Category Archives: Autism

Can the causes of autism also cause depression or anxiety?

I have been following the reaction to the Spectrum 10K project by monitoring the #Spectrum10K and #StopSpectrum10K hashtags (Spectrum 10K is a project aiming to gather the DNA of 10,000 autistic individuals). I have often seen the claim that the depression and anxiety which many autistic individuals experience cannot be caused by autism. Therefore, if we want to reduce the level of depression and anxiety in autistic people there is no point studying the causes of autism. In this blog post I will analyse this claim.

It is commonly claimed that many autistic individuals have such high levels of depression and anxiety because our society does not accept them or accommodate them. Some people can be hostile to other people who are different and this rejection can have a negative impact upon those people who are different. Also, society is set up to accommodate some people whereas other people whom it is not set up for face many barriers in daily living. These claims seem correct. The question is whether the cause of autism could cause depression or anxiety in addition to the depression and anxiety caused by societal response. I will now outline two possible reasons for thinking the causes of autism can cause depression or anxiety.

In relation to non-autistic individuals, depression and anxiety can result from many factors. Some people get depressed or anxious because how their life is going. To say they get depressed or anxious due to their genes would be a mistake. However, other people get depressed or anxious even though there life appears to be going well. There is no detectable environmental trigger of the depression or anxiety. In these cases it is plausible to say that causes internal to the individual are causing the depression and anxiety. So this is an example of depression and anxiety occurring for reasons other than how their lives are going. Could the same sometimes happen in autistic individuals? Now, it could be claimed that if this is happening in autistic individuals then this is due to the causes of depression or anxiety and not the causes of autism. However, do we actually know this? Is it possible that some causes of autism play a causal role in the characteristics associated with autism and can separately also play a causal role in producing depression and anxiety? I do not think this can be ruled out. The sheer complexity of genetics is immense, with any particular gene potentially acting very differently depending upon which other genes are present and what environmental interactions occur. We have only a very limited understanding of this. So it is possible that the causes of autism actually do also result in depression or anxiety which arise irrespective of an individuals life situation. To be clear, I am not saying this is the case, rather, I am saying we do not know enough to rule out this being the case.

Another reason to question the claim that autism cannot cause depression and anxiety is the possibility that what we call depression and anxiety in autistic people is different to depression and anxiety in non-autistic people. When an autistic person reports that they are depressed, and when an non-autistic person reports that they are depressed we take them as to both be experiencing the same thing, namely depression. However, it is possible that actually the phenomenological experience (how it actually feels) of autistic individuals when they report depression is different to the phenomenological experience of non-autistic people when they report depression. This phenomenological experience could arise because what autistic individuals and non-autistic individuals report as depression actually stems from different causes. This would give reason to think that actually what is being reported as depression is actually different in both cases. All this could also be the case for anxiety. If what we call depression or anxiety in autistic people arises from the causes of autism then the causes of autism actually can cause depression or anxiety in autistic people. Now, is there any evidence for this claim? The problem is that phenomenology is very difficult to do and it is very rarely actually done today, so this is not the sort of thing which we have much evidence upon (although I have often felt that my experience of depression seems different to phenomenology I have read relating to non-autistic individuals). Additionally, it raises a whole bunch of metaphysical and conceptual issues about what make two things instances of the same thing (I have published a couple of papers which partly discuss this sort of issue in psychiatry, but ultimately philosophers have discussed these issues for literally multiple millennia in areas other than psychiatry). Seeing depression and anxiety as the same, or not the same, in autistic and non-autistic people seems to be compatible with the evidence and with credible metaphysical and conceptual approaches. Again, I am not suggesting it is the case that depression and anxiety are different in autistic and non-autistic individuals, rather, I am showing that it could be a possibility.

In this blog post I am doing philosophy, aiming to show alternative possibilities rather than showing a particular interpretation is true. I aim to show that we cannot simply dismiss or discount the claim that autism can cause depression or anxiety. As such, I think this is not grounds, in itself, to object to biological research on autism. Of course, I have not addressed many of the other potential problems with biological research on autism in general and Spectrum 10K in particular which may give good reason to reject Spectrum 10K (I have discussed some scientific problems with Spectrum 10K in another blog post).

The potential and pitfalls of Spectrum 10K

There has been a lot of hostility to the Spectrum 10K project on twitter. In this blog post I will reflect upon this, alongside mention some additional aspects of Spectrum 10K which I have not seen commented upon.

