Looking to the future

Autism: A New Introduction to Psychological Theory and Current Debate - Tiffany Watt Smith 2019

Looking to the future

THIS BOOK IS based on a volume written and published in the early 1990s. Writing this re-worked and updated account of psychological theory in autism research has highlighted the vast quantity of new information that has been accrued in that time. When Francesca began in autism research 30 years ago, it was possible to read every paper ever written about autism; in 1988, fewer than 200 papers on autism were published, and the total literature amounted to well under 3,000 specialist publications. Now, a search on PubMed for papers with “autism” in the title or abstract returns well over 4,000 results published in 2017 alone — a rate of more than 80 new papers per week. In total, the scientific literature on autism exceeds 60,000 papers as we write, and it would be impossible for any student to read more than a small fraction.

Over that time, we can see how things have changed for the better for autistic people and their allies, at least in the UK and other high income countries. Public awareness of autism has increased enormously, resulting in better understanding in the community. Identification of autism has likewise increased, reflected in higher diagnostic rates. Tailored educational provision is available, including training for teachers in mainstream schools, resulting in better further education and employment outcomes for autistic people. The growing grasp in the academic community of the breadth and complexity of the autistic spectrum has been beneficial in promoting a more sophisticated approach to study design and interpretation.

However, if we take as our goal a world in which an autistic person experiences no disadvantage as a result of their autism, we have a very long way still to go. Autistic people still experience high rates of mental health problems and have lower life expectancy, including high suicide rates. Quality of life is often low, with autistic people reporting limited autonomy and low rates of employment and independent living. Some groups remain very poorly understood — we know very little about autistic women and girls, including their experience of key life stages, such as puberty and the menopause. There is little information about the ageing process for autistic adults, including whether or not autism increases vulnerability to dementia, and how to care effectively for autistic adults in old age. When suitable knowledge and support are available, access to services is far from universal, even within the UK. Disseminating current findings and practices globally remains difficult, and there is a huge amount to learn about how autism is manifest in cultures beyond North American and European societies.

Autistic people also experience profound disadvantage as a result of commonly co-occurring conditions, including epilepsy, learning disability and language disorder. Treatments and supports for these continue to have limited effectiveness, as well as frequently being created and evaluated without autistic input. Moreover, many of the challenges experienced by autistic people reverberate through the family and beyond — parents and siblings experience stress and lower quality of life (Tint & Weiss, 2016), teachers of autistic children report high rates of burn-out (Boujut, et al., 2017). Community awareness of autism may be high, but Sue’s recent UK survey showed that perceptions of understanding remain low (Macmillan et al., 2018). In this chapter, we explore some of the ways to move forward from this position.

1. Autistic rights and advocacy in research

The autism rights movement, founded on the work of pioneers, such as Larry Arnold, Martijn Dekker, Dinah Murray, Ari Ne’eman, Jim Sinclair and Donna Williams, has made enormous strides since the millennium. Autistic-led organisations have come to the fore, such as the Autism Self-Advocacy Network (ASAN) in the USA and Autism Rights Group Highland or the Participatory Autism Research Collective in the UK. While it would be false to say such organisations are thriving — funding sources remain scarce and many are sustained by the energy and goodwill of a handful of community leaders — they are becoming recognised by the academic and policy-making establishment as a key part of any autism-related initiative. The progress of autistic rights can be compared with other civil rights movements — the most common comparison is often with the LGBTQ+ community. Many parallels can be drawn between these groups: both experience disadvantages that are created, intensified and maintained by majority-group insistence on normative behaviour, for example. Homosexuality was once a diagnosable ’disorder’ within the DSM (only removed in 1973), and associated ’cures’ were promoted. As such, recent strides in gay rights give hope to the autistic community and provide a valuable reference point for non-autistic allies. It is instructive to ask oneself as an academic, if autism follows the same transformational pathway as homosexuality, will my work still be relevant to autistic people in five, ten or twenty years’ time?

