Return to SCE Homepage
Return to Session Program

Autism and Representation
Case Western Reserve Univ
Cleveland, Ohio
28-30 October 2005



Bruce Mills
Kalamazoo College
Do not cite without permission of author.

"Autism and the Imagination"

We associate much with the imagination: the power to evoke images of absent objects and people, to generate inner pictures of past events or memories, and to create artistic representations of important experiences and desires.   We attribute the highest order of moral, intellectual, and emotional growth to the ability to engage the faculty of the imagination; we see "imaginative play" as necessary for the development of empathy and thus healthy familial and social bonds.   In part, such associations and attributions arise from a history of understanding the imagination as more than the passive ability to recollect "images" and instead as an active and transformative faculty of the mind.   Emerson, Coleridge, and others argued that, through the productive or constructive power of the imagination, we literally create the world and/or give order or larger meaning to seemingly disparate and discrete elements.   In light of this understanding, an impaired imagination might manifest itself in a lack of empathy, a literal mindedness, or a tendency to get caught up in the individual and discrete impression.   Within a continuum of definitions that elevate the liberating (and humanizing) nature of the imagination, such qualities suggest profound limitations--and a vision of the world that closes off the possibility of great art and wisdom.

Given current definitions and descriptions of autism spectrum disorders (ASD), it is reasonable to conclude that those with the disorder lack the possibility of experiencing the full, imaginative life of "neurotypical" individuals.   In the Diagnostic and Statistical Manual of Mental Disorders, 4 th Edition or DSM-IV , clinicians address this deficit in the last of the triad of impairments.   Noting the "restricted repetitive and stereotyped patterns of behavior, interests, and activities" of those on the spectrum, the DSM-IV illuminates the nature of this impairment by underscoring how it presents itself through "delays or abnormal functioning" in the area of "symbolic or imaginative play."   While grounding such diagnostic criteria in empirical studies, this representation of autism still reveals underlying assumptions concerning the nature of the imagination and an "ideal" or "natural" creative process; it positions certain cognitive processes in relation to the imagination as "normal."   In its implied emphasis upon the synthesizing power of the imagination, for instance, the DSM-IV invites readers to devalue the heightened (and atypical) ability of those with autism to discern the sometimes subtle and distinct features of a discrete image, activity, or experience.   Instead, it favors the capacity to move quickly to a more generalized or symbolic understanding.   Moreover, if autism is theorized as an impaired ability (or inability) to develop what Uta Frith terms the "drive for central coherence," then those on the spectrum seem to lack definitive features of the imagination.   That is, Frith asserts that "[i]n the normal cognitive system there is a built-in propensity to form coherence over as wide a range of stimuli as possible, and to generalize over as wide a range of contexts as possible" ( Autism 2 nd Edition 159-60); such a view resonates with the belief that the highest manifestations of the imaginative faculty embody the capacity to unify seemingly distinct images and memories.

In this paper, I wish to trace resonating conceptual legacies.   In recent Western thought, poet-philosophers consistently linked notions regarding the imagination to the anatomical nature of the mind.   In the even more immediate study of the biology of mind, psychologists and scientists hypothesize a peculiar mentalizing ability (or theory of mind) within the particular neurological wiring of the brain.   With the increased incidence of autism and the insights arising from autists' self-reporting and artistic work, however, we might begin to re-think past paradigms that oppose typical/normal with atypical/abnormal creative processes.   In the continuum that marks the different cognitive processes that produce "art," we might begin to refine an understanding of the imagination in relation to autism.  

A Unifying Instinct: Romantic Notions of the Imagination

Considering the law of the human mind and equating the process of creation with the act of thought itself, Ralph Waldo Emerson voices an influential notion of the creative self in a language that recalls current models of intellectual development:

To the young mind every thing is individual, stands by itself.   By and by, it finds how to join two things and sees in them one nature; then three, then three thousand; and so, tyrannized over by its own unifying instinct, it goes on tying things together, diminishing anomalies, discovering roots running under ground whereby contrary and remote things cohere and flower out from one stem.   It presently learns that since the dawn of history there has been a constant accumulation and classifying of facts.   But what is classification but the perceiving that these objects are not chaotic, and are not foreign, but have a law which is also a law of the human mind? ("The American Scholar" 65-66).

