S. Baroncohen et al., DO CHILDREN WITH AUTISM USE THE SPEAKERS DIRECTION OF GAZE STRATEGY TO CRACK THE CODE OF LANGUAGE, Child development, 68(1), 1997, pp. 48-57
Normal toddlers infer the referent of a novel word by consulting the s
peaker's direction of gaze. That is, they use the Speaker's Direction
of Gaze (SDG) strategy. This is a far more powerful strategy than the
alternative, the Listener's Direction of Gaze (LDG) strategy. In Study
1 we tested if children with autism, who have well-documented impairm
ents in joint attention, used the SDG or the LDG strategy to learn a n
ovel word for a novel object. Results showed that although 70.6% of ch
ildren with mental handicap passed the test by making the correct mapp
ing between a novel word and a novel object, via the SDG strategy, onl
y 29.4% of children with autism did so. Instead, their reliance on the
LDG strategy led to mapping errors. In Study 2 a group of normal chil
dren, whose chronological age (24 months old) was equated with the ver
bal mental age of the 2 clinical groups in Study 1, was tested using a
similar procedure. Results showed that 79% of this normal group passe
d the test by making the correct mapping between a novel word and a no
vel object using the SDG strategy. Taken together, the results from bo
th studies suggest that children with autism are relatively insensitiv
e to a speaker's gaze direction as an index of the speaker's intention
to refer. This result is consistent with previous findings showing th
at children with autism are relatively ''blind'' to the mentalistic si
gnificance of the eyes. Discussion centers on how the absence of an SD
G strategy might disrupt specific aspects of language development in a
utism.