Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Neural mechanisms of visual SL in children with developmental language disorder.
Poster Session C, Thursday, October 1, 10:45 am - 12:45 pm, Wangari Maathai
Hanna Cygan1,2, Martyna Brylka2, Jakub Wojciechowski2, Cristiano Chesi1, Tomasz Wolak2; 1University School for Advanced Studies IUSS Pavia, Italy, 2Institute of Physiology and Pathology of Hearing Warsaw, Poland
Developmental language disorder (DLD) is characterised by impairments in both language production and comprehension. Among factors that may contribute to the emergence of linguistic deficits in DLD, impairments in sequential statistical learning (SL) have been identified frequently. SL is reflected in the ability of children to extract meaningful phonetic and grammatical elements from language streams during development. Importantly, the basic ability to extract probabilistic properties of events in the environment, was related to the functions of cortical and subcortical brain regions. Using a behavioural SL task and functional magnetic resonance imaging (fMRI), we tested SL ability in the visual domain and its neural correlates among children with DLD and their typically developing (TD) peers. We hypothesised that SL neural mechanisms related to visual sequence learning are atypical in DLD for both verbalizable and abstract stimuli. We also examined cognitive abilities, including visuo-spatial working memory, using the Stanford-Binet Scale (SB5) to test for possible interactions with SL. 46 TD children and 40 children with DLD, aged 6;6–9;6 years were included in the examination. During fMRI, children performed SL tasks involving sequences of two types of stimuli: easy-to-name (EN) objects and difficult-to-name (DN) objects. The children underwent a pre-training fMRI, one week of behavioural training, and a post-training fMRI. Groups did not differ in task performance during both experimental sessions. Both groups presented improvement in performance following training in the SL tasks involving both EN and DN objects. The fMRI results showed that, after training, the DLD group was characterised by greater involvement of the frontal cortex and temporal pole for EN objects. Furthermore, in the TD group, the left putamen, globus pallidus (GP), and thalamus were involved in the early stages of SL, whereas in the DLD group, these areas were involved in SL after training. Additionally, analysis of cognitive factors from the SB5, revealed a positive correlation between SL of EN objects and visuo-spatial working memory only in the DLD group. For DN objects, after training, the DLD group presented greater involvement of the parietal and precuneus regions in the SL task performance. Our results suggest that school-aged children with DLD perform equally as TD peers in visual SL. However, they may employ different neural and cognitive processes in SL than TD children, including WM, possibly as a supportive or compensatory mechanism. To further investigate sensitivity for probabilistic content and WM load, in the ongoing experiment, we are testing the interactions of childrens' performance in sentence repetition and comprehension with the probabilistic properties of syntactic elements in relative clauses (RCs) in Italian and Polish. We have designed a self-paced repetition and comprehension experiment of four types of RCs. This allows us to test the children's sensitivity to the probabilistic properties of RCs and WM load related to distance and interference.
Topic Areas: Disorders: Developmental, Language Development/Acquisition