Creativity has long been linked with psychological disorders albeit in very different ways. The hallucinations and delusions associated with schizophrenia were often regaled as divine inspiration and the clich of the mad genius was no doubt propagated by the eccentricities of eminent schizophrenic individuals. On the other hand, in autism the so called idiot savants demonstrated mentally retarded individuals with exceptional abilities in a specific area.
Though now thought of as quite different disorders, historically autism and schizophrenia have been tightly intertwined. Bleuler (1911) coined the term "autism" to describe social withdrawal as a negative symptom of schizophrenia. In his classic description of children with autism, Kanner (1943) made a convincing case that a pervasive disposition to aloneness could occur in the absence of psychosis but it wasn't until Kolvin's (1971) differentiation of autism and schizophrenia based on age of onset that this idea began to enter the psychiatric mainstream, capped in 1980 by the inclusion of autism as a diagnostic category distinct from schizophrenia in DSM-III.
Recently, a few researchers have begun to postulate the idea that autism and schizophrenia may in fact be linked disorders. Kraepelin originally diagnosed autism as the negative symptoms seen within schizophrenia and for a long time the disorders were merged as part of the same thing. Researchers today seem to be thinking that perhaps Kraepelin's original diagnosis was a lot more accurate than modern day psychologists suggest.
Nettle's evidence supports this idea and there have been some very recent genetic studies that further elucidate this thought. In particular, Crespi and Badcock have put forward a convincing argument that autism and schizophrenia are just opposite ends of the same spectrum of a dysfunction of social mind.
Although much of their hypothesis remains to be confirmed, a large portion of the research today is in support of this direction. If there truly is a link between autism and schizophrenia, and creative ability is seen in both disorders, it seems an investigation into what potentially links the creativity between them could only help elucidate the debate further.
CreativityDispelling the Myths
Creativity has been a topic of interest in psychology for over fifty years and yet despite literally hundreds of articles being written about it, we reach 2010 still without a consensus on what it is that makes a person creative or even a real definition of what creativity is. Many theorists have attempted solutions, some equating creativity with divergent thinking while others claim it is altered states of consciousness that cause our creative bursts. Other attempts have included the right hemisphere as the seat of creativity, the prefrontal cortex as the seat of creativity and creativity occurring only in states of unfocussed attention. The problem is that all of these attempts are right to some extent and yet by themselves also very very wrong. To quote Arne Dietrich,
"For every problem, there is one solution which is simple, neat and wrong."
There is nowhere this thought applies better than to psychology, where nothing is ever as simple as you'd hoped. The problem with all these supposed measures and definitions of creativity is that they don't even begin to encompass its complex nature. Also we too often focus on the extreme examples when we speak about creativity; names like Pablo Picasso and Albert Einstein are thrown around all too readily in the creativity debate. No one can deny the impact that the creative abilities of these two titans had in their respective fields but what about everyday creativity? What about the students that manage to solve a particularly troublesome maths problem or...
The focus on the extremes of creative ability is, I believe, part of the reason we lose sight of the notion of creativity and replace it with one single over simplistic measure of 'creativity' instead. Take the notion of divergent thinking for example. It has been proven many times over that individuals in artistic domains such as writers, musicians and artists are much more likely to score highly on tests of divergent thinking than their scientific counterparts, with the more artistic among them scoring more highly then the mildly artistic. This simple fact has since been mutated into the idea that the most divergent thinkers are the most creative with an implied connotation that the optimum level of creativity would be achieved by being a completely divergent thinker; a myth that has been so well propagated by the media and pop-psychologists that the terms have almost become interchangeable in serious psychology. The following quote was taken directly from the Wikipedia page on creativity:
"Divergent thinking is sometimes used as a synonym for creativity in psychology literature"
If we take a step back for a moment and consider the various ways normal individuals are creative then we can instantly see this focus on extreme divergent thinking meaning the same thing as exceptional creativity for the myth that it is. We can't say that creativity lies in the minds of divergent thinkers since no-one ever is a completely divergent thinker.
Of course people tend to lean one way or the other but I would argue not only that it's not possible to be a purely divergent thinker, but also that if one were to exist they wouldn't be creative at all. Creativity requires convergent thinking just as much as it requires divergent thinking. How can someone put their divergent thinking skills to good use without the ability to bring all their wonderful ideas together into a coherent creative act? How could a convergent thinker follow all the logical steps towards a problem's solution without making that one original, divergent association for the resulting 'eureka' moment?
