Language and literacy skills of bilingual and monolingual children with neurodevelopmental disorders

Language and literacy skills of bilingual and monolingual children with neurodevelopmental disorders

First Author: Esther Geva -- Applied Psych &Human Dev, OISE- U of Toronto
Keywords: Developmental neuroscience, Bilingualism, Dyslexia, Autism
Abstract / Summary: 

Geopolitical circumstances mean that various forms of bilingualism are prevalent around the globe. Regardless of the context of bilingualism some bilingual children may have identified or unidentified neuro-developmental disorders, including Autism Spectrum Disorder (ASD), Developmental Language Disorder (DLD), or learning disorders such as dyslexia. The underlying thread of the studies in this international symposium (conducted in Greece/UK, Canada, Norway, the US) is an in-depth examination of specific language and reading skills in children and adolescents with these disorders. The papers contribute to our understanding of how specific components of language, ranging from phoneme discrimination, morphology and syntax, to storytelling and inferencing develop in typical and atypical students, the cognitive processes underling their language and reading skills, and the extent to which these patterns are universal or vary as a function of bilingualism, the specific language involved, or the specific disorder. In addition to their theoretical merit these studies tackle methodological challenges concerning reliable assessment, identification and intervention.

Symposium Papers: 

Narrative abilities of English L2 learners with autism spectrum disorder, developmental language disorder and typical development

First Author/Chair:Krithika Govindarajan -- University of Alberta
Additional authors/chairs: 
Johanne Paradis

Purpose: Narrative skills are associated with reading skills and often used for clinical assessment. We investigated the narrative abilities of English L2 learners with Autism Spectrum Disorder (ASD), Developmental Language Disorder (DLD) and Typical Development (TD). Cross-disorder characteristics can elucidate syndrome-specific difficulties, and inform assessment practices; however, few cross-disorder comparisons have been conducted with bilingual children.

Method: A standardized English narrative test was administered to L2 children with ASD (L2-ASD), with DLD (L2-DLD) and with TD (L2-TD), The participants had diverse L1 backgrounds and were matched for age (mean=6;7), non-verbal intelligence and receptive vocabulary. Narratives were coded for the following macro-and micro-structure sub-skills: story grammar (SG), introducing referents, syntactic complexity, mean length of utterance, lexical diversity and story length in words.

Results: Regression modelling showed that L2-TD included more story grammar units in their narratives, produced longer and more complex utterances, introduced referents more often and accurately than L2-ASD. L2-DLD included more story grammar units for one story, but patterned similarly to L2-ASD for referring expressions, syntactic complexity, utterance length, lexical diversity, story length and individual story grammar units.

Conclusion:  Standardized narrative tasks show promise for differentiating children with TD from children with disorders, when L2 children are the reference group. However, this task did not differentiate adequately between L2-DLD and L2-ASD; these groups had similar narrative skills.  Additional coding categories, such as presence of irrelevant utterances, could better elucidate syndrome-specific characteristics in narrative skills.

Bilingualism effects in the comprehension performance of children with Developmental Language Disorder and Autism Spectrum Disorder

First Author/Chair:Ianthi Maria Tsimpli -- University of Cambridge
Additional authors/chairs: 
Eleni Peristeri

Purpose. The study addresses the contribution of language ability, executive functions (EF) and Theory of Mind (ToM) to the comprehension ability of monolingual and bilingual children with Developmental Language Disorder (DSD) and Autism Spectrum Disorder (ASD).

Method. Participants included two groups of 60, 11-year-old bilingual Albanian-Greek children with DLD and ASD, age-matched to monolingual children with DLD and ASD, and to typically-developing monolingual and bilingual children. Children were tested on reading and listening comprehension (Sideridis & Padeliadu, 2000; Chrysohoou, 2007) in texts of ascending difficulty. Comprehension questions targeted children’s memory for facts, inferencing, non-literal language interpretation and monitoring for inconsistencies. EF and false belief tasks were also administered.

Results. Comprehension was differentially affected across monolingual and bilingual children with DLD and ASD. Monolingual ASD children performed worse than monolingual and bilingual children with DLD in inferencing, while bilingual ASD children outperformed their monolingual peers with ASD in all texts. Bilingualism effects were less robust for children with DLD whose performance differed from monolingual DLD children only in texts of low morphosyntactic complexity. Language ability, EF and ToM contributed more to the comprehension performance of bilingual (vs. monolingual) children.

Conclusions.  Comprehension differences between groups appear to be an effect of their partly distinct language and cognitive control impairments. Monolingual ASD children’s performance reflects deficits in inferencing and non-literal language that could not be compensated by their language ability and intelligence, while comprehension challenges for monolingual and bilingual DLD children mainly stemmed from the lexical and morphosyntactic complexity of the texts.

