Autism at the biological level
WHEN THE ORIGINAL version of this book was written, it was still necessary to say that autism is not caused by psychogenic factors and to directly refute the ’refrigerator mother’ myth by giving evidence (then still limited) of the biological basis of autism — e.g. high rates of epilepsy. Thankfully, this no longer needs to be argued for. There is now good awareness, in most places, that autism is a strongly genetic condition arising from a complex interaction of genetic and environmental factors. Nonetheless, huge efforts have gone into finding the genetic and neural bases of autism, with what some might consider a disappointing lack of major breakthroughs. Research has identified plenty of biological features that differ between people with and without autism, and also features that vary dimensionally with a specific behaviour. But at the time of writing there are no biological features that provide a distinctive marker, or specific cause, of autism (Muhle et al., 2018).
Heterogeneity of etiology may be one reason why progress has been slow; many people now talk of ’the autisms’ to reflect the belief that different individuals have different biological paths to autism. The research funds spent on finding the etiology of autism are also a matter of controversy and debate, with many stakeholders concerned that biological research is focused on finding a ’cure’ and arguing for greater funding towards more immediately improving the lives of people on the spectrum. We recognise the strength of feeling among the community and reject the notion that autism might be something to be cured. But it is wrong to suppose that biological research, even that focused on ’causes’, cannot align with community priorities. Through a better understanding of biology, we can develop meaningful opportunities for intervention for difficulties that commonly co-occur with autism and that autistic individuals might choose to address. An example might be developing drugs to modulate how the brain responds to sensory input, to allow an individual to choose to dampen down sensory experiences which they find overwhelming. We will return to these debates at the end of this chapter, after briefly reviewing the current state of understanding of the genetics and neural bases of autism, and how cognitive theory can inform studies of the biology of autism.
1. The contribution of genetics
The first evidence that autism had a strong genetic origin came from seminal twin studies that showed that identical twins, who share 100% of their genes, showed a much higher concordance (i.e. probability that, if one has autism, the other twin will too) than fraternal twins, who share on average 50% of their genetic material (Bailey et al., 1995; Folstein & Rutter, 1977). This finding, and the resulting high heritability estimates, have been replicated many times in subsequent and much larger studies (Tick et al., 2016). Autism also runs in families, with an increased likelihood of autism occurring in the younger brother or sister of an autistic child.
The field of genetics moves very fast, and specific information about candidate genes would soon date and be unhelpful here. However, it is important to note that the present consensus is that in the majority of cases autism is the result of the action of hundreds of common genetic variants, each of very small effect. In this respect, autism would not be different from other quantitatively distributed traits, such as height. Research is therefore moving away from looking for ’autism genes’ and towards constructing ’polygenic scores’, where an individual can be given a score according to how many of the autism-associated genetic variants they have. Such a score would, once we have data from sufficiently large autistic and non-autistic samples, give a probabilistic (not deterministic) estimate of the likelihood someone will show autistic traits or meet diagnostic criteria for autism.
By contrast, for a minority of those with autism — usually those who have additional intellectual disability — a rare genetic mutation can be found that is believed to cause their autism, because most or all people with that mutation are autistic. Discovering such rare mutations of large effect can be hard; the causal role of a mutation never seen before is difficult to establish since we all carry many unique mutations. It may also be that even these ’big hits’ from rare mutations depend for their ultimate effect on the genetic background of the individual; the other common genetic variants they inherit may intensify or ameliorate their developmental outcome.
Thus, autism is due to a mixture of common inherited genetic variation across many genes each of small effect and rare mutations of large impact. The former would be considered ’familial’ and the latter ’de novo’ in most cases. How can genetic research on autism move forward given this huge heterogeneity? One approach is to try to map the many different implicated genes onto a smaller number of biologically relevant pathways (Geschwind, 2008). If indeed there is convergence on particular pathways of importance, this approach may give clues to the development of therapeutics that could target particular commonly co-existing symptoms, e.g. anxiety.
