Cross-sectional studies of grammatical morphemes in autistic and mentally retarded children.

The frequency of occurrence of functors in obligatory contexts was studied in verbal autistic and mentally retarded children matched for nonverbal mental age, and the percentages of correct use of functors were rank-ordered. The grammatical complexity of their language was also described using a transformational grammar. The data were compared to those obtained in a normal group matched for mental age and to the data presented by Brown (1973) and deVilliers and deVilliers (1973) in younger children. The autistic subjects omitted functors frequently and independently of the grammatical complexity of their language. The rank ordering of morphemes was consistent within both the autistic and mentally retarded groups but showed no correlation between the two groups or to the rank ordering described by deVilliers and deVilliers. It is suggested that functors in autistic subjects may develop in an atypical but consistent order and that this may be due to specific semantic deficits, particularly in the areas of person and time deixis.