Delivered to students at Harvard College in 1837, Emerson's assertions offer his era an image of the evolving intellect and thus an antebellum rendering of what William G. Perry, Jr. and others might term a form of "intellectual and ethical development in the college years."   Interestingly, he envisions such growth as the unfolding of an internal and eternal law; it is a transcendental vision rooted in biological fact.   The "duty" of the individual, then, is to attend to and foster the insights arising from this "unifying instinct" or "law of the human mind."   For Emerson, this creative life involved more than rigidly applying the thoughts of other times and writers; it demanded that the past be used to construct a response to the present, to develop new generalizations that harmonized the seemingly unrelated aspects of current and evolving conditions.   One could argue, given his reflections in "The American Scholar," that the ideal mind by nature promised flexibility and fluidity of thought.

Emerson's ideas represent just one example of the tendency to link speculation on the nature of thought and cognitive processes to the physiology of the mind itself.   Starting in the late eighteenth century, the discussion of the imagination often formed the site of this speculation.   By the second and third decades of the nineteenth century, prominent encyclopedia definitions of the imagination offer some of the clearest merging of the literary, metaphysical, and physiological.   According to The Cyclopaedia (1824), the imagination "is a power or faculty of the soul, whereby it conceives and forms ideas of things, by means of impressions made on the fibres of the brain, by sensation."   In fact, the definition of this faculty or power arises more from new findings in relation to the anatomy of the brain and the neurology of perception than past philosophical or literary models.   Thus, after distinguishing between the passive imagination (i.e., the "simple impression of objects") and active imagination (i.e., the capacity to arrange and combine received images in a "thousand ways"), the article devotes the rest of the initial paragraph to how the brain processes sensation of present and absent objects.   This is to say that modern notions of the imagination arise from the era's conceptions of cognitive processes, from the nature of the human mind.

Such notions resonate with Samuel Taylor Coleridge's influential distinctions between the primary imagination, secondary imagination, and fancy articulated Biographia Literaria .   In a sense, his distinctions define features of the imagination along a spectrum.   At the most advanced place on this spectrum, Coleridge posits the primary imagination or "the living Power and prime Agent of all human perception" and "repetition in the finite mind of the eternal act of creation in the infinite I am. "    The secondary imagination only differs from the primary in "degree" and "mode of its operation": "It dissolves, diffuses, dissipates, in order to re-create; or where this process is rendered impossible, yet still at all events it struggles to idealize and to unify."   Finally, Coleridge argues that the "Fancy, on the contrary, has no other counters to play with, but fixities and definites"; it is "no other than a mode of Memory emancipated from the order of time and space; and blended with, and modified by that empirical phenomenon of the will, which we express by the word choice "    ( Biographia Literaria 304-5).   Along this hierarchy, then, we move from the more limited capacity to attend to ("play with") objects provided by memory to the ability to create anew, i.e., to transform ("dissolve, diffuse, and dissipate) in order to given new order to the seemingly fixed elements of sensation and memory.   And, while not clearly articulated in these excerpts, Coleridge viewed those who display fancy to be lesser thinkers and artists.

In defining the "sensual man" as one who "conforms thoughts to things" and the "poet" as conforming "things to thoughts," Emerson captures aspects of Coleridge's distinctions.   "The Imagination," Emerson adds, "may be defined to be the use which the Reason [i.e., intuition or unifying instinct] makes of the material world" (44).   In a sense, Fancy can be understood as the conventional mind that strains to move beyond the simple mirroring of rigid forms and laws; the Primary Imagination is the power to construct novel approaches and narratives and to grasp larger symbols or underlying structures in the prelude to and/or process of creation.   In the Romantic Period, the power of the primary imagination offered one of the highest manifestations of a divine life.   Not surprisingly, the poet or artist were described in terms that captured this capacity to unify aspects of the world.   And, while not everyone had the ability to achieve such heights of the imagination, each person (created in God's image) contained the wiring for such a possibility.

By directly and indirectly connecting imagination to the mechanism of the mind itself (and how this mechanism or structure renders in material form the infinite dimensions of divine creation), Coleridge, Emerson, and others mark a definitive aspect of a modern turn in Western thought.   In tracing an "archaeology of the human sciences," Michel Foucault describes this epistemological transition or new episteme as one that positions man as an "empiro-transcendental doublet":

For the threshold of our modernity is situated not by the attempt to apply objective methods to the study of man, but rather by the constitution of an empirico-transcendental doublet which was called man ....   There are those that operate within the space of the body, and--by studying perception, sensorial mechanisms, neuro-motor diagrams, and the articulation common to things and to the organism--function as a sort of transcendental aesthetic; these led to the discovery that knowledge has anatomo-physiological conditions, that it is formed gradually within the structures of the body, that it may have a privileged place within it, but that its forms cannot be dissociated from its peculiar functioning; in short, that there is a nature of human knowledge that determines its forms and that can at the same time be made manifest to it in its own empirical contents. ( The Order of Things 319)

From the time of the Enlightenment, Foucault underscores, such human faculties as reason and imagination took into account the physical or material nature of the human body.   Though the ability to transcend the constraints of time and place through our imagination points toward a kind of metaphysical power, the ultimate check to such abilities, by definition, rests in the evolution and normal development of mind as much as in the formal education of the self.   In the anatomy of mind, in the very wiring of the brain, we might discover--or, certainly, people came to believe in the possibility of discovering--peculiar capacities for knowing.   In a sense, the imagination is a hardwired faculty that has its own rules.   Definitions of the imagination, then, can be traced back to the cognitive processes evolving from the brain.