Why hasn’t more progress been made towards understanding the genetic causes of autism? There are several major challenges, including heterogeneity within the constellation; the ’fractionated triad’ idea would suggest, for example, that searching for genes affecting one dimension of autism, such as an individual’s social interaction profile, will be more productive than searching for genes predisposing to autism as a whole. Less contentious reasons for slow progress include low statistical power. Compared to many other clinical groups, studies of autism genetics are at an early stage, with relatively small sample sizes. Hundreds of thousands of DNA samples may be needed, especially if each genetic variant has a tiny effect and hence a weak signal. In addition, genes may interact with one another and be turned on or off by genetic or environmental factors throughout development. Research into epigenetic factors that moderate the action of genes is at an early stage, partly because, unlike our DNA — which is the same in every cell of our bodies — epigenetic signatures will differ in different tissues and over the lifespan. Getting DNA samples from saliva or even blood is relatively easy compared to getting brain tissue for epigenetic analysis!
The ultimate aim of all this work is to find biomarkers that would aid in so-called personalised or ’precision medicine’ instead of a one-size-fits-all approach, understanding the different possible biological bases of autism should in principle lead to better individualised therapies (Geschwind & Staite, 2015). One example might be a drug to improve synaptic function, ultimately giving a person with a learning disability more capacity to learn new information. Of course, it will be vital that this work is informed by stakeholder views, helping to identify drug targets that are acceptable and important to the autism community, and avoid mere ’normalisation’ as an end in itself. One risk is that some biomedical researchers may feel that their work, being relatively far-removed from the everyday experience of autistic people and their allies, doesn’t require significant stakeholder input. However, we might also note that basic biomedical research takes many years — sometimes decades — to deliver results with a practical application. In which case, there is a particularly pressing reason to work with stakeholders from the very outset. Basic scientists must stay abreast of the latest thinking about autism to avoid their hard work and scientific innovations producing a treatment that nobody wants.
2. Candidate neurological underpinnings
Just like the search for the genetic basis of autism, the hunt for what might be different in the brains of people in the autism constellation has attracted huge interest but, perhaps, not progressed as far as some might have hoped. Once again, this is probably due, in part, to the biological heterogeneity that has led some authors to talk about ’the autisms’ rather than ’autism’. The ’fractionated triad’ notion may again relevant; Figure 4.1 shows how different brain regions may relate to different diagnostic features of autism. For example, one of the better replicated neurobiological findings in autism is early brain over-growth in the first four years of life (Zwaigenbaum et al., 2014). However, only a minority of autistic children show this pattern of accelerated early growth of head circumference (Ecker, 2017). There are also important contextual factors to consider: are head circumference norms up to date, applicable to different regions or populations and is overall growth and body size taken into account? Nevertheless, this finding has been interpreted as important and researchers have speculated that perhaps the normal pruning of the brain (by programmed cell death: apoptosis) and the ’use it or lose it’ principle of synaptic competition are disrupted in toddlers with autism.
Figure 4.1 Brain regions in relation to diagnostic features of autism
Reproduced from Amaral et al., 2008, with kind permission of the author
As well as possible differences in total brain volume, differences in specific brain regions have been reported, and researchers have attempted to link these to the behavioural features of autism (Ecker et al., 2015). The most frequently implicated regions include frontotemporal and frontoparietal regions, the amygdala — hippocampal complex, cerebellum, basal ganglia and anterior and posterior cingulate. However, while structural neuroimaging studies may show group differences between those with autism and without, there is wide variability within groups too, and there are no findings to date that reliably mark autism as different at the individual level. Individual differences in brain and behaviour, as well as what we know of the functions of different brain regions, have led researchers to link specific regions to autism features. For example, the frontotemporal regions and amygdala have been linked to socio-emotional processing, and the frontostriatal system (orbitofrontal cortex and caudate nucleus) have been implicated in repetitive/stereotyped behaviours. Whether these differences are a cause or an effect of autism is open to question; the amygdala is also enlarged in non-autistic children with anxiety disorders, for example. We know the brain is changed by what we do and learn, as in the well-known finding of larger hippocampal volume in London taxi drivers (Maguire et al., 2000). Therefore, it is likely that brain differences reflect, at least in part, differences in the lives (more stressful) and behaviours (more repeated) of autistic people.