What characterizes the morphosyntax of Norwegian children with Developmental Language Disorder?

First Author/Chair:Jannicke Karlsen -- University of Oslo

Purpose: Knowledge of the language-specific symptoms of children with developmental language disorder (DLD) in various languages is increasing. However, in some languages such as Norwegian, research has been scarce: This is the first study to examine  broader aspects of morphology and syntax in Norwegian-speaking children with DLD. The study aimed to compare morphological and syntactic structures in the sentence production of Norwegian-speaking children with DLD and controls with typical development (TD).

Method: Participants were 27 children with DLD (5-12 years) and 27 children with TD who were  matched (pairwise) on gender, age and nonverbal IQ. As a theoretical framework and an analytical tool, we used Processability Theory, (PT, Pienemann 1998), a cognitive theory of language development that predicts a hierarchy of stages in the development of certain morphological and syntactic phenomena. We analysed the sentences produced by the children in the Formulated Sentences subtest of the CELF-4, with a focus on lexical morphology, phrasal morphology, sentence structure, and subordination.

Results: The majority of children with DLD (23) performed at lower PT stages than their matched control with TD. Only two children with DLD produced subordinate clauses, compared to 18 of the matched controls. 

Conclusion: Our findings suggest that Norwegian children with DLD do not have problems with morphology per se, but with the processing of morphosyntax in general, including syntax.

Developmental reading disorders in U.S. high school L2 learners?

First Author/Chair:Richard L. Sparks -- University of Cincinnati

Purpose: U.S. high school students studying a L2 comprise a clinical, or special, population because they live in mostly monolingual environments, delay L2 learning until high school, and learn to read and speak/comprehend the L2 simultaneously. Previous studies have found that most U.S. students do not become literate or fluent in a L2. Study examined growth and prediction of their L2 reading comprehension skills over three years of Spanish.

Method: Two hundred and sixty-three monolingual English high school students completed Spanish I and II, and 51 of the participants finished Spanish I, II, and III. All participants had been administered L1 reading and vocabulary measures at the beginning of Spanish I. At the end of each year, subtests from the Woodcock-Muñoz Pruebas de aprovechamiento, a standardized measure of Spanish, were administered: word decoding, reading comprehension, listening comprehension, vocabulary. Latent growth modeling was used to assess growth and prediction of L2 reading comprehension skills.

Results: Individual differences in L2 decoding and L2 listening comprehension skills predicted initial L2 reading comprehension skills, but L2 listening comprehension and L2 vocabulary predicted subsequent growth of L2 reading comprehension skills. All participants exhibited very limited growth in L2 vocabulary over three years of Spanish.

Conclusions: L2 vocabulary appears to be a strong predictor of L2 reading comprehension skills in U.S. L2 learners. However, U. S. learners acquire very little L2 vocabulary, even after 2-3 years. The limitations in vocabulary development are likely to explain their L2 reading disorder. Findings are discussed in the context of the Componential View of Reading. 

A longitudinal study of phoneme discrimination in ELLs and EL1s who are typical readers or at-risk for reading disability

First Author/Chair:Miao Li -- University of Houston
Additional authors/chairs: 
Esther Geva; Fataneh Farnia; Emiko Koyama

Purpose: English Language Learners (ELLs) need to acquire new unfamiliar phoneme contrasts. Not much is known on the longitudinal development of phoneme-discrimination (PD) skills among typically developing (TD) children, and ELLs and monolinguals at-risk for dyslexia (AR-DYS). The study examined overall trajectories of PD in ELLs and monolinguals, and compared these trajectories in TD and AR-DYS readers.

Method: Participants were 200 ELLs from various home language backgrounds, and 156 monolinguals attending public schools in Toronto. PD (same-different judgments of pseudoword pairs) was assessed in the Fall and Spring of grade 1 (Time1 and Time2), and again in the Fall and Spring of grade 2 (Time3 and Time4). AR-DYS status was assigned for those below the 30% percentile on grade 3 phonological awareness, word reading, and pseudoword decoding. 

Results: ELLs performed more poorly than monolinguals at Time1, and both groups improved over the 2 years. The trajectory of the monolinguals was linear, while the trajectory of the ELLs was quadratic: growth was steeper in grade 1, but decelerated in grade 2. The gap between ELL and monolingual TD closed by Time4, but the ELL-AR-DYS did not close the gap by Time4. ROC analyses showed that for monolinguals performance on the PD at Time1 is the strongest predictor for AR-DYS status in Grade 3, whereas for ELLs it is at Time3.

Conclusions: Results underscore the importance of considering phoneme discrimination challenges faced by ELLs, the utility of considering development trajectories, the “double whammy” of ELL- AR-DYS, and the importance of providing early intervention.