Some autistic people are against all biological investigation of autism. For example, they might think it is eugenicist, a waste of time or a far lower priority compared to changing society. So it makes sense that people with these views would reject Spectrum 10K. However, given the level of hostility to Spectrum 10K, it seems that lots more autistic individuals are against all biological research on autism than I realised or a lot of people think there are specific problems with Spectrum 10K which are not present in at least some other biological research on autism. It is the latter group I have in mind for this discussion.

One reason why I was surprised by the hostility is that Spectrum 10K seems less problematic than most biological research (which is not to say it is problem free, I will discuss problems below). Most biological research on autism aims to correlate biological factors with autism itself, i.e. with people who meet the diagnostic criteria. So far as I can see, Spectrum 10K is aiming to do something different. It aims to correlate biological factors present in autistic people with environmental factors. In this regard, Spectrum 10K is studying an area which is completely ignored by much biological research on autism. It is actually including the environment in scientific research. If you think, as I do, that issues of flourishing and disability ultimately involve both biology and the environment (and much besides them) then this is a good thing.

It opens a range of possible practical benefits compared to most biological research on autism. For example, imagine it was discovered that some biological factors are correlated with some sensory issues caused by particular environments. This could lead to two options. Firstly, change the environment. Secondly, develop a medication which specifically targets that biological factor, alleviating the sensory issue. Contrast Spectrum 10K’s approach with typical biological research on autism. It generally just finds endless biological complexity. But imagine if typical biological research on autism found a relatively strong biological factor present in most autistic people. What practical benefits could follow from this? It does not give us information about what part of the environment to change. It also means that any medication developed would probably affect lots of autistic individuals in a major way. That starts to look like a so called ‘cure for autism’ which so many autistic individuals are against. So it seems to me that Spectrum 10K has more likelihood of producing practical benefits and less likelihood of being eugenicist compared to most biological research on autism.

There are certainly significant potential issues with Spectrum 10K. Firstly, scientists have lots of sophisticated tools for investigating biological data. Are they going to use equally sophisticated tools for investigating the effect of the environment? They need make sure that they study a wide range of environmental factors and categorise those environmental factors in a nuanced way. They also need to understand how these environments affect autistic individuals through using sophisticated measurements of wellbeing and through understanding the phenomenological experience of autistic individuals. Secondly, Spectrum 10K will have much more statistical power compared to most biological studies given the sample size. This could be a significant positive but it also could cause more harm than good. A larger sample size can often drown out individual variation. It might only find commonalities at a broad level and miss smaller groupings of commonalties which only exist on a smaller scale within the sample size. Thirdly, there is the problem of epigenetics. Spectrum 10K will look at the relationship between biology and the environment, such as establishing that people with a specific gene have problems in a specific environment. However, another relationship is possible, namely that the environment might turn on the gene. So Spectrum 10K might mistake one causal relationship, biology plus environment resulting in lower wellbeing, for another, biology being present due to the environment and is unconnected to lower wellbeing. I do not know how Spectrum 10K will tackle these problems but certainly many studies are deficient in these matters.

Finally, some people think that, in contrast the claims of Spectrum 10K, that this data may end up in the hands of other organisations which will misuse it. This is a potential problem inherent in all biological research on autism. If anyone has any detail on what track record Cambridge has in relation to breaking promises or on data breaches, I would certainly be interested to read it.

Overall, if someone rejects all biological research on autism then it makes sense to reject Spectrum 10K. However, for all its potential problems, it seems to me a step in the right direction compared to most biological research on autism. It might, like most biological research, produce nothing of use. However, as much as I think we need to radically redress the balance of funding between biological investigation of autism compared to non-biological investigation, I do think this type of biological research has more potential than most biological research on autism.

Report on subtyping autism workshop

On the 19th of November 2019 a workshop on subtyping autism was held at LSBU. The hosting organisation was PARC (Participatory Autism Research Collective). With help from individuals from PARC and LSBU, myself, Chloe Farahar and Annette Foster organised this workshop. The workshop lasted two and a half hours. There were five talks, each lasting twenty minutes followed by ten minutes of questions. Some papers argued against subtyping, some argued for subtyping and some looked at specific subtypes (or, in my own case, looked at multiple specific subtypes when I outlined the history of subtyping). I was pleased to see that the workshop was well attended, counting 37 people present at one point on a quick head count.

