On the Spectrum

Chapter 8
association of maternal vitamin D deficiency during pregnancy and later development of ASD
(Vinkhuyzen, Eyles, Burne, Blanken, Kruithof et al. 2016). Although replication of this finding
is warranted, it would be interesting to evaluate whether prenatal vitamin D supplementation
could decrease the number of babies born with ASD.
Importantly, it remains essential to better identify young children at risk for developing
symptoms of ASD as early as possible. There is evidence that early rigorous intervention,
consisting of intensive in-home therapy in language and social reciprocity, leads to better
outcomes in childhood (Estes, Munson, Rogers, Greenson, Winter et al. 2015).
CONCLUSION
In this thesis, we assessed multiple aspects of the neurobiology of ASD along a continuum
of traits. We discussed the use of continuous phenotypes of ASD in the search for the
underlying neurobiology and considered heterogeneity in symptoms and etiology as
potential complicating factors in this type of research. So, to conclude, we can ask the
question that was proposed by Happé and colleagues: is it time to give up on a single
explanation of ASD (Happe, Ronald and Plomin 2006)? The answer is probably yes; it is time
to give up on a single neurobiological explanation. It is probably best not to speak of the
search for a “neural signature” of ASD, which implies a specific pattern of brain differences
with nearly perfect sensitivity and specificity for the disorder (Lilienfeld et al. 2015). It is
more likely that the inconsistency that we observe in studies of the neurobiology of ASD
reflects actual heterogeneity across the autism spectrum itself. It, therefore, is unlikely that
a single consistent neurobiological phenotype will be found that adequately describes the
entire autism spectrum (Bernhardt et al. 2016). The neurobiology of ASD is likely highly
idiosyncratic, in the sense that every patient with ASD may show their own deviation from
typical brain development (Hahamy, Behrmann and Malach 2015). However, even if ASD is
a broad diagnosis, large-scale neuroimaging studies of patients with the diagnosis can lead
to identification of relevant and common brain correlates. Ultimately, on the individual level,
each person with ASD may exhibit their own profile on each of these parameters, and perhaps
such profiles could ultimately guide treatment decisions. Therefore, it would be too early to
quit the search for the underlying neurobiology of the broader spectrum of ASD altogether,
or to completely abandon the case-control design.
Importantly, behavior originates in the brain and can ultimately only be understood from
the brain, even if differences are not obvious or even visible. As analysis technology improves,
neuroimaging research remains an obvious and promising way to tap into this complex
system. Continuous and categorical phenotypic approaches should complement each other
in the search of neurobiological markers of psychopathology. At the same time, the search
204