As Dietrich (2007) discusses, at the very best, measures of divergent thinking correlate with a certain type of creativity in a certain type of person but what we really need to do is discover what part of the creative process it is that divergent thinking tests tap. In order to better understand creativity we need to discover the particular parts of divergent thinking that associate with creativity and then we need to find out why. We need to investigate the underlying mechanisms that facilitate creativity in the form of divergent thinking.
Interestingly when we attribute creative ability to one brain state or another we often don't pause to think of all the instances individuals are in such brain states and produce nothing worthwhile or creative. Creativity can't be exclusively associated with divergent thinking or defocused attention or any of the other half-hearted attempts at defining it because all these things provide a basis for uncreative thinking as well.Redefining creativity
This section is devoted to explaining Dietrich's 2004 model of creativity. This is what I will be basing my final explanation of the link between creativity in autism and creativity in schizophrenia on.Creativity in autism
This section will summarise the stereotypical kinds of creativity seen in autism (scientific, mathematical, convergent thinking) but will also discuss how other types of creativity are also seen within this disorder. Donna Williams is given as an example of an artistically creative autistic individual and her work is compared below with the work of Peter Myers, another autistic artist. I make the suggestion that it is highly likely that a large amount of artistic creativity within ASD individuals would still fall under the heading of cognitive or mechanistic or scientific due to its emphasis on exceptionally detailed replication of an object or scene in real life (see Stephen Wiltshire). Peter Myers' artwork would also potentially fall under this heading due to the detailed and dogmatic replication of pattern.
However, there are always exceptions to the rule and Donna William's artwork is representative of this with its slightly abstract, representational style.
The point of this section is to highlight the tendency of creativity writers to pigeon hole the autistic population into the 'science' box when there are clear exceptions to the rule that a comprehensive theory of creativity needs to address.Creativity in schizophrenia
Similar to the creativity in autism section this section summarises the kinds of creativity typically seen in schizophrenia and bipolar disorder as generally being confined to the arts such as painting, writing etc. Notable exceptions are then discussed such as John Nash in order to make the point as above.
Spectrums of disordersThe autistic spectrum
This section summarises the large number of disorders that display autistic symptoms or have been linked to autism and puts forward a suggestion for a much more inclusive autistic spectrumThe psychotic spectrum
Similarly this section summarises the large number of disorders that display psychotic symptoms as well as discussing the evidence for the link/ spectrum model of schizophrenia and bipolar disorder. An inclusive spectrum involving bipolar and schizophrenia amongst others is suggested here.An autism and schizophrenia spectrum?
A view that is beginning to emerge in certain researchers across the literature is that autism and schizophrenia are intrinsically linked in some way and some have begun to suggest that they may be diametric disorders that represent two extremes on a cognitive spectrum that has normality at its centre (see figure).
The original suggestion for this idea can be traced back to Nettle's 2005 paper in which he was investigating the relationship between mental health and creativity. Discussing the fact that poetry and artistic creativity were more associated with divergent thinking and schizophrenia while mathematics was associated with convergent thinking and autism, Nettle observed that the two disorders seemed to mirror each other and concluded with the statement,
"The constellation of autism, systemising and science appears to be in many respects the opposite tail of the distribution to the constellation of arts, unusual experiences and affective and psychotic disorders explored in the present study."
Since then the view as acquired a number of supporters and a large amount of data has been gathered in support of the idea. Two ardent proponents of this view are Bernard Crespi and Christopher Badcock and they have written multiple papers on the matter. In their 2008 paper they presented a table illustrating the diametrically opposed phenotypes of autistic and psychotic conditions which is summarised in table 1 on the following page. The main emphasis of this model is on the two disorders being opposite dysfunctions of the social mind but other anatomical data is also included.
The first sign both of the presence of a disorder and the beginnings of how they differ can be detected at birth. There are a large number of growth deficiencies with the brain size being smaller than average due to reductions in grey matter (Narr et al, 2005), white matter (Kuperberg, et al, 2003) and lower cortical thickness (McDonald et al, 2005) as well as a lower than average birth weight in infants that later go on to develop schizophrenia. In comparison, infants later diagnosed with autism consistently have an increased brain size and are often bodily larger than then their peers when born and during development as well (Mills et al, 2007). In line with this and directly relevant to the social defects seen within the two disorders are the alterations to the amygdala and hippocampus within the two disorders. Baron-Cohen et al (2005) discuss the differences which follow in initial trend of being larger in autistic individuals and smaller in schizophrenic individuals.