Given this pattern of thought arising in the eighteenth century and gaining wider prominence in the nineteenth century, it is possible to argue that notions of the imagination were progressively connected to the laws of cognition.   In other words, the metaphysical was merging with the physical, the philosophical with the biological.   Artistic genius simply manifested the highest order of these inherent laws; moreover, the artistic productions of this creator betrayed the laws of this unifying impulse, this drive toward greater visions of coherence and wholeness.   Hence, with Coleridge and Emerson, we see the greatest praise directed toward those individuals who have done more than record the discrete or individual sensation or memory.   For the writers of the period, the imagination redeemed the seemingly unredeemable aspects of finite existence; it was the power that gave order and promised union where before had existed disharmony and disconnection.

Central Coherence, Theory of Mind, and the Imagination

Our current understanding of autism draws from theories that echo in striking ways eighteenth and nineteenth notions of imagination that have become "normalized" over two centuries.   And, while such views of what constitutes the imagination might not directly inform studies of autism, they form, at the very least, a conceptual and cultural framework that conditions how we understand the nature of creativity.   We hear past efforts to define the peculiar faculty of the imagination in two concepts concerning information processing that have influenced how we "see" autism: the notion of central coherence and mentalizing or theory of mind.

In her first edition of Autism: Explaining the Enigma published in 1989, Uta Frith explained differences between the drive for central coherence and local coherence in terms that cannot help but recall Coleridge's distinction between the primary imagination and fancy.   In a particularly vivid conceptualization of her ideas, she imagines information processing as a force not unlike the main channel of a large river.   For most individuals, this force "pulls together large amounts of information (many tributaries)."   However, she also posits a drive toward what she terms "local coherence," arguing that even smaller pieces of information must be "pulled some locally acting cohesive force" (97).   While Frith does not discount the importance of this drive toward local coherence, she does--at least in this first edition--see the more embracing pattern-making force as on a higher level.   Not unlike how Coleridge (and Emerson) framed the higher mental faculty of the imagination, she sees the drive for central coherence as the capacity to give order, to unify; more specifically, she reframes notions of the primary imagination (or Emerson's unifying instinct) in the terms of cognitive neuroscience.   In this context, the focus on small pieces and thus an impaired capacity to give a unifying explanation to "local" information mirrors the shortcomings of the fancy.

With the notion of mentalizing or theory of mind, we have another framework within which to explain the particular set of impairments described in the DSM-IV , a framework that clearly resonates with or is related to powers attributed to the imagination.   Frith prefers the term "mentalizing," defining this theory as "the ability to predict relationships between external states of affairs and internal states of mind" (2 nd Edition 77).   In Mindblindness: An Essay on Autism and Theory of Mind , Simon Baron-Cohen suggests that "the phrase 'theory of mind'. . . has come to be shorthand for the capacity to attribute mental states to oneself and to others and to interpret behavior in terms of mental states" (55).   Significantly, Baron-Cohen postulates four important "mindreading" mechanisms within the brain: Intentionality Detector (ID), or "a perceptual device that interprets motion stimuli in terms of the primitive volitional mental states of goal and desire" (32); Eye-Direction Detector (EDD), a mechanism that allows individuals to detect and make inferences from another's eyes (39); Shared-Attention Mechanism (SAM), or a mechanism that enables individuals to develop "triadic representations" or "relations among an Agent, the Self, and a (third) Object" (44); and, finally, Theory of Mind Mechanism (ToMM).   Serving an especially important role in relation to these other mechanisms, ToMM "has the dual function of representing a set of epistemic mental states and turning all of this mentalistic knowledge into useful theory" (50).   To see ToMM as performing tasks formerly assigned to the imagination would not be unreasonable.

