Contextual influence on the language production of children with speech/language impairment.

To investigate the possible contextual variability of language and interaction, data from 10 children with speech/language impairment and three different types of conversational partners (a parent, a peer and a clinician) were collected. The dialogues were analyzed with respect to the characteristics of the dialogues as whole, to the dominance and the productivity of the three types of conversational partners, and to the productivity, fluency and grammatical structure of the language production of the children with speech/language impairment. Productivity was measured as mean length of utterance (MLU) in words, number of utterances, number of different words and proportion of complete and intelligible utterances. Fluency was measured as the proportion of utterances containing a maze. Grammatical structure was analysed with respect to the use of grammatical morphemes, word order patterns, the occurrence of expansions and the complexity of verb forms. Significant differences, in particular between child-child and adult-child dialogues were found. The adult-child dialogues were asymmetrical, where the adults dominated through asking many questions and talking much. The peer dialogues were more dynamic and equal. In these dialogues the children with speech/language impairment took a more active role as a conversational partner. The children produced more utterances and different words with the adult partners, but variables related to fluency and grammatical structure did not vary as a function of the conversational partner. The results imply that children with speech/language impairment practise different aspects of their communicative ability with different types of conversational partners. In particular, it is important to provide them with opportunities to interact with peers, in order for them to develop their skills as independent conversational partners.

Conversation versus narration in pre-school children with language impairment.

The study focuses on two elicitation methods for language sampling in children with language impairment: conversation and narration. It has been noted in other studies on different clinical groups that language elicited in different speaking contexts varies in aspects such as MLU, fluency and syntactic complexity. The purpose of this study was to compare genre effects on different aspects of language production in a group of pre-school children with language impairment. The results show that there are differences in language production during conversation compared with narration. Intelligibility and fluency were found to be higher in conversation than in narration, whereas MLU in words was higher in narration. The narrative task elicited more phrasal expansions and grammatical morphemes per utterance than the conversation. However, the children used more complex verb forms in conversation than in narration. The results are discussed in relation to recent research.