Questioning prescriptive power: A re-thinking of theory of mind

(Over the next couple of months, I will be hosting a series of guest posts on the social study of autism. In this second post of the series, it’s great to have Matt Willis. Matt has recently completed an honour’s degree in social anthropology at La Trobe University in Australia. His research, which was based on a careful reading of prior research and literature, focused on autism and the anthropological perspective.)

I’m going to be writing about the unbalanced construction of autism, drawing primarily on theory of mind as an example. My aim is to get people thinking about how cognitive theories, such as theory of mind, are constructed and applied by neurotypical people to autistic people, or those labelled as autistic. How does this affect the validity of the theories? Who has the power in the decisions to use and how to apply these theories?

Before I dive into that though, I want to establish a precedent to my point: intelligence tests. Anthropologists and psychologists, among others, are generally critical of the practise of applying intelligence tests developed in one culture to people in another. Psychologist Patricia Greenfield considered ability tests “cultural genres” in that their construction is determined by particular cultural symbols and normalised practices. Taking one cultural genre and trying to superimpose it on a different culture is like trying to fit a square block into a circular hole—it’s not going to work! Robert Sternberg outlines a bunch of great reasons as to why this is the case, including different cultural evaluations of what ‘smart’ means, the familiarity of assessment materials and delivery, children might develop skills that serve them better in their environments than other environments, and people from different cultures might simply think about things quite differently (see reference list at the end of the post if you’re interested in reading more).

Now, these tests don’t necessarily have much traction within the cultural genres in which they are constructed either (if you even accept the notion of “cultural genres”), but that’s another can of worms. The question is: what does any of this have to do with autism (aside from the fact that autistic people likely frequently have to take such tests and are then erroneously labelled as this or that)? Consider once more the idea of constructing an aptitude test based on a particular cultural genre and attempting to use this test to explain the abilities of a group of people from a different cultural genre. Who is constructing this test? Why have they done so? Why are they applying it in the context of a different cultural genre? In other words, where is the power and how is it being used?

Theory of mind (the theory that typical minds have an ability to perceive states of mind in other people, such as intentions, knowledge, and desires) is frequently said to be impaired in autistic people. But why? Consider this: neurotypical children commonly fail theory of mind tests. In most of these instances, attempts are made to find alternative explanations and discover other instances in which the kind of empathy associated with theory of mind is present. Yet people who are labelled as, or in the process of being labelled as, autistic generally don’t receive the same scrutiny. It’s more a case of ‘Alright, there you, there’s the evidence!’

Damien Milton, an autistic person himself, as well as a doctoral social researcher at Birmingham University, has written a great critique of theory of mind, entitled On the Ontological Status of Autism: the ‘Double Empathy Problem’. In this article, Milton points out that many autistic people are made to feel uncomfortable or threatened during theory of mind tests because of the way the test is administered. This highlights, fairly strongly, the fact that tests like theory of mind are designed and implemented by neurotypical people, based on what they believe is appropriate methodology. Now, I’m certainly not trying to suggest that autism is equivalent to a culture, although different movements surrounding autism may constitute cultures. Even so, I think theory of mind is potentially another kind of invasive aptitude test, constructed by one group of actors and imposed upon a separate group of actors.

This is not to say that a degree of absence in theory of mind might not be observed in an autistic person. I have an autistic friend who has readily admitted that it is somewhat the case for him. But that “somewhat” is important. Ben has already written about the diversity of autism. We all know that autism is not homogenous, that autistic people are as different from one another as are neurotypical people from each other. You cannot apply one, totalising theory to a diverse group of people. Nor can you ignore the fact that absences of theory of mind may very well manifest to some degree in many neurotypical people.

In Milton’s article, his main point is that the neurotypical researchers who investigate the presence of theory of mind have a distinct position of power in deciding whether their fellow communicator is perceiving states of mind, or empathy. Indeed, neurotypical people in general hold this power. The kind of empathy under question is determined by neurotypical people, based on what they believe is normal behaviour. Under this model, a neurotypical person’s empathy is difficult for an autistic individual to grasp. Yet, as Milton argues, it is equally true that an autistic person’s empathy is difficult for a neurotypical person to grasp. Communication works both ways, and the rules of communication should not be determined by only one actor’s ideas of what is normal and what is not.

So what can we say about theory of mind after all this? That it is constructed by neurotypical people, assumes that neurotypical cognizance (if there is such a thing at all) is free from an absence of theory of mind, assumes that autistic people universally display a degree of an absence of theory of mind, yet does not seem to engage in communication with autistic people on level ground. Therefore, the foundation upon which the notion that autistic people have a deficit in theory of mind is built, is not as solid as many people assume.

What this all means for me is this: people need to stop generalising cognitive normalcy and start considering that many people communicate, think, and feel in different ‘non-typical’ ways. These ways are not invalid and are not representative of entire categorisations of people. Put another way: neurodiversity deserves all the momentum it can get.

Publications cited

Milton, Damien 2012, ‘On the ontological status of autism: the “double empathy problem”’, Disability & Society, vol. 27, no. 6, pp. 883-887.

Greendfield, Patricia 1997 ‘You can’t take it with you: why ability assessments don’t cross cultures’, American Psychologist, vol. 52, no. 10, pp. 1115-1124, <;

Sternberg, Robert J. 2008, ‘Culture, instruction, and assessment’, in J Elliot and E Grigorenko (eds.), Western Psychological and Educational Theory in Diverse Contexts, Routledge, Oxon, pp. 5-22.