A key component of any endeavour to improve the lives of autistic people and their allies is to engage with those communities throughout the research process. Our position is that the incorporation of autistic perspectives into research is a matter of moral principle — neatly encapsulated by the disability rights slogan Nothing About Us Without Us. However, working with stakeholders also yields practical benefits, including facilitating recruitment to studies, and the fact that delivering research that matters to people will increase likelihood of translation into practice. Adopting an autistic rights framework is easiest when using participatory or collaborative methods to work alongside autistic people and their allies.

Engagement with autistic and autism communities (the former — people with a diagnosis and those who self-identify as autistic;1 the latter — their friends, family and the professionals who work with them) is not straightforward, however. Finding ways to include autistic children, people with a learning disability and those who experience barriers to communication is a particular challenge which must be overcome. In cases like these, the role of the parent in advocating for their children is absolutely crucial. In fact, one key way in which the parallel between the LGBTQ+ community and the autistic community breaks down is in the role of parent advocates. Parents are often at the coal-face of autism support, providing year-round, sometimes 24/7, care to children and adults with complex and intense support needs.

1We note that there is discomfort among both academic and non-academic stakeholders about the status of self-identification of autism. People mis-identifying themselves as autistic (perhaps when they in fact are experiencing a different form of neurodiversity) is a concern, as is the possibility that unscrupulous individuals might set themselves up as autistic advocates illegitimately. However, we also recognise the many reasons why an individual might want to identify as autistic without seeking independent medical endorsement.

Indeed, when reflecting on the history of autism support in the UK and beyond, it is clear how fundamental parents have been to the creation of infrastructure and knowledge about autism. Parents founded many of the leading autism organisations in the UK — the National Autistic Society, Scottish Autism and Autism Initiatives to name a few — and have been major donors to and fundraisers for autism provisions (e.g. schools, adult centres) and research. So we can see that understanding and recognising the parent perspective is also vital. But, when asked directly, parents and autistic adults frequently offer contrasting opinions on topics in autism research and practice, and this can make it hard for researchers to determine a clear, community-influenced direction for their work (e.g. Fletcher-Watson et al., 2017). One rarely acknowledged but growing group — as adult diagnosis rises, and those diagnosed under more recent, broad criteria grow up — is autistic parents. Since autism is highly heritable, they are more likely than most to have autistic children. Their perspectives and needs as a group are poorly characterised in the literature, despite the fact that autistic parents of autistic children provide a unique insight into both roles.

People often ask who is the ’right’ group to work with in a research project — and the answer is that all stakeholder groups have something to offer. If working on a study pertaining to autistic adults who are cognitively able, you might still want to involve professionals, service providers and family members — though this might be partners or children rather than parents. For example, if the project hopes to produce information relevant to mental health services, it will be important to have the perspective of clinicians represented, to maximise the chances of getting your findings translated into practice. What about non-speaking autistic people, or young children? Who speaks for them? Again, we would suggest that a range of voices should be present. Consider, as a parallel, a charity deciding what is the best way to help Muslim refugees in a faraway warzone. They can’t talk to them directly, so instead, they might bring in representatives of the Muslim faith, people who have lived in that part of the world and people with professional experience of supporting refugees. Some might belong to multiple such categories. It is clear to see that a person who shares the same faith will have insights that are complementary to those of an international aid worker — neither’s expertise trumps the other. The optimal solution needs to be found by all parties working together.

There are now some excellent guides for researchers who want to engage with the autistic and autism communities to help overcome this problem and others (see Fletcher-Watson et al., 2018). Next, we summarise a few examples of ways to engage at different stages of a typical research project in Table 10.1. These presume that most autistic collaborators are outside academia, but of course, this doesn’t have to be the case. A key goal for the future should be to improve access to research training, methods and resources for autistic people; encourage neurodivergent individuals into undergraduate and post-graduate study; and ensure that autistic-led research is a prominent part of the knowledge landscape.

Table 10.1 Ideas for participation in research for each stage of the project cycle

Research Stage


Formulating ideas

Spend time with some autistic people: as a befriender, volunteer, on twitter or socially

Writing a grant

Talk informally to the autistic people you know

Do an online survey about your research questions

Cost for autistic consultancy and advisers

Starting a project

Consider: do you need an autistic person on your interview panel?

