Search Abstracts | Symposia | Slide Sessions | Poster Sessions
Understanding the role of the right hemisphere for language processing in autism – A normative modeling approach
Poster Session E, Friday, October 2, 11:00 am - 1:00 pm, Wangari Maathai
Marjolein Mues1,2, Sabine Hunnius2, Jan Buitelaar1,2, Christian Beckmann1,2; 1Radboud University Medical Center, 2Donders Institute for Brain, Cognition, and Behavior, Radboud University
Language abilities of autistic children are highly heterogeneous: some autistic children are non-speaking while others develop above average language abilities. Difficulties with language and communication have debilitating effects on autistic children’s developmental outcomes and quality of life and a better understanding of the basis of language development in autism has crucial translational and clinical importance. It is not clear why some autistic children develop language difficulties and others do not, but preliminary studies show that the underlying neurobiology may play a role. While language is typically left-lateralized, some autistic children show stronger right-hemispheric involvement, leading to a less-lateralized neural pattern during language processing. Researchers hypothesize that the right hemisphere may either be an underlying cause of language deficits in autism or that a right hemispheric network may be compensatory in nature, facilitating an alternative route for successful language development. However, previous studies have so far mostly examined underlying neurobiology in isolation and not in the context of phenotypical language skill while this is imperative to better understand the role of the right hemisphere for language processing in autism. In this ongoing study, we will further examine these hypotheses. In this project, we will create normative models of neurobiology underlying language processing in the left and right hemisphere using structural, diffusion and resting-state functional MRI in typically developing children using existing neuroimaging data from multiple Open Access datasets (i.e., Human Brain Connectome, ABIDE, etc.). Data of autistic children has been previously collected in the Preschool Brain Imaging and Behavior Project (PIP), part of the international AIMS-2-TRIALS consortium. Neuroimaging data of autistic preschoolers of the five data acquisition sites of PIP (Radboud University, The Netherlands; King’s College London, UK; Ghent University, Belgium; Karolinska Institute, Sweden and Assistance Publique-Hôpitaux de Paris, France) will be compared to the normative curves of typically developing young children to examine how far removed each autistic individual is from the curve. This “distance score” will then be related to standardized measures of language that were collected. We will outline our normative modeling approach and the variables that we will include in this presentation. We will follow the dual stream hypothesis for language processing and examine both dorsal and ventral variables, including functional connectivity within dorsal and ventral networks, fractional anisotropy in the arcuate fasciculus (dorsal) and the inferior fronto-occipital fasciculus (ventral) and volume of the superior temporal gyrus (dorsal) and medial temporal gyrus (ventral). Using a normative modeling approach, we aim to overcome current overreliance on single measures of either brain function or structure and group-level analyses that are insufficient to characterize the great heterogeneity across the autism spectrum. Our goal is to better understand atypical language performance in autistic children and clarify the effects of a potential right-hemispheric network for language in autism.
Topic Areas: Disorders: Developmental, Language Development/Acquisition