One aim of the workshop was to obtain the views of autistic people on the question of subtyping. I feel this was a success since some of the speakers identified as autistic whilst there were also autistic individuals in the audience. Some autistic people thought that questions about how to best formulate the diagnosis of autism should be largely determined by autistic people whereas other autistic people felt the views of autistic people should be one of many relevant factors. Either way, I hope this workshop went some way to facilitated autistic people, as well as the non-autistic, in providing views upon this issue.

Quite a number of speakers seemed to share a common vision of what a good psychiatric diagnosis should do, even though they disagreed about whether subtypes would help achieve those goals. In my presentation and in other presentations it was argued that there is more to autistic people than just a broadly defined autism. Rather, it is important to understand the specific ways in which a specific autistic person manifests autism (alongside understanding aspects of them unrelated to their diagnosis). Also, I and others argued that understanding how autistic people can change over time is important. It is also important to understand that different environments can influence how autistic people think, feel and behave. So in this reguard there was a level of shared commonality when it came to questions over what a good diagnostic system should do. Some people (including myself) felt that subtypes would help with understanding these factors whereas other felt this information could be best understood without subtyping autism.

I outlined a history of subtyping. I divided my history into three different areas, the 1930s to the 1970s, the 1980s until recently (around 2010) and finally I discussed the present day. I focused upon Lauretta Bender and Leo Kanner in the first period, then I focused upon DSM-III and Lorna Wing in the second period and finally I focused upon DSM-5 in the final period. I outlined how the subtypes which were employed heavily reflected the values held by psychiatrists in each era. I suggested that DSM-5 is still largely working within the values present in the 1980s even though our values are currently in a process of shifting away from those 1980s values. I then suggested that adopting subtypes would help better reflect the values we are currently moving towards.

The full list of talks was as follows:

Autistic observable and unobservable experiences, and erroneous subtyping: Introducing the Internal and External Autistic Space (Chloe Farahar  & Annette Foster )

Exploring an autistic derived classification of autism (Mary Doherty)

Online discourse on autism does not need autism subtyping (Dafne Zuleima Morgado Ramirez)

An historical overview of subtyping (Sam Fellowes)

Demand Avoidance Phenomena: The circularity, integrity and validity of PDA: a commentary on the PDA Conference held by the National Autistic Society (2018) (Richard Woods)

Subtyping autism workshop

On the 19th of November (16.30 to 19.00 at London South Bank University (LSBU)) I am hosting (with collaborators) a workshop on subtyping autism. Below are some reasons why I am interested in this topic and more details on the event (note that these are reasons why I, rather than my collaborators or the presenters, am interested in the topic of subtyping).

My main academic interest is showing that psychiatric diagnoses are in principle scientifically legitimate. I argue that some (though not all) psychiatric diagnoses can be used in a similar way to how models are used in more respectable sciences. They cannot do everything a typical model can do in other sciences. Also, what they can do is usually done with less precision than occurs in other sciences. However, what level of similarity that they do have means that they have significant scientific worth (and this is true even though they typically cover a heterogeneous symptom picture and are causally heterogeneous).

Of course, psychiatric diagnoses can vary significantly in their scientific worth. Various ways of formulating them can impact this. The main diagnosis I am interested in is autism and one alternative way of formulating it is to add subtypes. This could alter its scientific worth as well as having many practical impacts outside of the scientific domain. This workshop will hopefully generate interesting discussion about these issues. Also, many speakers and attendees have a diagnosis of autism (or are in the process of being assessed) so lived experience might contribute some insight to both scientific and practical consequences of subtyping. Below is the initial abstract that was part of the call for papers and then the list of speakers.

This was the initial call for papers: Prior to and after the publication of the 2013 DSM there was much controversy over whether Asperger’s syndrome should be removed from the diagnostic manual. Members of the neurodiversity movement made contributions to this debate, drawing upon their lived experience of being autistic and their knowledge of the ways in which being diagnosed can be (or is not) beneficial. This conference aims to consider the question of autism and subtypes more broadly. What benefits or disadvantages are there for autistic people in adding substantive subtypes to autism? It is commonly stated that autistic people can present in many different ways. Might subtypes help add accuracy and nuance to clinical pictures? On the other hand, the boundaries between subtypes are typically vague and they can be placed in multiple places. This raises the danger that they are arbitrary impositions which do not reflect autistic experience. This conference draws upon the experience of autistic people to help resolve these issues. Possible areas for discussion include whether self-understanding for those diagnosed would be increased through being diagnosed with a more specific subtype or if this would unhelpfully impose unnatural constraints? Would the non-autistic be better able to understand different ways autism can manifest if there are significant subtypes, or would it help perpetuate unhelpful stereotypes based on an imposed limited number of ways autism would be proposed to manifest? Would the neurodiversity movement benefit from the subtypes bringing greater focus upon different ways autism can manifest or would this lead to unhelpful fracturing of the autistic community?