Agree clear terms of reference and communication modes for collaborators

Designing materials

Review with autistic people and other relevant stakeholders

Consider aspects like contrast, font, layout and language, as well as content

Data collection

Check the space: do a sensory review, check for disrespectful posters

Use the Standard Participant Question Response (SPQR) to invite feedback on the experience2



Best scientific practice is best autism practice

Pre-register your hypotheses and analyses


Tell the people who took part and thank them

When talking about your research be respectful: don’t assume there’s no-one autistic in your audience

Share as creatively, widely and openly as you can

2. Deriving meaning from the constellation

If there’s one thing everyone can agree on, it is that autism is complicated. This complexity is especially striking in many of the areas of autism where our knowledge is most lacking: presentations that don’t adhere to the ’classic’ male type, multiple potential candidate genes and gene networks and their relation to behaviour, variety in developmental trajectories in childhood and adulthood and interactions between autism and cultural differences. All of these must be examined in relation to a shifting and subjective set of diagnostic criteria, further complicated by a rise in people self-identifying as autistic, without a formal diagnosis.

The role of the broader autism phenotype (BAP) also feeds in to this complexity. The connection, or division, between people with autism versus those who simply manifest many (subclinical) autism-linked traits remains unclear. Should we conceptualise autism as the tail end of a normal distribution? If so, then analysis of the relations between autism traits and other behaviours in the general population should have relevance for autism. For example, if we found that AQ scores mediate a positive correlation between Facebook use and depression, we could extrapolate that autistic people will be at higher risk of depression if they use Facebook and act accordingly. If, however, we believe that autism is qualitatively distinct from the general population’s trait distribution, then studies of the BAP in people without autism may have limited relevance for autism itself.

This debate about the relation between autism and autistic traits has caused a great deal of consternation in the community. Understandably, autistic people are angered by statements, such as “aren’t we all a little bit autistic?”, even when this is well meant. It would clearly be unacceptable to make the same statement about clinical depression, on the basis that everyone feels sad once in a while. On the other hand, ADHD is conceptualised as a dimensional, rather than a categorical, diagnosis. ADHD traits cause difficulties only for people who find themselves at the extreme end of the distribution, and even then, the degree of those difficulties depends on the environmental demands.

If we consider this question at each of our levels of explanation, the answer does appear more nuanced. At the behavioural level, it is certainly the case that there is a smooth continuum of scores on questionnaires, such as the autism quotient or social reciprocity scale: diagnosed people endorse more items but neurotypical people endorse some, and there is no clear water between. At the biological level, as discussed in Chapter 4, autism is in most cases a quantitative trait resulting, like height, from a combination of hundreds of common genetic variants, each of tiny effect. The same genetic influences therefore seem to operate on diagnosed autism and variation in autistic traits in the general population. However, some autistic people have pointed out that using this kind of information to suggest common ground between autistic and non-autistic people is a bit like saying someone is “a little bit pregnant” if they have a backache or swollen ankles.

At the psychological or cognitive level, the picture may be more complex and depend upon which theory one considers. For example, the ToM account suggests a possibly categorical difference; neurotypical people automatically and unconsciously track others’ mental states, but autistic people do not (or do so only through conscious calculation). Meanwhile, weak central coherence is hypothesised to be a cognitive style characteristic of autism but also seen in neurotypicals — notably the close relatives of those with autism and perhaps in individuals with musical, artistic, memory or maths talents. So whether or not it makes sense to see autism as on a continuum with neurotypicality depends on the level of description and the psychological theory considered.