Why is this drive toward central coherence and the presence of a theory of mind mechanism so important?   In answering this question in evolutionary terms, Frith and Simon Baron-Cohen begin to offer ideas that position autistic cognitive tendencies as marking an abnormal mind, one that typical development has marked as fundamental to survival.   (In a sense, the capacity to mentalize can be understood as "presence" and the inability as marking an "absence.")   According to Frith, "...without this type of high-level cohesion, pieces of information would just remain pieces, be they small pieces or large pieces" and "[a]s pieces they would be only of limited use in the organism's long-term programme of intelligent adaptation to the environment" ( Autism [1 st Edition] 98).   Baron-Cohen sees his work as a contribution to "evolutionary psychology," a field that "looks at the brain (and thus the mind) as an organ that, via natural selection, has evolved specific mechanisms to solve particular adaptive problems" (11).   Drawing from an essay by L. Cosmides, et al., he writes that

[the authors] use the metaphor of the brain as a Swiss Army knife to make this point.   Each blade of the Swiss Army knife was, clearly, designed for a specific purpose: the corkscrew for pulling corks, the screwdriver for driving screws, the scissors for cutting thin materials, the saw for sawing thicker materials, and so on.   It makes no sense to try to use the corkscrew to drive screws, or the screwdriver to pull corks, if the knife has a different "mechanism" for solving each of these problems.   So it is with the brain, say Cosmides et al.   We do not use our color-vision system to talk, or our language system to see color.   We use specialized modules for the functions they evolved to solve.   (11).

In light of such assumptions and beliefs, we can see ToMM as an especially critical mechanism; it is an evolutionary adaptation that enables individuals to go beyond superficial (e.g., mechanistic) interpretations of their environment in order to posit and process interior causes for exterior actions.   In many respects, we can again understand ToMM as related to and/or depending upon the functions associated with the imagination.   That is, it is the mechanism that takes specific information, creates representations of mental (not real) states, and forms propositions.   Moreover, it has a unifying function, i.e., it takes these representations of possible mental states and constructs a "theory" of human behavior upon which to base conclusions and potential choices.

Not surprisingly, psychologists and neuroscientists point to studies of children's play (or lack of it) as important evidence of an impaired imagination/theory of mind.   If, as Baron-Cohen argues, normal development shows the ability to "pretend or recognize the pretending of others" (53), then a failure to show such play implies some dysfunction, impairment, or "absence."   As stated in the DSM-IV , autism by definition manifests itself in play that fails to offer evidence of a normally-developing theory of mind.   Rather than "pretending," children simple perform ritualistic actions, e.g., the repetitive lining up of toys.   They appear to lack a mechanism that builds representations founded in an understanding other minds; they seem to have no drive to account for or give meaning to playmates' interior states.   In short, individuals with ASD are "blind" to other minds.   They do not interact with others because they cannot build a shared representation, a pretend scenario, without ToMM.   It is not that they do not want to interact; it is that they lack the mechanisms of mind for developing such an interest.

From Deficits to Styles of Knowing

If this admittedly brief history of the imagination and abridged summary of current cognitive theories of autism gesture toward one common denominator, it is the normalizing of styles of processing information that reflect less attention to the local and discrete.   Or, perhaps to frame this idea differently, these preceding reflections point toward the tendency to see the individual thing as a means to an end not as an end itself.   According to these influential views, to be human is to demonstrate a capacity for organizing broad pieces of information and giving special importance to inner or psychological states.   

But what if we were to define the imagination with less attention to the "highest" manifestation of some unifying faculty and more attention to distinct preferences for types of information and styles of pattern making?   What if the drive for central coherence and theory of mind (at least the capacity and powerful inclination to account for psychological states in social interactions) could be read not as signs of normal development but as features along a spectrum of possibilities?   In other words, we might consider that some individuals thrive on the local and more private (or what Frith calls "self-limiting taxonomies" [ Autism 2 nd Ed .163]) and others might orient themselves toward global arrangements and classifications.   The expressive outcomes of such predispositions would inevitably vary.   On this continuum of the creative and/or cognitive, the poles would represent especially limiting manifestations of inner "wiring" and outward nurturing, places where individuals cannot easily build bridges toward other ways of knowing and present peculiar challenges for the listener, viewer, and/or reader.