The cognitive differences between the two disorders, as highlighted in the table, emphasise how central the social mind is and are believed to stem, at least in part, from the alterations to the amygdala previously discussed.
Conclude with potential problems but explain how it seems to be going in the right direction
Explain how these opposite expressions can come from the same gene
There have also been a fairly large number of genetic findings that have linked the two disorders via their associations with the same genes. Not only do the majority of the genetic studies provide evidence linking the two disorders but many also directly support the diametric model of schizophrenia and autism. Crespi, Stead & Elliot (2010) have reviewed the current literature relating to the genetic underpinnings of the two disorders in order to establish whether the genetic evidence fits with a diametric theory. A large part of the data analysed by Crespi et al supported the hypothesis of autism and schizophrenia being diametric conditions. In order to support this model it needed to be demonstrated that there was a statistically significant difference between the number of deletions on the gene in one disorder compared to the number of duplications on the same gene for the other disorder. For the model of diametrically opposed disorders to be supported via genetic studies it needs to be shown that for any abnormality seen on a gene implicated in both disorders, an opposite abnormality is seen on the gene for the other disorder, i.e. if there is a deletion on a certain gene in autism then in order for the diametric model to be supported, the same gene in schizophrenic individuals must show a duplication. In this way it is possible for us to see how the same characteristic, such as selective attention for example, can be dysfunctional in both disorders but in opposite ways, i.e. in one disorder it would be more selective than controls while in the other it would be less selective.
This phenomenon is well illustrated by a variant on the genes implicated in Williams syndrome.
Of the number of rare copy-number variants (CNVs) loci implicated in schizophrenia and autism, Crespi et al deemed seven to have enough data gathered to enable an analysis as to whether there was a statistically significant difference between the deletions in one disorder and the duplications in the other. The CNV loci analysed were 1q21.1, 15q13.3, 16p11.2, 16p13.1, 17p12, 22q11.21 and 22q13.3 and of these the following four were found to support a diametric model; 16p11.2 and 22q11.21 showed deletions to be associated with an increased risk of autism and duplications to be associated with an increased risk of schizophrenia while 1q21.1 and 22q11.21 demonstrated the opposite relationship. These reciprocal variants
Creativity in autism and schizophreniaFitting the disorders with the model
Dietrich's model can be easily shown to incorporate the thinking patterns of those with psychological disorders and, in doing so, we can perhaps further elucidate the neural networks behind both the disorders and creative thinking. Initial processing of emotional content occurs in the structures of the limbic system such as the amygdala, the product of which is then processed by the cingulated cortex and the ventromedial prefrontal cortex. Cognitive, mechanistic processing on the other hand occurs in other parts of the limbic system, namely the hippocampal formation and also requires the temporal, parietal and occipital cortices. Full reintegration of the cognitive and emotional information doesn't occur until the dorsolateral prefrontal cortex. I argue here that not only that the systems described by Dietrich as underlying spontaneous, emotional processing best fit with the known data on the systems implicated in autism but also that the deficits seen in damage to this system mimic those seen in autism. As such the style and type of creativity seen in the opposing camp of deliberate, cognitive thought fits with the styles generally associated with autistics. Conversely I believe that it is also the case that the systems underlying such cognitive, deliberate thought are potentially damaged in schizophrenia and provide evidence of the known brain systems implicated in schizophrenia to support this claim. The opposite style of spontaneous, emotional thinking fits much better with the creativity often seen in schizophrenic individuals.
Deliberate mode - cognitive structures as Dietrich calls them are said to be responsible for the type of creativity shown by Thomas Edison's systematic approach to inventing or by the methodical piecing together of the of structure of DNA by Crick and Watson. These two examples were, I assume, picked fairly arbitrarily by Dietrich and yet a little research discovers that Thomas Edison was thought to be autistic himself. Not evidence for this theory but certainly interesting to consider. Dietrich goes on to suggest that a prerequisite for this kind of creativity is a large number of domain-specific items stored in the parietal, occipital and temporal cortices since this allows the maximum amount of relevant items to be considered within working memory. This expertise in one specific area is extremely symptomatic of an autistic individual who focuses on one or two things nearly obsessively and generally know all they can about their chosen interest. This prerequisite for knowledge, explains Dietrich, is why few scientific discoveries or inventions were made by individuals uneducated in the field.
