Referential Cohesion in the Narratives of Bilingual and Monolingual Children With Typically Developing Language and With Specific Language Impairment

Abstract

Purpose The study explores referential cohesion in the narratives of bilingual preschool children with typical language development (TLD) and with specific language impairment (SLI). Referential cohesion requires integration of multiple discourse factors and is expected to pose a challenge for children with bilingual SLI due to weak proficiency in both languages. Method Narratives were elicited from 45 bilinguals speaking Russian as the home language (L1) and Hebrew as the societal language (L2; 15 with SLI), 20 Hebrew-speaking monolinguals (10 with SLI), and 20 Russian-speaking monolinguals (10 with SLI) using a story retelling procedure. Bilinguals were tested in both languages. Analyses examined the effect of impairment (SLI vs. TLD) in bilinguals and monolinguals. Language effects were examined in cross-language comparisons of bilinguals (L1 vs. L2) and in differences between monolingual groups (Russian vs. Hebrew speakers) for the use of referential expressions. Results Bilingual children with SLI used a higher proportion of pronouns for character introduction and had fewer pronouns, which have been described as "adequate" ( Colozzo & Whitely, 2014 ) than bilingual children with TLD. No language effect emerged for bilinguals, who performed similarly in their L1 and L2, but a significant cross-linguistic difference emerged in the monolingual data: Russian-speaking children mainly used nouns to introduce and maintain characters, whereas Hebrew-speaking children mainly used pronouns for introduction and maintenance of characters. Conclusion The difficulty of children with SLI in creating a referential connection between a pronoun and a noun phrase is discussed in light of the interaction of local and global processes in narratives, which is shown to be weaker in children with SLI.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 31 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.