In revising her chapter "A Fragmented World" for her second edition of Autism: Explaining an Enigma , Uta Frith suggestively directs readers toward these possibilities.   Citing Francesca Happé's study of central coherence, she begins to reframe her earlier attempts to make sense of the fragmented perceptions of those with autism:         

Happé's work has been crucial in redirecting research interest away from the focus on deficits and toward the strengths in autism.   She collected evidence for detail-focused processing in autism in three quite different domains: visual perception, auditory perception, and verbal semantics.   She formulated the hypothesis that central coherence is a cognitive style that varies from weak to strong in the normal population and is reflected in a normal bell-shaped distribution of scores on relevant tests.   The population of individuals with autism would also be expected to show such a bell-shaped distribution, but with the mean shifted toward the weak extreme. (162)

Turning to work by Simon Baron-Cohen and his team in Cambridge, she notes his identification of "an information-processing style" that he terms "systemizing" (163).   Such a style, she summarizes, "is based on an intuitive understanding of how mechanical things work, and a preference for information about the physical as opposed to the psychological world."   According to the study, "[s]tyle and content not only fit individuals with autism but also many normal people..." (163).   As people begin to assess and understand the meaning of "spectrum" in Autism Spectrum Disorders even more fully and to include control groups and family members that point toward affinities rather than the presence and absence of particular abilities, the limitations of past theories will give way to a more constructive configuration of what constitutes strengths and weaknesses.

Perhaps one outcome of such thinking is the need to conceive of creative processes and outcomes in terms of teams and networks that not only foster the "arts" of local coherence but also facilitate a communication and presentation of such artistic products and processes.   It is important to emphasize that, while such a conclusion may not be profound, it demands a fundamental valuing of the small and local and the physical and mechanistic.   Thus, the neurotypical drive toward central coherence, manifested in various aspects of educational curricula and treatments, might give less room for an attention to the local, e.g., to an incessant focus toward multiple perspectives on a single object and how it looks or functions.   Preventing those with autism from getting "stuck" (an important and necessary goal) must be balanced with an eye toward the developmental advantages of nurturing the narrow vision.   Anecdotal evidence of the importance of finding this balance arises in the narratives (autobiographical and biographical) of or about those who have autism.   Perhaps two of the most widely read texts, Temple Grandin's Emergence and Thinking in Pictures , trace the positive outcome of identifying how certain "fixations" might open the door for meaningful insights, interactions, and careers.   Without question, Grandin's obsession with the "squeeze machine" (i.e., the device that enables her to apply calming pressure to her body and thus help address sensory needs) became the medium of an "emergence" from the more profoundly limiting aspects of her way of processing the world.

In many respects, then, this paper has simply endeavored to identify legacies of normalizing certain styles of knowing and to suggest their overt and subtle tendencies to foster a view toward autism that overlooks or tries to "correct" narrow or mechanistic tendencies.   This is not to say that most people do not reflect a drive toward central coherence or a theory of mind or that ASD does not present itself in impairments that have devastating effects on individuals and their families and communities.   In other words, these reflections are not meant to dismiss the very real challenges of enabling individuals on the spectrum to address perseverative behaviors and sensory processing in ways that release them from destructive fixations.   However, I would argue that a conceptual shift is necessary before the special skills and potential contributions of those with autism can be realized.   Moreover, in relation to the imagination, I wish to encourage a rethinking of assumptions behind the DSM-IV's perspective on imagination and autism.   The nature of play--and its symbolic and imaginative dimensions--might vary in relation to the particular manner in which the "player" processes the world.   The viewer also brings certain predilections to such observations.   To begin to see the nature of the imagination through insights gleaned from "autistic" creation--e.g., its rendering of the particular in ways that enables the single thing, so to speak, to be seen in all its peculiarity--might be one step.   Becoming aware of a potential mindbindness in the drive toward central coherence is another.


Baron-Cohen, Simon.   Mindblindness: An Essay on Autism and Theory of Mind .  

            Cambridge: MIT Press, 1997.

Coleridge, Samuel Taylor.   Biographia Literaria, or Biographical Sketches of My

            Literary Life and Opinions .   Vol. 7 of The Collected Works of Samuel

            Taylor Coleridge .   Ed. James Engell and W. Jackson Bate.   Princeton:

            Princeton UP, 1983.

Emerson, Ralph Waldo.   Selections from Ralph Waldo Emerson .   Ed. Stephen E.

            Whicher.   Boston: Houghton Mifflin, 1957.

Frith, Uta.   Autism: Explaining the Enigma .   Oxford: Blackwell, 1989.

---.   Autism: Explaining the Enigma. Second Edition .   Oxford: Blackwell, 2003.

Grandin, Temple.   Emergence: Labeled Autistic .   New York: Warner Books, 1986.

---.   Thinking in Pictures, and Other   Reports from My Life with Autism .   New York:

            Vintage, 1995.

Whiten, Andrew, ed.   Natural Theories of Mind: Evolution, Development and Simulation

            of Everyday Mindreading .   Oxford: Basil Blackwell, 1991.

Wolfberg, Pamela J.   Play and Imagination in Children with Autism .   New York:

            Teachers College Press, 1999.


Return to Top of Page

Return to SCE Homepage
Return to Session Program