Beyond measuring the size of different brain regions, differences in structural and functional connectivity have been reported. At the micro level, many researchers think that autism will prove to be defined by differences in synaptic functioning, maybe having to do with a modification of neural excitation and inhibition in which the neurochemicals GABA (gamma-Aminobutyric acid) and glutamate are key. At a macro level, it has been suggested that local connectivity is increased and long-range connectivity decreased in autism (Ameis & Catani, 2015). Diffusion Tensor Imaging (DTI) allows visualisation of the integrity of white matter, including the major neural highway connecting the right and left hemispheres, the corpus callosum. Abnormalities in white matter have been reported, but since development of white matter is inherently affected by grey matter development, it is unclear if these are primary or secondary in autism. Ideally, information from structural and functional neuroimaging would be combined with histological studies of post mortem tissue, but to date, there are still relatively few donations in international brain banks.
When the previous version of this book was written, it mentioned the promise of functional neuroimaging studies, which had just begun with PET (positron emission tomography). The far less invasive technique of magnetic resonance imaging (MRI) was yet to be applied to autism. Since that time, there has been an explosion of functional imaging research using fMRI. Among other things, this has identified a network of brain regions reliably activated during Theory of Mind (ToM) tasks in neurotypical volunteers, including medial prefrontal cortex, posterior superior temporal sulcus/temporoparietal junction, praecuneus and temporal poles. Studies with autistic volunteers engaging with the same ToM tasks, typically show reduced and/or less coordinated activation across this ’mentalising network’. Interestingly, this network overlaps with the so-called default mode network: the functionally interconnected set of brain regions (posterior cingulate cortex, praecuneus, medial prefrontal cortex, temporoparietal junction and hippocampus) that are less active when volunteers engage in any task and most active when volunteers are ’at rest’. Differences in the intrinsic functional connectivity of the default mode network have been reported in autism, with the suggestion of a developmental shift from hyperconnectivity between default mode network nodes in childhood to hypoconnectivity in adolescence and adulthood (Padmanabhan et al., 2017).
Other forms of neuroimaging are also being used to explore brain function in autism, including electroencephalography (EEG), magnetoencephalography (MEG) and near infrared spectroscopy (NIRS). A recent systematic review of autism research using EEG and MEG (O’Reilly et al., 2017) found “a general trend toward an under-expression of lower-band wide-spread integrative processes compensated by more focal, higher-frequency, locally specialized, and segregated processes” as well as fairly consistent reports of atypical lateralisation (increased left>right functional connectivity ratio). Beyond imaging the active brain, techniques such as transcranial magnetic stimulation (TMS) and transcranial direct current stimulation (tDCS) can be used to modulate the functioning of specific brain regions. These methods are being used to interrogate brain functions and test cognitive theories in neurotypical volunteers. Experimental therapeutic uses are also being explored in a range of clinical groups, including autism, although too few properly controlled trials exist to draw conclusions at this time.
3. Other biological influences
Although the genetic contribution to autism is large, even identical twins don’t show 100% concordance, and gene-environment interactions may be hidden in traditional estimates of heritability. Are there environmental factors that contribute to the etiology of autism? Almost certainly. Although to date the evidence regarding aspects of our environment to which we are widely exposed (e.g. pollution, everyday chemicals or foodstuffs) is weak. There are rare exposures that can be directly linked to increased rates of autism; for example, the anti-epilepsy drug valproate appears to be linked to autism if taken during pregnancy (Christensen et al., 2013).