Setting aside for a moment the continuum from autistic to autistic traits, heterogeneity within the constellation remains a profound challenge. As scientists, how might we respond? Option one is to try to parse autism into meaningful sub-groups, at the biological, psychological and/or behavioural level. Sub-grouping on the basis of behaviour can be useful for a practitioner in a specific setting (e.g. seating children with a similar reading level at the same table in class), but this is unlikely to be functional for research, because behaviour varies widely depending on the physical environment, current activity, age and so on. To date, studies looking for behavioural sub-types have rarely provided evidence for much other than a split between autism with, versus autism without, intellectual impairment and/or language disorder. Attempts to use biological information to define sub-groups have also foundered. A combination of the sample sizes and expensive and intrusive data types required to derive biologically defined clusters has been prohibitive to date, though efforts are underway (Charman et al., 2017; Loth et al., 2017). These approaches also suffer from the fact that they inevitably rely on models from personalised medicine. This system aims to identify sub-types of a particular disease in order to relate them to a specific treatment — as in Type I versus Type II diabetes, or when prescribing specific cancer drugs according to a patient’s genetic markers. This framework has limited applicability to autism, where we aim to improve self-defined quality of life, rather than offer a ’cure’, though it might have relevance for unwanted co-occurring conditions, such as epilepsy.

Setting aside behaviour and biology, then, can psychology do any better at identifying meaningful underlying sub-groups? So far, our review of the theoretical literature suggests we are still a long way from understanding autism in these terms. However, psychology may have a valuable role to define, measure and provide supports for the additional needs of autistic people. DSM-5 calls for characterisation of the individual, not according to a diagnostic sub-group (e.g. Asperger’s syndrome, PDD-NOS, as in DSM-IV), but instead using a detailed picture of their strengths and challenges — sleep problems, high IQ, supportive family, depression. To do this, we need three things: precise measurement, understanding of autism-specific manifestations (e.g. is depression in autism the same as depression in neurotypical people?) and matched autism-enabling interventions. Psychology offers methods and frameworks to achieve all of these.

Plan B for tackling the challenge of heterogeneity in autism is simply to embrace it. Rather than attempting to break down the concept into internally consistent sub-groups, can we endorse a model of autism as a common thread which unites diverse people? In much the same way that our nationality both defines us and simultaneously offers almost infinite variety. In this scenario, the role of psychology might be to identify what are the common threads that unite autistic people, and what are the boundaries of this form of neurodiversity?

3. Neurodiversity, co-morbidity and intersectionality

The future of autism research is further complicated, not just by variability in presentation of autism itself, but also variability in presence and manifestation of accompanying diagnoses — clinical anxiety, epilepsy, ADHD, etc. — and the intersectional influence of facets, such as race, gender or sexuality. It is well known that autism frequently presents alongside other medical diagnoses, but new data are showing that a large proportion of the autistic community also identify outside hetero-normative categories in terms of gender and sexual orientation (George & Stokes, 2018; Gilmour et al., 2012).

There’s no doubt that psychology has an important role to play in probing the experience and impact of co-occurring conditions and intersectional experiences. However, there is a risk that this complexity is erased by strict inclusion and exclusion criteria for studies attempting to achieve a statistically significant result — which is easier when the sample is relatively homogeneous. One possible solution is to focus less on diagnostic categories and more on trans-diagnostic domains of functioning, as in the Research Domain Criteria proposed as a new research framework by the National Institute of Mental Health in the USA (Insel et al., 2010). For example, one could design a study enrolling anyone with anxiety, regardless of their original diagnosis. A major challenge here would be identifying a funder and recruiting families — both of these often align under a specific diagnostic banner which may drive their decision-making. Either way, there is a push for larger and larger samples to be able to understand multiple influences on experience and outcome, and to derive recommendations that can be tailored to specific individuals. Of course, the opposite is also a valuable approach: small groups recruited to qualitative studies that take a ’deep dive’, to provide rich and detailed pictures of the multiple influences on lives and outcomes.

Regarding the influence of intersectionality in autism, psychology may play a role to characterise lived experiences and make informative comparisons between groups. However, this work must take place as part of an interdisciplinary approach. Psychologists who wish to address or take account of intersectionality in their research or practice need to stay on top of the latest philosophical, sociological, legal and political debates to ensure that their work is current and relevant to the broader issues experienced by autistic people of all stripes. More than that, ideally autism research would be at the forefront of developing these debates, with results feeding back into a socio-political agenda for change.