Spontaneous mode-emotional structures are responsible for the kind of creativity that occurs when emotional information is suddenly represented in working memory and there is evidence to suggest not only that autistic individuals have problems representing emotions but also that they also have problems with spontaneity. These deficits would make it unlikely, although still possible, for an autistic person to be creative in the deliberate-emotional domain or the spontaneous-cognitive domain and doubly unlikely for them to be creative in the spontaneous-emotional domain. People on the autistic spectrum are characterised by their social difficulties which stem from an inability to read others emotions and gauge appropriate social responses. A lot of research has been conducted implicating dysfunctions of the amygdala in these autistic deficits and, importantly showing that the dysfunction is due to an inactivation rather than over activation of the amygdala. In line with this, research has shown a disruption of white matter tracts between other limbic areas responsible for emotion and social processing namely the ventromedial prefrontal cortex and the anterior cingulate. Kennedy and colleagues (2006) found a direct correlation between severity of social impairment in autistic spectrum individuals and atypical activation of the vmPFC. These parts of the limbic system are the same as those implicated by Dietrich as being responsible for the emotional types of creativity thus it should come as no surprise to see a limited number of autistic individuals performing this type of emotional, artistic creativity. Of course that's not to say it doesn't happen because of course there are emotionally creative autistic individuals but this model can help to explain why they tend to be the exception rather than the rule. To investigate the spontaneity of autistic individuals we need to move away from Dietrich's model briefly to look at the Default Mode Network (DMN). The DMN is a network of brain regions that are active when an individual is not focused on anything in the outside world and the brain is at wakeful rest. The DMN activates when individuals focus on internal tasks such as daydreaming, and deactivates during goal-oriented activities. Ingvar (1979) was the first to propose the notion that spontaneous thoughts occur in relation to specific 'rest state' activation in the brain. Specifically he suggested that a discovered hyperfrontal activity pattern reflected "spontaneous conscious mentation". Further research has gone on to show that spontaneous thoughts are much more likely to be generated in this brain state, as demonstrated by Binder et al's (1999) experiment which demonstrated that rest, as compared to individuals participating in a task, was associated with both increased DMN activity and nearly six times more spontaneous thoughts. This network has been shown to be disrupted in both autism and schizophrenia. Buckner et al (2008) argue that the DMN is less active in individuals with autism and as such their rest state or imagination state differs to that the typical population. This data certainly agrees with the observed lack of imagination seen in those diagnosed with autism. The implication of these findings on the integration of autism with Dietrich's model is to illustrate the decreased likelihood of spontaneous creativity occurring in individuals on the autistic spectrum. In the same way that dysfunction in the emotional regions of the brain doesn't completely rule out emotional creativity in autistics, the decreased activation of the DMN doesn't completely rule out spontaneous creativity. I am not arguing that autistic people cannot have spontaneous creative thoughts but simply emphasising the fact that the general trend would be towards deliberate creativity in the ASD population. Of course it is fairly clear to see that the likelihood of an autistic person producing creativity in the spontaneous-emotional domain is severely reduced and if we think of the kind of creativity that would typically go in such a domain, emotive poetry perhaps or abstract, representational artwork, we can see that it is typically the kind of creativity associated with the schizophrenic end of the spectrum rather than the autistic end.