A review of meta-analyses and systematic reviews, at the time of writing, found no link between autism and vaccination, thiomersal exposure, maternal smoking or assisted reproduction (Modabbernia et al., 2017). Birth complications associated with reduced blood supply/oxygen or trauma to the infant showed strong links to autism, while other aspects of pregnancy, such as birth by caesarean and maternal obesity or diabetes showed weaker connections, with evidence insufficient to date on the role of nutrition. Evidence of a link between autism and older paternal age is strong, and may be due to genetic changes in sperm produced later in life. Modabbernia and colleagues consider a range of mechanisms for the highlighted environmental factors including “non-causative association (including confounding), gene-related effect, oxidative stress, inflammation, hypoxia/ischemia, endocrine disruption, neurotransmitter alterations, and interference with signalling pathways”. They also highlight the limitations in research in this area and the need for “prospective design, precise exposure measurement, reliable timing of exposure in relation to critical developmental periods and … genetically informed designs”.
Of course, discovering the effects of possible environmental factors is difficult because we can’t easily or ethically run RCTs of exposures. For this reason, animal models are important; for example, scientists can control the age of mouse fathers to check that paternal age effects in humans are not just a reflection of autistic traits, making finding a partner and having children happen a bit later in life. Mouse models are used to study the effects of genetic perturbations, as well; ’knock-out’ animal models help to uncover what key genes do and what the effects are of specific mutations. Journalists are fond of headlines saying, “Scientists Create Autistic Mouse”, but, of course, these are wildly misleading! Animal models are probably most useful in tracing the mechanistic effect of identified genetic changes, through proteins formed and their functions in an organism. Looking for ’autistic behaviour’ in an animal is far less likely to be productive; even if repetitive grooming or burrowing can be seen in a mouse, how do we know that has anything in common with repetitive behaviours in autism? One of the reasons some scientists are keen to see a connection is because such animal analogues provide the possibility of trialling interventions, either genetic or pharmacological. For example, studies with mouse models of Rett syndrome, a degenerative disorder caused by mutations in the X-linked gene MECP2, have shown the ability to effect remarkable reversal of impairments closely aligned to the clinical symptoms in humans (e.g. motor and breathing problems), even in mature animals (e.g. Tillotson et al., 2017). However, the idea of treating or preventing autism — as opposed to alleviating the disadvantageous things that often accompany autism, such as intellectual disability, language impairment, epilepsy, anxiety — is not only ethically controversial but deeply problematic.
4. Cross-pollination between neurobiological and cognitive explanations
We still have no clear neurological or broader biological explanatory model of autism, neither to explain onset nor to predict progression across the lifespan. In fact, we are still struggling to understand the typical brain. So how can the neurobiological data we have impact our understanding of autism?
First, they can be used to debunk myths about the causes of autism. Twin studies in the 70’s put an end to the, then-dominant, psychogenic models of autism that placed responsibility for autism on parent behaviour. Findings of brain changes that date to the first trimester of pregnancy suggest that later environmental exposures cannot be sole causes of autism.
Second, they can be used to identify treatments with potential to impact epilepsy and possibly intellectual disability. If biomarkers for these could be found, early or even preventative treatment might be possible.
Third, they can potentially inform our psychological theories, since some psychological theories make specific biological predictions. One example is the ’broken mirror’ theory of autism, which locates the origin of social and communication difficulties in a faulty ’mirror neuron system’ (Williams et al., 2001). Mirror neurons, first discovered in monkeys, are neurons that fire when the animal performs or sees a specific action performed. The broken mirror hypothesis proposed that a ’faulty’ mirror neuron system in autism impedes imitation and social cognition. This psychological theory, therefore, makes predictions testable through neuroimaging, as well as through psychological experiments (e.g. reduced interference of observed movements on own movements). However, robust evidence that autistic and neurotypical groups differ in the functioning of the mirror neuron system has not, ultimately, been forthcoming (Hamilton, 2013).