Related Posts:


8 thoughts on “Questioning prescriptive power: A re-thinking of theory of mind

  1. The author’s use of language is problematical to put it mildly..
    For example he says, “I’m certainly not trying to suggest that autism is equivalent to a culture”. No disagreement there but he seems to be suggesting that this means that there is no such thing as autistic culture.
    In fact there is an emergent autistic culture.

    He points out that theory of mind “is constructed by neurotypical people” whilst seemingly missing the obvious point that autism was constructed by neurotypical people.
    Autism is a construct and autistic people played no part in its construction.
    By failing to take this into account the tone of the piece becomes unintentionally patronising.
    He ends by saying that “neurodiversity deserves all the momentum it can get”.
    This is simply nonsensical.
    Neurodiversity is a scientific fact.
    I have to assume that he means that the neurodiversity movement (or possibly the neurodiversity paradigm) deserves all the momentum it can get.
    Yes indeed.
    But let’s use the correct language please.

    There is another howling error….
    He also says that “autism is not homogenous”.
    If one assumes that he means that “autistic people are not a homogenous group” then it makes sense.
    But he really loses the plot with “autistic people are as different from one another as are neurotypical people from each other”.
    Autistic people are MORE different from one another as are neurotypical people from each other.
    Neurotypicals are the typical group!
    We, by our very atypicality are unlike each other.

    1. Autistic UK, thanks you for your response. I understand your points and frustration. I certainly meant no offense by my post. I had adopted a casual writing style and, considering how much Ben has already written about the details implicated in my post, I did not want to be too repetitive and get bogged down in clarifying points I had assumed most readers would already understand. This was foolish of me and I certainly must concede that I could have been clearer with some of my language. I apologise for that. Nevertheless, if I may briefly respond to your points, perhaps some of the tension can be resolved.

      The presence and content of this very blog already spells out the fact that autistic cultures are very real. Furthermore, in my post I do explicitly state that “different movements surrounding autism may constitute cultures.” This, to me, was sufficient acknowledgement of the fact that autistic cultures do exist.

      I did mean “neurodiversity deserves all the momentum it can get.” The post is about how professionals have automatic power in constructing explanatory theories of, in this case, autistic cognizance. The scientific fact of neurodiversity is what such professionals, and the general public alike, seem to be unaware of or uninterested in. Therefore, “neurodiversity”, as a scientific fact, as a concept, needs more momentum behind it. This is, clearly, the role of the neurodiversity movement. I am advocating its approach by acknowledging what the neurodiversity movement does. At least that was my intention.

      I don’t believe stating that “autism is not homogenous” carries no meaning. Autism, which is a category constituting roughly similar ways of being, is not a homogenous category because of the fact that its “ways” of being are plural. As stated in in the opening of my response, I was writing with a couple of erroneous assumptions and this is certainly a statement that could have benefited from some clarification. I sincerely apologise for that.

      As for your final point, I certainly understand what you’re saying. However, I also believe it is easy to oversimplify “neurotypical” as well. There is far more difference between neurotypical people than many seem to recognise. Indeed, professionals who categorise and theorise about autism are guilty of this oversimplification of neurotypicality as well. Neurotypical people are also neurologically and behaviourally diverse within themselves. Immensely so. It’s a more complex point than it first seems.

      It’s important to distinguish between talking about autistic people as representing a diversity of neurological type beyond the normalised kind, and talking about autistic people as being more diverse than other types within themselves. The spectrum of neurological diversity does not manifest only in non-typical neurologies; it is present in neurotypical people as well. The latter may be true, but I’m not comfortable categorically stating this in lieu of systematic evidence to that effect, the gathering of which I do not advocate. Attempting to estabalish which group of people is more neurologically diverse within themselves is not only superfluous, but I believe counter-productive to the wider message of neurodiversity advocates. Human beings (not one group or another) are a neurologically diverse group of organisms. It is not a competition. This is why I chose the words “as different”. In the words of Thomas Armstrong (, “we are all somewhere along continuums related to literacy, sociability, attention, learning, and other cognitive abilities, and thus all of us are connected to each other, rather than being separated into ‘normal’ and ‘those having disabilities.'” It is for this reason that I also refrained from using the word “neurotypicals”. This carries the same implications of oversimplification and overcategorisation as using the contested word “autistics”, and promotes the notion that one neurological type should be privileged over another. Again, this runs counter to the core message of the neurodiversity movement.

      As I have said, I certainly concede that my use of language was perhaps a little careless. This is only because I was in a more relaxed, casual writing mode, as this is the mode I thought most suitable for Ben’s blog. I maintain that, despite the casual style, my points are still academically sound and, despite some not-quite-explicit language, my choice of words are not so disastrous as to misconstrue, misrepresent, or obscure the message of the neurodiversity movement.

      I honestly thank you for your critique, Autistic UK. A writer who believes there are no aspects of their writing they can improve upon is a fool. I hope my response generates some kind of resolution to your grievances.

    2. Thanks for pointing out the elephant in the room (not just in this article): WHO constructed and WHO maintains the constructs i/r/t neurodivergence? Are neurotypical and neurodivergent, as constructs, embeddedness in social power relations? Whose interests are served by constructing human differences through the rhetorical lens of neuroscience, as if that lens is constructed from practices and beliefs and language free of reification—perhaps appropriated from a planet where all minds have solved the problem of subjectivity ?

      And is communication such an easy thing to define in terms of individual minds, such that we can know whether we are inquiring about communication—or social domination?

      These are the directions of analysis that may reveal the ultimate means-end orientation of inquiry, the same strategic notions of communication “success” measured by “effect” … as if cause and effect in communication make any sense at all—well, at least when human actors are still human, rather than programmed automatons.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s