The examples given by Dietrich for the kinds of creativity seen in the emotional-spontaneous domain are the artistic expressions of Picasso and Samuel Coleridge's poem "Kublai Khan." Again, presumably, the examples we are given are arbitrary and yet Picasso has been linked with schizophrenia numerous times while Coleridge has been suggested to have had hypergraphia, a condition often linked with schizophrenia. Just to be clear on this point, I'm in no way viewing this as evidence for this type of creativity being associated with schizophrenia but it is just another finding that makes you think instinctively that perhaps this is the right direction. Dietrich describes the process by which spontaneous-emotional creativity occurs as happening when the neural activity of structures that process emotional information becomes spontaneously represented in working memory. It is well known within the psychological community that working memory capacity is very limited and as such these spontaneous representations must compete in order to reach consciousness (Pinker, 1999). Since emotions are directly relevant to us and tend to be attached to biologically significant events, they are more likely to present with a strong or 'loud' signal in order to enter consciousness. Emotions are what permit people to attach values to certain stimuli and it is widely accepted that schizophrenic individuals often attach salience to stimuli randomly and inappropriately which contributes to the delusions of the disorder. Related to this is the fact that it has been shown by Myin-Germeys et al (2000) that schizophrenia subjects experience more intense and more variable negative emotions than controls. Taken together these results show that someone on the schizophrenic spectrum would be likely to have strong emotional signals attached to a variety of stimuli and as such it is a lot more likely that such thoughts would be spontaneously represented in consciousness, leading to a larger amount of spontaneous-emotional creativity from people on this psychotic spectrum. Indeed considered from the opposite point of view it would seem very obvious that these kinds of spontaneous emotional thoughts occur frequently in schizophrenic individuals based on the disorders inherent characteristics. The emotional nature of an insight signifies its importance and the unintentional nature means it is much more likely to be described as some kind of epiphany or divine inspiration. It isn't difficult to see how strong salience, randomly attached to some kind of stimulus, combined the spontaneous nature would lead a schizophrenic person to believe that perhaps a god was speaking to them or some kind of malevolent demon putting thoughts into their minds and as such it may be the regular occurrence of these kinds of thoughts that promote the hallucinations and delusions so common in schizophrenia.
There is a large amount of preliminary data suggesting that the DMN is overactive in the schizophrenic population and this has many implications for how we can explain the position of the disorder within Dietrich's model. First and foremost it provides a potential explanation for the spontaneous emotional thoughts of schizophrenia. Harrison et al. (2007) recorded increased activation of the DMN in passive states in schizophrenic patients as compared to controls. This result was replicated in a study by Garrity et al (2007) and furthered when they found that the positive symptoms of schizophrenia (e.g., hallucinations, delusions, and thought confusions) were correlated with increased default network activity during the passive rest moments, specifically implicating the medial PFC and the posterior cingulate cortex (PCC). This overactivation of the DMN during rest or passive states presents a potential explanation for the spontaneous emotional thoughts required to place schizophrenia into the corresponding quadrant. Since we have already discussed the role of the DMN in the creation of spontaneous thoughts, the overactivation of this network in schizophrenia could explain the prevalence of such thoughts within the disorder. This idea is supported by the link made by Garrity discussed earlier between the positive symptoms and the DMN activity. Delusions and hallucinations are emotionally loaded stimuli that present spontaneously to consciousness in the same way that creativity inducing spontaneous emotional thoughts would. The fact that overactivity of the DMN correlates with an increase in these positive symptoms certainly supports the idea of an overactive DMN producing an excess of spontaneous emotional thoughts and as such placing schizophrenia within Dietrich's spontaneous mode-emotional quadrant. There has been a large amount of evidence that the amygdala in schizophrenic individuals is dysfunctional; often overactive. The amygdala, cingulate and ventromedial prefrontal cortex are all involved in the processing of emotional information in the brain. The amygdala is responsible for the initial processing of affective content (Le-Doux, 1996) while the cingulate cortex and the ventromedial prefrontal cortex have been implicated in the higher level of emotional processing (Damasio, 1994). Given the involvement of the cingulate and VMPFC in the DMN and their strong links to amygdala, it seems that the overactivation of the DMN can also be used to explain the emotional content of the spontaneous thoughts.
Deliberate mode-cognitive structures are responsible for the kind of logical step by step processes that often lead to creativity within the scientific or mathematical domains. There are a number of reasons why a schizophrenic individual may not fall naturally into these categories. Firstly, many studies have shown that the dorsolateral prefrontal cortex in schizophrenia is dysfunctional and there have been many further investigations looking at how this affects the disorder. The DLPFC has been implicated in working memory, directed attention and temporal integration (Fuster, 2000a; Goldman-Rakic, 1992; Knight & Grabowecky, 1999; Posner, 1994) and a slightly more recent study by Perlstein et al (2001) has demonstrated not only that the dysfunction of the DLPFC in schizophrenics cause working memory problems but also that it contributes to the disorganised thinking seen in the disorder as well. These problems are what could prevent the majority of the schizophrenic population from falling into Dietrich's deliberate-cognitive quadrant of creativity.Conclusion