It is also worth highlighting that the link between psychological and biological research is a two-way street: our cognitive theories can also help direct research into the biology of autism. Perhaps the most obvious example is the use of theory and well-designed cognitive tasks in functional neuroimaging (Philip et al., 2012). Psychological theory can also aid in genetic studies, for example helping us rethink who is ’affected’ in a family pedigree. Based on the hypothesis that unusual eye for detail may underlie specific talents in autism, overlapping genetic influences were found on parent-reported autistic traits and special skills (in music, maths, art or memory) in 8-year-old twins in the Twins Early Development Study (Vital et al., 2009). If you wanted to explore genetic influences in a family with an autistic child, you might therefore want to note who has talents, as well as who has autism (or both).
Fourthly, animal models need to be informed by psychological theories of the behaviours concerned. For example, if you are interested in whether knocking out a specific gene might contribute to face recognition deficits seen in autism, you need to think about your theory for face processing differences. If your theory is that people with autism do not become face experts, due to looking less at eyes and faces, you might want to test identity recognition in your knock-out mouse; this is best done through olfactory recognition tests, since mice use smell more than sight to identify other mice. However, if your theory is that face processing is different in autism because of detail-focused cognitive style, you wouldn’t test your mouse’s olfactory skills — you can’t separate global and local processing in smell — instead you’d use a visual task where responding to either parts or wholes was rewarded.
5. The quest for biomarkers
One specific manifestation of the application of neuroscience methods to understanding autism is in the quest for early biological markers (Loth et al., 2016). Such biomarkers might be used to identify autism pre-natally or in infancy, or to identify sub-types of autism (one of “the autisms” perhaps). This normally involves recruitment of a sample who are more likely than the general population to get an autism diagnosis and tracking them pre- and post- natally. One such group is the infant siblings of autistic children. We will review the findings and (methodological, statistical) challenges of this technique in Chapter 7, but here we want to draw attention to some conceptual issues for the field.
First, the pursuit of biomarkers raises a series of ethical issues. Most pressing is the question of how to apply the new knowledge, if a pre-natal marker is identified. Autistic activists have noted the accepted practice in some countries of early termination of pregnancy in the event that a chromosomal disorder is identified in the foetus — as often happens in the case of Down syndrome. Autistic people and their allies are understandably deeply concerned that the same practice could become available, even usual, if a pre-natal marker for autism were detected. It is hard for neurotypical people to imagine how distressing it must be to think parents in the future could be offered the option to terminate a pregnancy if their unborn child had the same condition as you. Researchers and clinicians have pointed out that there is scientific value in identifying very early, including pre-natal, markers of autism and that this research quest should not be construed as a ’prevention of autism’ agenda. However, it is understandable that the community remain concerned. We should all remain cognisant of these issues and be vigilant in ensuring that new discoveries result in understanding and practical applications that support and enable the autistic community.
Additional ethical issues are raised by the methodology commonly used to probe for early signs of autism: namely, the recruitment of infants in high-likelihood groups, and the collection of detailed data from them and their families over sometimes very long periods of time. Sue recently conducted research into the attitudes of the autism community to this work (Fletcher-Watson et al., 2017a, 2017b). The international survey uncovered overwhelming support for the endeavour in general, and an impressive level of endorsement of scientific priorities (e.g. earlier diagnosis) but also some important details. For example, phrases like “higher chance” or “higher likelihood” were endorsed above the term “at risk”, which is commonly used to describe infant siblings recruited for studies. Parents specifically valued transparency from the research team and wanted to be fully informed about their child’s scores at each data collection point — something that is rarely permitted by ethical review panels. The results highlight the value of asking people about their attitudes and experiences, and, as far as possible, aligning research protocols and study communications with community values.