These are the papers that are being presented:

Autistic observable and unobservable experiences, and erroneous subtyping: Introducing the Internal and External Autistic Space (Chloe Farahar  & Annette Foster )

Exploring an autistic derived classification of autism (Mary Doherty)

Online discourse on autism does not need autism subtyping (Dafne Zuleima Morgado Ramirez)

An historical overview of subtyping (Sam Fellowes)

Demand Avoidance Phenomena: The circularity, integrity and validity of PDA: a commentary on the PDA Conference held by the National Autistic Society (2018) (Richard Woods)

Reviewing Re-Thinking Autism

Re-Thinking Autism contains, to my knowledge, the most sustained attack upon the diagnosis of autism found within a single book. The editors claim it is the first book in the field of critical autism studies. It contains 17 different papers which are grouped into three areas, namely ‘What is autism’, ‘Deconstructing autism’ and ‘Challenging Practice’. The editors see critical autism studies as focusing on two questions. Firstly, is the diagnosis of autism valid and secondly, is it useful. Most articles are very critical of the diagnosis of autism. Since I felt I benefited immensely from being diagnosed with autism I was probably always going to have reservations. However, in the spirit of engaging with the critical autism studies movement, I will suggest that the book did not address some important questions and issues which would significantly help any attempted challenge to autism.

In relation to validity, multiple articles criticised the scientific foundations of autism. Some give an useful overview of the very heterogeneous causal underpinnings of autism. Whilst this is a very relevant point when assessing autism I feel the consequences were insufficiently explored. There were relatively few details about why the causal heterogeneity makes autism deeply flawed and in need of replacement. Some articles contain statements that autism is not a biological entity or a scientific entity but these are usually asserted rather than argued for. I think these claims face some challenges. Firstly, virtually all psychiatric diagnoses are causally heterogeneous. Therefore, I feel that critical autism studies scholars need either explain why autism need replacing whilst other diagnoses do not or need explicitly endorse a whole rejection of DSM type psychiatric diagnoses (perhaps in favour of person centred approaches). A few articles do seem closer to the latter approach though there are few explicit statements. Additionally, there is little discussion about why causal heterogeneity is problematic and why it makes something not a biological or scientific entity. There was also little mention of other important factors when assessing a scientific theory (such as simplicity, tractability or coherence). These are admittedly deeply philosophical questions which are debated by philosopher of psychiatry (and I was pleased to see a few articles reference philosophers of psychiatry) but I feel these are the sorts of questions which critical autism studies needs engage in. Perhaps the arguments which critical autism studies employ are fully defensible but they currently need more development.

In relation to usefulness, some articles question whether it is helpful to give people a label and other articles question whether it helps in educational settings or support settings. Some of these authors making these arguments have a wide range of experience and expertise. Consequently, I felt they had a viewpoint that was at minimum worth hearing. That said, there was very little input from people diagnosed with autism (only one author was described as being autistic). Very few of the reasons why I find being diagnosed so useful were mentioned. Of course, being autistic does not make me automatically correct on this issue. Perhaps my positive feelings about how useful being diagnosed was are based upon flawed reasoning. Despite this, it would have been good if the book engaged in reasons why some autistic people find being diagnosed so useful even if only to then challenge those reasons.

I feel the biggest problem was a lack of alternatives being outlined. If autism is deeply flawed then what should replace it? Even some vague suggestions (i.e. split it up, add subtypes, merge it with other diagnoses) would have been helpful if I am to assess whether an alternative to autism would be preferable. Additionally, outlining alternatives would give an easier route to challenge autism. There would be no need to argue autism is deeply flawed, rather, there would instead be the easier task of arguing that autism can be useful but an alternative diagnosis is even more useful.

It is worth nothing that the articles are quite diverse and so the degree to which these criticisms are applicable to any given article will vary significantly. Additionally, I felt around a quarter of the articles were pretty good (these were usually the ones which critically analysed autism rather than wished to replace the diagnosis).