4. Delivering results that matter

It is our hope that the future of autism research will include not just a maintenance of the staggering productivity which has characterised recent years, but also a consistent and pervasive focus on delivering findings that matter to autistic people. As one example, ten research priorities, published in 2016 by Autistica and the James Lind Alliance, provide an excellent starting point for UK researchers from which to look to the future (Table 10.2). The list was developed in partnership with a range of stakeholders, with careful attention paid to representation and diversity in the sample. The topics proposed can be addressed from a broad range of disciplinary perspectives and have relevance to autistic people of all ages, abilities and background.

Table 10.2 Community research priorities, Autistica and the James Lind Alliance


Which interventions improve mental health or reduce mental health problems in autistic people? How should mental health interventions be adapted for the needs of autistic people?


Which interventions are effective in the development of communication/language skills in autism?


What are the most effective ways to support/provide social care for autistic adults?


Which interventions reduce anxiety in autistic people?


Which environments/supports are most appropriate in terms of achieving the best education/life/ social skills outcomes in autistic people?


How can parents and family members be supported/educated to care for and better understand an autistic relative?


How can autism diagnostic criteria be made more relevant for the adult population? And how do we ensure that autistic adults are appropriately diagnosed?


How can we encourage employers to apply person-centred interventions and support to help autistic people maximise their potential and performance in the workplace?


How can sensory processing in autism be better understood?


How should service delivery for autistic people be improved and adapted in order to meet their needs?

Identifying the goals of research is only half of the solution, however. We also need to consider how research is carried out. In this chapter, we have highlighted the importance of adopting an autistic rights framework, embracing heterogeneity and recognising diversity in the autistic community. These things are easier said than done, especially when resources are scarce and professional environments are competitive. So what needs to change?

First, the research establishment. The current emphasis on research impact — the extent to which new findings can be shown to make a difference — must continue and expand. Training researchers to value impact and to understand how to achieve it should be a core part of their professional development. Funders also need to recognise that impact may take many years to unfold and find a way to follow-up with researchers, holding them to their impact plans while accommodating this slow process. Achieving impact includes collaboration and outreach, but it is also linked to scientific quality. Reproducible results, consistent measurement across studies and collaborations to share data and develop large samples are all essential to build a strong evidence-based case for change.

Second, public understanding of science. A common concern within the scientific community over partnership with stakeholder groups is that doing so compromises the scientific integrity of the work; will ideas be ’dumbed down’ and findings simplified for public consumption? One solution is to nurture a scientifically expert community of practitioners, parents and autistic people. Of course, it still behoves academics to share jargon-free summaries of their work and present it to general audiences — it is unrealistic to expect community members without specialist training to read the latest issue of Nature Neuroscience or even Autism Research. But a degree of scientific understanding can facilitate partnerships and, crucially, give members of the public the confidence to question the headlines.

For practitioners, including clinicians and teachers, understanding the limits of research is essential to permit translation of findings into practice. Published studies in psychology are often very good at making generalised statements about averages across groups, but the practitioner must translate such generalisations into specific actions for their client or pupil. This is somewhere that the much-maligned medical approach should act as a beacon for practitioners in fields such as education and social work. Medical undergraduate training emphasises the changing nature of knowledge and the importance of keeping in touch with the latest research findings in order to apply them in practice. Medics are trained to understand basic science and to conduct evaluations in their services. In contrast, practitioners in social care and education are given little or no support and resources to do this, both during their training and once qualified, with negative consequences for the quality of support provided to autistic people and their families.

5. Current debates


Psychology research in autism has delivered thousands of papers based on millions of pounds of research funding, but a huge amount remains to be done before we can state that being autistic entails no systematic disadvantage in life. In part, this is because of the gap between research and practice. Research is often focused on long-term gains, benefit to future generations and generalised statements derived from group averages. In contrast, practice settings are focused on immediate needs for current service users, and tailored person-centred supports which take into account contextual and individual detail.

