This study examined the effects of adult “recasting” in sign language on the acquisition of specific syntactic-semantic structures by six deaf children between 9 and 76 months who were primarily at the one-sign utterance stage of development. In “recast” replies in conversation, the child's utterance is redisplayed in an altered sentence structure that still refers to the central meanings of the first sentence. Syntactic-semantic structures targeted for input intervention by teachers and parents using recasts included subject–verb relations, attribution, negation, subject–verb–object relations, conjunction, and conditionality. Recasting triggered the acquisition of new syntactic-semantic structures in American Sign Language and English which were evident in the spontaneous production of previously non-used sign utterances.

This study examines the ability of good and poor spellers at grades six and ten to generate orthographic and phonetic derivatives for three predominant vowel alternation patterns characteristic of internal derivational morphology. Results support the hypothesis that a productive knowledge of these patterns in orthography precedes a productive knowledge of these patterns in phonology. Further, orthographic (visual) information was found to be superior to phonetic (aural) information in accessing the appropriate derivational morphological rules. An order is identified for the acquisition of a productive knowledge of the three vowel alternation patterns in both orthography and phonology. Based on these results and analyses, instructional implications for both spelling and vocabulary are offered.

The spelling performance of a 17-year-old developmental dysgraphic. K. M., is compared to that of T. P. (Hatfield & Patterson, 1983), an acquired dysgraphic. Both make errors which are phonologically valid and spell regular words better than irregular words. Within the words, relative difficulty in spelling is also similar and may result from the effects of length, frequency and. for exception words, the presence or absence of a similarly spelt word. The spelling performance of K. M. may be interpreted as reflecting a phonological routine. The correspondences involved in this system do not differ from those used by normal children of the same spelling age. Subtle deficits in reading may have contributed to spelling difficulties.

Linguistic research on children's acquisition of writing concerns itself largely with the sentence as the highest unit of analysis. However, an adequate account of written language acquisition must deal with the text-productive ability, including devices for cohesion and compactness. The intersentential cohesive devices used by Dutch-speaking children aged 8–9 (N=14) and 10–11 (N=14) in narrative texts are analysed and compared in this report. The coding of the cohesive devices followed Halliday and Hasan (1976). The writing samples were also analysed for degrees of cohesiveness and compactness, using the formulae suggested by Scinto (1983). Our findings indicate a developmental trend in the construction of more cohesive and compact narrative texts.

Deaf students' recognition of indefinite pronouns and quantifiers was tested using written materials in the form of comic strips that provided pragmatically appropriate context. One hundred and eighty-seven profoundly hearing-impaired students, aged 7–23 years, served as subjects. There were significant developmental trends for both the indefinite pronouns and the quantifiers, with the quantifiers significantly more difficult than the indefinite pronouns. A comparison of the results with predictions drawn from theoretical linguistics and with predictions drawn from Developmental Sentence Scoring (Lee, 1974) data for hearing children indicates that theoretical predictions are more accurate for hearing-impaired students. This may be due to differences in methodology (DSS reports spontaneous spoken language; the present study reports comprehension of written English) and to educational practices with hearing-impaired students.

Blank-trial probe discrimination learning tasks were used to evaluate the hypothesis-testing abilities of 15 language-disordered and 30 normally developing children matched for mental age and language age. Children were presented with a series of two-dimensional learning set and orthogonal problems. No significant group differences were found in the learning set problems. All the children reached learning set criterion quickly, used a high proportion of simple object hypotheses, and maintained a high proportion of confirmed hypotheses. On the orthogonal problems, however, the language-disordered and mental-aged- (MA) matched children performed significantly better than the younger normal children. These findings suggest that the cognitive and linguistic deficits language-disordered children exhibit do not reflect an underlying failure to generate and test hypotheses. This conclusion, however, does not seem to apply to all language-disordered children. In the final section of the paper, language-disordered children's cognitive strengths and weaknesses are interpreted within the context of an information-processing model.