cultural variations in folk-psychology count against the nativist hypothesis
(Lillard, 1998), it does not really do so – on two counts. First, what
evidence there is concerning mind-reading development in children cross-
culturally suggests a common developmental trajectory (Avis and Harris,
1991; Naitoet al., 1995; Tardif and Wellman, 1997). Second, variable
expression in varying circumstances is just what one might predict of an
innate mind-reading module, in any case – just as the language-module
develops diVerently in the context of diVerent natural languages.
4.2 Abnormal developmentIf the human mind-reading capacity is indeed innate and domain-speciWc,
then it is plausible – as argued in chapter 3 – that there is a modular
processing system on which it runs. Where there are domain-speciWc
modular systems, there are possibilities for special cognitive impairments.
And there does indeed seem to be such an impairment in the case of
mind-reading. It is known asautism, a developmental disorderWrst iden-
tiWed in the 1940s (Kanner, 1943; Asperger, 1944). Autism is known to
have a substantial genetic component, and is clinically deWned in terms of a
triad of impairments:
(1) Abnormalities of social behaviour and interaction;
(2) Communication diYculties, both non-verbal and in conversation;
(3) Stereotyped, unimaginative behaviour and failure to engage in pretend
play.
(Wing and Gould, 1979; American Psychiatric Association, 1987; World
Health Organisation, 1987). Autistic children seem ‘aloof’, often treating
other people as objects; they have diYculty in interacting socially with
both adults and other children; and while many can successfully acquire
language, they have diYculty with the pragmatic aspects of language-use
(for example, not realising that when Mother asks whether the fridge is
empty, you do not answer ‘No’ on the grounds that it contains a single
mouldy lettuce leaf). They also have a tendency to become engrossed in
repetitive and stereotyped tasks which normal children wouldWnd dull and
monotonous. But their central problem seems to bemind-blindness–a
speciWcdeWcit in mind-reading ability.
The diYculty that autistic children have with mind-reading has been
revealed by their very low rate of success on the false-belief task, even at
comparatively advanced ages (Baron-Cohenet al., 1985). The poor perfor-
mance of the autistic children in these tests can hardly be explained by any
general learning diYculties or general cognitive impairments, since the
autistic children do markedly less well than children with Down’s syn-
Developmental studies 95