for S.

It is clear that the notion of personality plays a significant, but not readily specifiable, role in folk psychological explanation. By attributing a certain personality to someone, we are implicitly saying something about the way that person will behave in certain situations; the attribution is like stating a generative rule about the person’s dispositions. Furthermore, the personality attributed seems to determine at least in part what behaviour we would consider to be rational given some set of circumstances. Morton suggests that the connotations of intelligibility inherent in this use of the concept of rationality in turn depends on imaginability:

“… one’s understanding of what particular people are like … consists very roughly, of what one would have to know in order to begin imagining how they do what they do.” (Morton 1980, p.149)

Being able to understand concepts or personality and character, Morton conjectures, requires mastery of the concept of mind used in one’s culture. How this might occur he does not say (beyond that it requires “the correlation of terms of character with styles both of one’s own and of other people’s actions”), but it is tempting to suppose that it is not unlike the route to language acquisition proposed by Chomsky. That is, the complexity of the tacit theory of mind we employ, and the difficulty we have in articulating it, contrasts markedly with the effortlessness with which it is acquired (Sloman makes a similar point (Sloman 1987)). Chomsky takes the parallel features of language to be evidence for an innate, general grammatical machinery which linguistic experience fits to the specifics of any particular natural language. Evidence that there may analogously exist innate machinery underlying the development of a concept of mind comes from the recent finding that autistic children lack such a conceptual framework. The hypothesis is then that the developmental abnormalities criterial of autism damage the neural structures relevant to that framework’s mastery.

To be told about someone’s personality is to gain some understanding of that person. How does this come about? Morton’s view is that

“The function of concepts of character in psychological explanation is … to block, qualify, or emphasize the connections between the beliefs and desires that an explanation appeals to and the actions it explains. The idea is essentially that the use of these concepts has an overriding aim, that of fitting general patterns of explanation to the quirks of the particular case.” (Morton 1980, p.152)

It is interesting to note the similarity between the roles of character concepts and scientific theories. The latter also relate the particular to the general, by showing how apparently disparate phenomena exhibit an underlying unity (Popper 1979, pp.262–3). Both yield an understanding which can be defined in terms of explanatory and predictive power. The association of the general with the particular in character attributions is not at all straightforward, for “if a character term, say ‘haste’, characteristically serves to pick out a class of intentions from which the person to which it applies may act (e.g. haste rules out the motives for double-checking what one does, and reinforces the motives of speed and efficiency), then the range of actions in whose explanation it can figure may not have any simple unity at all: they may, for example, be the same as the actions associable with some other term” (Morton 1980, p.153).

The question of how there is a predictively reliable folk psychology at all is, I think, closely related to Mischel’s concerns about personality already mentioned. Indeed, if it is right that folk psychology subsumes personality and character attributions, then an account in relation to folk psychology is all that is sought. Dennett’s view is surely plausible: the intentional stance works because “the internal processes of the system (i.e. the individual agent) mirror the complexities of the intentional interpretation, or its success would be a miracle” (Dennett 1987, p.60). But, he says, the mirroring which occurs need not be in any way straightforward; it is not obvious that there must be an exact correspondence between the propositional attitudes attributed to rational creatures on the one hand and their brain states on the other, such “that the logical form of the propositions believed will be copied in the structural form of the states in correspondence with them” (p.34) — Fodor’s ‘language of thought’ hypothesis. To think that it is obvious, Dennett contends, is to confuse the behavioural patterns picked out by folk psychological explanations with the pattern of the originating brain states. Dennett’s point finds echoes in Morton’s caution that

“It (folk psychological explanation) is also by its nature a business that must often fail. For the facts, the real psychological facts, about an agent’s state of mind and relation to the surroundings, are what makes an explanation apply. A scattered set of true ascriptions of beliefs and desires can rarely be expected to capture enough of these facts, even when these beliefs and desires are in some other person responsible for a similar action.” (Morton 1980, p.161)