A key focus for the pursuit of early biomarkers for autism, agreed upon by researchers and many in the autism community, is earlier diagnosis. This leads to our second issue for the field: can we uncritically accept earlier diagnosis as an important target? On the one hand, families are keen to reduce waiting times from first concerns to confirmed diagnosis. If a biomarker were identified it could streamline the process and produce economic savings if fewer professionals were involved, compared with the multi-disciplinary assessment that is currently considered best practice. However, replacing the current system with a single blood test or brain scan would take away the rich and complex appraisal of strengths and needs that is a key feature of the diagnostic pathway. Diagnosis should be more than just a label, ideally, it’s about families working with an expert team to understand the individual and make a plan for their future support needs. So even if a biomarker were identified, we would still want to accompany this with a series of assessments and many aspects of today’s, slow, laborious, but thorough, diagnostic pathway should be retained.
What about earlier diagnosis in the sense of “earlier in the child’s life”, as opposed to “more rapidly after concerns are raised”? Would this be a positive outcome for autistic people and their allies? Again, this might be a mixed blessing at best. One worry is that diagnosis that precedes parents raising their own concerns could have a negative impact on the family, compared with a diagnosis that offers an explanation for a difference that has already been noticed. While we’re not aware of research directly addressing this question, it easy to imagine that a diagnosis that provides a helpful explanation of something which is already apparent to the family would be received more positively than a label applied out of the blue.
But wouldn’t earlier diagnosis allow earlier intervention? And that would be beneficial for families? Indeed, support during periods of developmental plasticity in infancy and toddlerhood could target the fundamental skills that psychological theory and research say underpin later language, cognitive, social and learning outcomes. This is true, but we should also note that there are no robust evidence-based interventions for very early life. As it stands, parents might receive a diagnosis in infancy but then be left without suitable support for months or years. Although work is being done to develop options for use in infancy (Green et al., 2015, 2017; Rogers et al., 2014), we know from other areas of the autism literature how gradual progress is when creating new interventions and especially translating these into practice (Dingfelder & Mandell, 2011). Huge swathes of the autistic population still do not receive any specialist support even at older ages (Salomone et al., 2016) and non-conventional approaches without supporting evidence are rife (Salomone et al., 2015).
This is a chicken-and-egg problem to some degree: we can’t develop the interventions until we have a diagnosed population, and we ought not to diagnose the population until we have the interventions. Furthermore, in any intervention we have a responsibility to look for possible harms, and to seek the involvement and endorsement of the autism community. This is especially challenging and important when intervening in infancy: participants cannot advocate for themselves and the long-term effects of modifying the early environment are not known, and could be profound. Autistic advocates have raised serious concerns about some practices, for example, which they point out systematically train autistic children to comply with adult instructions, and extinguish behaviours (such as hand-flapping) which are part of the child’s self-expression and self-calming repertoire (Bascom, 2012). We ought not to recommend interventions for infancy without a deep understanding of the long-term and interactional effects of the change — and this will take many years to describe accurately.
6. Current debates
Large sums of research funding are currently spent on trying to understand the biological and neurological profiles associated with autism, whereas there has been a relative paucity of funding for studies into service, intervention and societal issues (Pellicano et al., 2013). It would be good to see an equal investment in autism research likely to have a direct and positive impact on people alive today.
While plenty of group differences and trait relationships have been found, there are currently no clear biological markers, nor powerful bio-based explanatory models of autism. The pursuit of a biological explanation can add value, not least by de-bunking some dangerous myths about autism, and may contribute to treatments for things like epilepsy. They may also be important in understanding variability within the autism constellation, or help us make sense of differences between ’the autisms’. However, to make the most of this scientific endeavour, neurobiological studies need to be informed by psychological theory (and, eventually, vice versa). Moreover, it is essential that these discoveries are embedded in a strong ethical framework ensuring that any findings are put to positive use.
In pursuit of a model of the many autisms, which are the promising routes to take: genetic? neurological? psychological? behavioural? Which methodologies are more likely to arrive at a place where we can meaningfully and usefully identify sub-groups?
