Habilidades narrativas e Transtorno de Desenvolvimento da Linguagem (TDL): uma avaliação de crianças com comprometimento sintático à luz da teoria gerativa

Abstract

This dissertation focuses on the narrative skills of individuals at risk of Developmental Language Disorder (DLD), who exhibit syntactic impairment. It aims to verify the extent to which a correlation can be drawn between their linguistic, narrative and cognitive skills, particularly inhibitory control. DLD is a disorder that affects children during language acquisition, causing greater costs and a characteristic delay. The literature in the area has observed that children with DLD have difficulty, both in comprehension and production, with a group of so-called high-cost sentences, including passives, WH-interrogatives and relatives, as well with morphological issues. Based on the theoretical model of language proposed by the generative theory (Chomsky, 1965), updated in the Minimalist Program (PM) (Chomsky, 1995), and considering the procedural theory of language acquisition, proposed by Corrêa (2009), issues concerning linguistic processing and computation are assessed. MABILIN I (Language Skills Assessment Modules) (Corrêa, 2000) is adopted for assessing linguistic skills and identify possible children at risk of DLD. The MAIN (Multilingual Assessment Instrument for Narratives) (Gagarina et al., 2012) is adopted as an instrument for evaluating the narrative discourse of monolingual and multilingual children with or without language problems. Two tests assessing inhibitory control are applied: Flanker and GO–no go, in a version adapted by Rodrigues et al. (2024). Two groups participated in this evaluation: 12 children at risk of DLD, and 12 children without language difficulties. MAIN evaluates production, considering elements of macrostructure, microstructure, completeness of presented events and use of mental state terms, and comprehension. Two out of the four stories proposed by the instrument were used, for telling and retelling formats. Our results indicate that, the TDL group showed worse performance in macrostructure evaluation in production, but there was no significant difference in comprehension. In microstructure, the DLD group used a smaller number of complex sentences, had more disfluencies, a greater need for observer intervention and a significantly higher use of ungrammatical and incomplete sentences. In the inhibitory control tests, there was no statistically significant differences. Nevertheless, the DLD group achieved lower averages in the Flanker test. In terms of correlations, there were moderate correlations between the MABILIN and the cognitive tests, as well as between the MAIN and those tests. Positive correlations between MABILIN and the MAIN were also obtained, considering both groups together, but these correlations were weaker when only the TDL group is observed, which is probably due to the fact that these children have a worse performance in general. Nevertheless, the correlation obtained between MABILIN and MAIN confirms the effectiveness of this instrument for evaluating children with DLD, focusing on narrative skills