What has been clarified by this philosophical interlude? The most important thing, I think, is that by situating the concept of personality within folk psychology, Mischel’s concerns are shown not to be groundless; if he is mistaken, it is not for trivial reasons. Dennett and Morton caution us that the mapping between observable behaviour and internal structure need not be straightforward. Consider a new-born child. Viewed abstractly, it is an exceedingly complex system whose internal structure is the result of genetically specified developmental processes unfolding within a certain environment. It is not too inaccurate to say that the genes determine the structure which would result from canalization under some notional environment, and that the precise details of the actual environment dictate the pattern of deviation from that hypothetical structure. The uniquely structured brain that results, and which is coupled to the world via sensorimotor systems, effectively equips the child with an embryo personality, a set of cognitive proto-skills and motivations which interaction with the world will differentiate. The way the child begins to interact with the world depends on its internal structure, and every transaction with the world changes both that structure and the child’s situation; the inner and outer structures form a single closed system. This is the basis for what Woodman and Schoenfeldt call the ‘interactionist perspective’ on creativity (Woodman and Schoenfeldt 1987). Implicit in that approach is the belief that creative acts do not occur spontaneously and irrespective of time or place, and that the more regard one pays to the cultural position, background and more immediate circumstances of creative individuals, the less it is possible to see their works in such a light. The same spirit informs Basalla’s evolutionary approach to the history of technology (Basalla 1988). However, there is no contradiction between maintaining such an attitude and seeking to identify what it was about certain individuals that led to their being at a certain place, at a certain time, thinking or doing certain things. That, I maintain, is about how the inner mind-brain engages with the outer world, and this we describe using the language of personality.

Notes

[1] Journalist Matthew Parris discusses the same phenomenon in his book Fracture: Stories of How Great Lives Take Root in Trauma (Profile Books, 2020)

[2] This is not surprising, given that Perkins was writing several years prior to publication of the follow-up study.

Bibliography

Albert, R.S. (1971) Cognitive Development and Parental Loss Among the Gifted, the Exceptionally Gifted and the Creative. Psychological Reports 29: 19–26.

Basalla, G. (1988) The Evolution of Technology. Cambridge: Cambridge University Press.

Boden, M.A. (1991) What is Creativity? University of Sussex, CSRP 195.

Boden, M.A. (1993) Multiple Personality and Computational Models. University of Sussex, CSRP 299.

Dennett, D.C. (1987) The Intentional Stance. Cambridge, MA: MIT Press/Bradford Books.

Geschwind, N. (1984) The biology of cerebral dominance. Cognition 17: 193–208.

Getzels, J. and M. Csikszentmihalyi (1976) The creative vision: A longitudinal study of problem finding in art. New York: John Wiley.

Gregory, R.L. (ed.) (1987) The Oxford Companion to the Mind. Oxford: Oxford University Press.

Hollingdale, R.J. (ed.) (1970) Schopenhauer — Essays and Aphorisms. Harmondsworth: Penguin.

Kagan, J., J.S. Reznick, and N. Snidman (1987) The Physiology and Psychology of Behavioural Inhibition in Children. Child Development 58: 1459–1473.

Kuhn, T.S. (1962) The Structure of Scientific Revolutions. Chicago: University of Chicago Press.

MacKinnon, D.W. (1962) The personality correlates of creativity: a study of American architects. Proc. Fourteenth Congress on Applied Psychology, vol. 2, Munksgaard, pp.11–39; reprinted in abridged form in P.E. Vernon (ed.) Creativity, Harmondsworth, Penguin, 1970.

Morton, A. (1980) Frames of Mind. Oxford: Clarendon Press.

Perkins, D.N. (1981) The Mind’s Best Work. Cambridge, MA: Harvard University Press.

Popper, K.R. (1979) Objective Knowledge — An Evolutionary Approach (revised edition). Oxford: Clarendon Press.

Price, D. de S. (1975) Science Since Babylon (enlarged edition). New Haven: Yale University Press.

Roe, A. (1952) A psychologist examines sixty-four eminent scientists. Scientific American 187: 21–25; reprinted in P.E. Vernon (ed.) Creativity, Harmondsworth, Penguin, 1970.

Sloman, A. (1987) Motives, Mechanisms, and Emotions. Cognition and Emotion 1: 217–233.

Storr, A. (1989) Solitude. London: HarperCollins/Flamingo.

Weisberg, R. (1993) Creativity — Beyond the Myth of Genius. New York: W.H. Freeman.

Woodman, R.W. and L.F. Schoenfeldt (1987) Individual Differences in Creativity — An Interactionist Approach, in A Handbook of Creativity: A Festschrift in Honor of E.P. Torrance, New York: Plenum.

Zuckerman, M. (1991) Psychobiology of Personality. Cambridge: Cambridge University Press.