To conclude, I felt many articles avoided important questions which are relevant for assessing scientific concepts. I would like to see the critical autism movement engage more with philosophy of science and philosophy of psychiatry. Also, it would be more helpful if criticisms of autism were also accompanied by actual concrete suggestions for alternatives to autism. Despite this, I am glad I bought it and read it. If you want a single book which contains multiple different criticisms of autism then this is the book to go for. I suspect I cite multiple articles from this volume (admittedly, mainly to criticise).

Ursula Le Guin (1929-2018), autism and neurodiversity


It was announced today that Ursula Le Guin died on the 22nd of January 2018. She was my favourite author in my late teens and early 20s whilst I put her in my top three today. I first read my favourite Le Guin novel, The Left Hand of Darkness, aged 18 and it probably had quite an influence on my thinking and views. To my knowledge, Le Guin never wrote about autism or neurodiversity but many of the themes in The Left Hand of Darkness seem very relevant.

The basic idea is a man from a western philosophy culture visits a world where the inhabitants are meant to be symbolic of eastern philosophy. The visitor sees things in contrastingly binary terms, such as light vs dark, society vs nature and male vs female. The world he visits has an eastern philosophy approach where different elements are in balance with one another, they do not form a sharp contrast. The strongest example is gender, since the inhabitants of this planet are genderless, so there is no binary contrast between male and female. However, the visitor, with his western philosophy, cannot help but put the inhabitants into binary terms of male and female, he struggles to shed his inbuilt preconceptions and see the inhabitants of the planet for how they truly are. The visitor constantly fails to read the motives of the inhabitants of the planet and often does not realise this is happening. However, during a political crisis which he also misreads he is eventually forced to cooperate with one of the inhabitants and, through a long and difficult process, eventually gain some insight into the inhabitants.

I read the novel as an exploration of the challenge of understanding others, be it people you have known all your life or people from cultures you are unfamilar with. I take Le Guin as thinking we often interpret people, especially people from other cultures, in unhelpful, simplistic, binary terms. She was influenced by her interest in anthropology and, writing in the late 1960s, the feminist movement which she made a major contribution to. These notions seems very important lessons for today given the existence of Donald Trump who characterises Mexican migrants as ‘rapists’ and African nations as ‘shitholes’ whilst ex-Daily Mail columnist Katie Hopkins describes migrants as ‘cockroaches’.

I think Le Guin’s general approach is also applicable to some of my interests in psychiatry. The problem of how to accurately understand someone who is different is quite central to philosophy of psychiatry. There has been a long history of negative connotations being associated with psychiatric patients and those in mental distress. As this articles shows, disability has often been associated with deviance and immorality. Whilst hopefully this situation is improved today, with more people talking about mental health, Le Guin’s insights are also important in a world which tries to be tolerant of those who are different. In relation to autism, modern science suggests autistic individuals often have unusual ways of thinking, unusual ways of perceiving the world and unusual ways of understanding others. Thus it is very easy for a non-autistic person to apply their non-autistic expectations to an autistic individual, think they have understood them but fail to. On the other hand, autistic people are not simply defined by the symptoms of autism, each one is a unique person with many unique traits and views. So there is an equal danger that a non-autistic person interprets an autistic person solely in terms them being autistic and thus again fail to understand who they are.

Additionally, all this is applicable from my perspective as an autistic person. Autistic people often struggle to pick up on social nuances or see the perspective of others, and these are certainly true of me. Additionally, my default position for many years was to assume people think and feel as I do but, in my experience, this just leads to typically failing to understand others without realising it. Of course, some autistic people go to considerable efforts to try see things from the non-autistic perspective but, in my experience at least, this is really difficult to do. Non-autistic people also have many diverse traits and views so this approach can easily lead to painting those around me with a big, rather inaccurate, brush.

Le Guin’s concerns over binary categories and importance of balance seems important here. She shows how, in relation to gender, we can find a helpful middle ground, a balance with more harmony if (primarily though not exclusively) men can make more effort to not hold unhelpful gender stereotypes and if both genders see that gender roles are a product of wider socio-cultural forces rather than how men and women truly are. Similarly, both autistic people and non-autistic people can try and find a helpful middle ground, that more harmonious balance. Even if most the problems are caused by the larger of two groups (the non-autistic) and even if many of those problems are caused by wider socio-cultural factors (such as misleading narratives from the media, expectations about what constitutes acceptable socialising or expectations that useful work must follow specific uniform procedures), I believe the best way to find that harmony is for both autistic people and non-autistic people to challenge the ways they perceive the other and where required make changes. This image of what neurodiversity should aim for is different to some notions of neurodiversity I have encountered which desire that only non-autistic people make changes. A mutual change towards more harmonious understanding feels to me more in line with the lessons of The Left Hand of Darkness, my favourite novel of the recently departed Ursula Le Guin.

