Sentence comprehension in hearing-impaired and typically-developing children: Linking between morphosyntax and semantics

ANALÍ TABOH; CAROLINA GATTEI; DIEGO SHALOM

Moscú

Otro; 7th Summer Neurolinguistics School; 2020

Center for Language and Brain, National Research University Higher School of Economics

BackgroundUnderstanding who does what to whom in semantically reversible sentences requires integrating semantic information provided by the lexico-semantic structure of the verb with morphosyntactic information such as word order and morphological case marking. Hearing-impaired children who access spoken language through cochlear implants or hearing aids (HIC) often have difficulties comprehending syntactically complex sentences such as those with non-canonical word order (Friedmann & Szterman, 2006; Ruigendijk & Friedmann, 2017). They often omit case-marking morphology or use it incorrectly as well (Szagun, 2000; Svirsky et al., 2002). This cross-sectional work looked into how these children establish the linking between morphosyntax and semantics in comprehension, in order to assign the thematic roles to the arguments correctly (Bornkessel & Schlesewsky, 2006b; Van Valin, 2005). We aimed at (i) evaluating whether and how the comprehension patterns are different between HIC and typically-developing children (TDC); and (ii) analyzing the acquisition curve of this linguistic ability in HIC and TDC.MethodWe compared the performance of Spanish-speaking HIC (N=32) and TDC (N=13) with 3 to 6 years of hearing age (i.e. time of exposure to oral language) in a sentence comprehension task with psychological verbs. In Spanish, linking between morphosyntax and semantics may differ according to the type of psych verb. While Subject Experiencer Psych Verbs (SubjExp) present direct linking, i.e., the most salient thematic role (Experiencer) coincides with the most salient syntactic constituent (subject), Object Experiencer Psych Verbs (ObjExp) impose indirect linking (Belletti & Rizzi, 1988). Furthermore, while the canonical word order for SubjExp is SVO, as is usual in Spanish, the canonical order for ObjExp is OVS. Indeed, evidence from studies with adults shows that comprehension is enhanced when the sentence follows the same word order as the lexico-semantic structure of the verb. In the current study, we used the truth-value judgment paradigm (Crain & Thornton, 1998) and measured accuracy rates according to two independent variables: word order (SVO/OVS) and linking type (direct/indirect). ResultsVisual inspection of the data shows that, at hearing age 3-4, comprehension in both groups of children is facilitated when both the syntactic and the semantic orders coincide with the canonical structure, which is only the case in SubjExp SVO sentences. A difference between groups appears at age 4-5: while HIC perform at chance level, TDC show better performance when word order reflects the lexico-semantic structure of the verb (SVO for SubjExp, OVS for ObjExp). The performance of HIC only starts resembling the TDC comprehension pattern at age 5-6, even though accuracy is still lower.Since data collection is still on-going and had to be interrupted due to the COVID-19 pandemic, we lack definitive statistical analyses. However, preliminary statistical analyses show a significant interaction between word order and verb type at age 5-6 (p