Psychology in the future needs to work harder to ask questions and employ methods that can increase translation of research into practical impacts on autistic people and their allies. Making these impacts meaningful and positive means partnering with autistic people in research and supporting autistic academics to achieve and lead new research programmes.

In doing so, we must not abandon what sets our discipline apart. Psychology builds strong theoretical frameworks that can distinguish between surface consequence and causal root. Psychologists have expertise in robust methods that merge scientific rigour with ecological validity, from RCTs to diaries and interviews, often in the same study. Psychology is a hub discipline, at the interface of biology, medicine, education and social sciences. For all of these reasons, psychologists are well placed to provide the raw material the autistic community needs to build the best possible future.

Big questions

Applying psychological theory to questions that matter is easier said than done. How can we strike the right balance between ’pure’ psychological enquiry to develop theoretical models and delivering practice-relevant and community- endorsed projects?

How can we — scientists and communities together — change standards in the academic establishment to reward science that is rigorous and makes a difference in people’s lives? Is there a way to develop science reporting in the mainstream media to be more accurate and responsible?

How can we partner effectively with autistic people, including those with learning disabilities, language disorder, or other barriers to communication? How can we balance input from autistic adults, parents of autistic children and those who fall into both categories? When there are disagreements within and between these groups, how should we respond?


Post-diagnosis most people go through a period of information seeking. I was no different when I was finally diagnosed in 1997 at 12 years old. I was desperate to understand what autism was and why this made me different. When I met other people on the spectrum, it was clear we all had things in common with each other, but we were also markedly different from each other. Why was that?

I remember thinking ’I don’t understand this, but there’ll be some experts somewhere who have got this nailed’. Around this time, psychological theories for autism were becoming increasingly common and the first text of this book had just been published. When I read about those theories, I remember thinking how illogical it was as someone growing up alongside other autistic people to try and use psychological theories to qualitatively describe autism in a homogenous way. I kept wondering if there was something I didn’t get — surely academics get that it’s not this simple?! It’s welcome to see the authors embrace heterogeneity in this text.

After a PhD and post-doc in autism science, I am now the director of science at Autistica — the UK’s autism research charity. Autism research has hugely matured during that time. One of the main things I spend a lot of time thinking about at Autistica is outcomes — how do we deliver a long, healthy, happy life for all autistic people? Psychological theories are central to that hope and can underpin the change we need. But to be the driver of that change they must:

1. 1 Embrace complexity. It is increasingly becoming clear that autism is a useful construct, but that we are moving towards an era where we need to think more broadly about neurodevelopment or neurodiversity. It’s evident both from research and the experiences of myself and others that the difficulties that autistic people face are neither simply a matter of the disabled person or society but an interaction between both. We also know that co-occurring conditions are enormously common and that this area is under-researched and offers opportunities to look at more specific populations.

2. 2 Embrace the community and partners. Researchers who operate in isolation often fail to ask themselves “why?” They can get lost in the theory, failing to focus on delivering a better life for autistic people. Researchers need to work with autistic people and allies and listen to their priorities. They should also work with experts in involvement, policy and communications to ensure their big ideas, enormous expertise and unique skillsets are utilised to harness transformative change.

We have come a long way since the first edition of this book, and I am incredibly hopeful for the future of autism research. I know by working together and embracing the challenges we face, we can truly change lives. If we do that, when the time comes for the next edition of this book, I’m sure we will be able to look back on autism research and reflect on an era where autism research delivered positive change.

Recommended reading

Cusack, J., & Sterry, R. (2016). Your questions: Shaping future autism research. London, UK: Autistica.

Fletcher-Watson, S., Adams, J., Brook, K., Charman, T., Crane, L., Cusack, J., Leekam, S. R., Milton, D. E. M., Parr, J., & Pellicano, L. (2018). Making the future together: Shaping autism research through meaningful participation. Autism. doi: 10.1177/1362361318786721

Sinclair, J. (1993). Don’t mourn for us. Our Voice: The Autism Network International, 1(3).