Philosophical analysis of Neurotribes

My article, ‘Putting the Present in the History of Autism’ has been published in Studies in the History and Philosophy of the Biological and Biomedical Sciences (a pdf of the uncorrected proofs can can be found here). Though the article title mentions history, there is much philosophical content in article.

I focus upon Silberman’s extremely positive message about autism, discussing his portrayal of the past as being mistaken about the diagnostic criteria for autism and how the modern diagnostic criteria for autism has effectively got it right. I felt strangely conflicted about Silberman’s argument. One on hand, I thought he was unfair to many historical diagnostic criteria for autism. On the other hand, I still broadly agreed with his position. I agree with Silberman with that autism is getting something right about the world, being a worthy scientific concept and describing the world to at least a reasonably degree of accuracy. On this basis I significantly share Silberman’s positive message. However, as much as I believe in modern autism, I might believe even more in an alternative diagnostic approach to autism (whether it be one similar to one used in the past or something new), such as one with a greater number of subtypes or one with slightly altered boundaries. I felt Silberman’s positive message unfairly downplayed such alternatives. I think we need do research and consider our values to decide if the current diagnostic approach is superior to alternative diagnostic criteria and possible subtypes. We need study alternative approach and subtypes to see if we can make autism get even more right about the world or, alternatively, become more confident about modern autism by showing it works better than possible alternative views. So whilst I share Silberman’s positive views, I felt they risked reducing interest in scientifically investigating alternative approaches to autism, hence I both agreed with Silberman yet was critical of his position. I outline my views fully in the article.

Reviewing Peter Hobson’s The Cradle of Thought

Peter Hobson’s The Cradle of Thought is a philosophically and scientifically informed discussion of how thought develops in early life. Hobson’s main argument is that “interpersonal engagement contributes to the development of the mind – and [that] disordered interpersonal relations affect development of thinking” (p.143). He discusses many scientific studies which show how the level and nature of interpersonal relationships can impact the capacity for thought, suggesting deficient interpersonal relationships can lead to impoverished thinking.

Hobson highlights this through discussing individuals who are often not as capable of interpersonal interactions as most humans. He primarily discusses autism in detail, suggesting the usual thinking of autistic individuals (such as theory of mind differences) arise from lack of normal social and emotional interactions in early life. He heavily emphasizes that autism has a genetic component which results in biological differences (rejecting notions of poor mothering causing autism which some psychoanalysis used to believe) but argues the abnormal thinking itself is not primarily just due to biological difference. Rather, the biological differences result in difficultly interacting normally socially and emotionally, and this abnormal social and emotional experience results in abnormal thinking.

One might ask, why not just say the biological differences are responsible for both the abnormal interactions and for the abnormal thinking? Hobson uses a novel strategy to answer this question, primarily by looking at other types of individuals who also can face difficulties interacting normally. He considers individuals who are blind from birth and individuals who had very little social or emotional interaction in early life when raised in Romanian orphanges. These individuals often could not relate to other people in early life in the same way as most children. He then shows that these individuals sometimes develop some symptoms of autism, at substantially higher probabilities than would occur at random. Autistic individuals have biologically abnormal minds, blind individuals lack sight and Romanian orphans presumably have no major biological differences, yet all can exhibit some similar behaviour we associate with autism. Therefore, Hobson argues, some autistic behaviour is not directly the product of the biologically abnormal mind, but the biologically abnormal mind sets up abnormal interpersonal relationships and those interpersonal relationships result in symptoms associated with autism. Hobson also provides some evidence from how mothers with boarderline personality disorder interact with their children and how chimpanzees lack some parts of human socialising, suggesting both these cases can contribute to less than fully developed thinking.

Even though considered a developmental disorder, there is often a suggestion within scientific literature that the psychological development purely follow biological developments, rather than biological development resulting in psychological developments which then result in further psychological developments. I think a simple biological leading to psychological approach often present in modern science is far too simplistic, but I am unsure whether I think Hobson is correct or if I prefer a middle ground between Hobson and that modern science picture. Reguardless, the book is highly recommended for raising some important questions, presenting a solid evidence basis (often from very diverse sources which are not usually discussed in the context of autism) and for being highly accessible, being effectively popular philosophy and popular science.