52 Chapter 4
For the great majority of children with classical autism who do not have
a known medical disorder, genetic factors seem of primary importance,
with twin studies demonstrating a heritability of over 90%, almost cer-
tainly due to multiple genes of small or moderate effect rather than a single
major gene. The heritable phenotype seems to be broad, stretching from
classical autism at the one extreme to mild partial variants at the other
extreme. The recurrence rate in siblings is roughly 3% for narrowly
defined autism, but is about 10–20% for milder variants. Family studies
suggest greater genetic overlap for the social/communication features and
language features than for the restricted/repetitive activities. Genome-
wide association studies and candidate gene studies have identified various
loci, some of which have been independently replicated. Perhaps such
findings may one day illuminate the pathogenesis of autism or contribute
to screening. Genetic factors may be less important in the aetiology
of the autistic features associated with severe and profound intellectual
disability; these may be primarily determined by widespread brain damage
(phenocopies). Obstetric adversity is of dubious aetiological significance.
Though extreme and prolonged early deprivation in grossly inadequate
institutions may result in features of autism, there is no evidence that
‘ordinary’ psychosocial adversities play any part in the aetiology of autism.
There is no evidence for theories that autism is caused by an early
traumatic event, or by parents’ insensitivity or lack of responsiveness to
their child. Nevertheless, these views are still held in some quarters, and
cause parents unnecessary distress.
Many researchers have hypothesised that autism results from a primary
fault in just one neurological system or just one psychological function.
It is equally plausible that autism reflects a distinctive combination of
structural or functional abnormalities. Neurobiological studies have not
identified a characteristic focal deficit: almost every portion of the brain has
been implicated by some neuroimaging or neuropathology study and no
localisation has been consistently replicated. However, the structural and
functional relationships between different brain regions may be abnormal.
Since individuals with autism have, on average, larger head circumfer-
ences and bigger brains, widespread neurodevelopmental abnormalities
may turn out to be more important than focal abnormalities.
Attempts to identify a primary psychological deficit in autism have fared
slightly better. Though no one theory has won universal acceptance, two
theories have been particularly influential. One theory suggests that the
primary deficit in autism is in ‘Theory of Mind’, that is, the capacity to
attribute independent mental states to self and others in order to predict
and explain actions (see Box 4.1). This sort of ‘mentalising’ deficit would
disrupt abilities that depended on the capacity to see another person’s
point of view, but would not interfere with abilities that simply required a
mechanical or behavioural understanding of objects and people. Another
influential theory is that the primary deficit in autism is in executive func-
tion, with the sorts of problems in planning and organisational skills that
result in poor performance on ‘frontal lobe’ tests. Other suggestions for the