If a biomarker for autism was found, but was (as is likely) not 100% reliable — how would we handle mismatch between people who have the biomarker, and people who show high levels of autistic traits but don’t have the biomarker? Would people with real needs be denied help because they don’t meet this new biological criterion for autism?
Biology in itself doesn’t tell us how to make the world more autism- enabling, or what tack a teacher should take. Taking the example of Down syndrome, knowing the genetic cause hasn’t yet resulted in bespoke supports to optimise individuals’ development. Similarly, knowing the brain differences in the case of rare conditions such as agenesis (failure to form) of the corpus callosum doesn’t in itself tell a teacher how to teach to a child’s strengths.
How can we test and develop the translational endeavour? There is a requirement for better communication between scientific disciplines, and experts who combine a deep understanding of both molecular biology, clinical medicine and beyond.
COMMUNITY CONTRIBUTION: ANYA UTASZEWSKI — AUTISTIC, COMPOSER AND DATABASE ADMIN
In 2008, a local charity forwarded a letter from Brighton and Sussex Medical Schools, inviting people with a diagnosis of Asperger’s to take part in a study looking at why some people showed differences in social behaviour. I was curious to find out more and to have the experience of being a participant in a study whose topic interested me, and that I hoped would also be a learning opportunity.
I strongly oppose a coercive ’cure’ for autism and am deeply perturbed by the idea that one day a pre-natal test for autism could exist. My first email to one of the researchers was to enquire as to whether their research would in any way be used to contribute to these. I was assured that this would not be the case and that the researcher himself was supportive of autism rights.
Having completed some psychological questionnaires, I was ready for the MRI scan. A friendly assistant gave me a gown to change into and reminded me to remove anything of metal from my person. I lay down on the flat ’bed’ that would slide me into the scanner and was given four buttons to press with my dominant hand, correlating to available answers to questions I would be asked once in position.
The ’bed’ moved into the MRI ’tunnel’. To begin with, I was to lie still and then shortly afterwards, look up towards a mirror on which was reflected a screen displaying images accompanied by multiple choice questions.
The only thing I had known about having an MRI was that it was a very noisy process. I was quite anxious about this as I’ve always had very sensitive hearing. Once the scanner started making its noises, I was pleasantly surprised. The rhythmic repetition, the deep bass pitch and textural density, accompanied by occasional gentle juddering sensations, were beautiful. Then suddenly the sounds changed to something reminiscent of a ’60s science fiction weapon, or firing a spaceship’s guns on an old DOS game! I felt sonically immersed; this was ’surround sound’ in the most literal sense I could imagine.
After over an hour in the scanner, the researchers had the data they needed and the process was complete. I asked if I could have a picture of my brain (shown in Figure 4.2) and if I could be emailed a copy of the study’s findings once they were ready. Both requests were kindly granted.
Since this experience, I have taken part in many other studies, two of which involved MRI scans. I continue to enjoy both the scans and the opportunity to learn from these opportunities and hope that I might, in some small way, be contributing to helping others.
Figure 4.2 Images of Anya’s brain from the MRI this writing describes
Ecker, C. (2017). The neuroanatomy of autism spectrum disorder: An overview of structural neuroimaging findings and their translatability to the clinical setting. Autism, 21(1), 18—28.
Geschwind, D. H. (2015). Gene hunting in autism spectrum disorder: On the path to precision medicine. The Lancet Neurology, 14(11), 1109—1120.
Pellicano, L., Dinsmore, A., & Charman, T. (2013). A future made together: Shaping autism research in the UK. London: Centre for Research in Autism and Education, University College London.
Vorstman, J. A., Parr, J. R., Moreno-De-Luca, D., Anney, R. J., Nurnberger Jr, J. I., & Hallmayer, J. F. (2017). Autism genetics: Opportunities and challenges for clinical translation. Nature Reviews Genetics, 18(6), 362.