Building Morphosyntactic Networks in Preschoolers With and Without Developmental Language Disorder

About This Grant

Project Summary/Abstract Difficulty using morphemes of tense and agreement (e.g., The girl sits, The dog barked) during the preschool years is a hallmark symptom of developmental language disorder (DLD). Children with DLD omit these mor- phemes more often and for a longer period of time than children with typical language development (TLD). The mechanistic factors driving the difficulty in acquiring consistent use of these morphemes in DLD are not well understood. Network science is a complex systems approach that models how simple relationships (known as edges) between items (known as nodes) in a network give rise to complex patterns of behavior. Network ap- proaches to language have indicated that the network structure of different language sub-systems (semantics, phonology, syntax) explains variability in language skills that traditional language measures cannot. Although difficulty using grammatical morphemes during the preschool years is a key indicator of DLD, network science has not yet been applied to examine the link between structural representations of morphosyntax and produc- tive morphosyntactic abilities. To address this gap, the current objective is to build morphosyntactic networks of preschool-aged children with DLD and TLD that explain differences in morphosyntactic productivity. The cen- tral hypothesis is that the networks of children with stronger language skills will have structural signatures driven primarily by linguistic skills that support morphosyntactic development. First, morphosyntactic networks (i.e., networks that model grammatical relationships between words and bound morphemes such as -s and - ed) will be built from spontaneous speech of preschoolers with TLD and DLD to determine structural properties of a group morphosyntactic network (Aim 1). The hypothesis is that the morphosyntactic network will have high connectivity like other networks of natural language systems. Next, regression analysis will be used to examine how within-group (TLD and DLD) individual differences in language skills known to support the development of morphosyntax predict the structure of the morphosyntactic network (Aim 2). The hypothesis is that the predic- tors of structural differences in the morphosyntactic networks will be different between TLD and DLD given that morphosyntax is a key area of difficulty in DLD during the preschool years. Whether the hypotheses are sup- ported or refuted, this work will advance the field’s understanding of how differences in productive morphosyn- tax in healthy and disordered language development may be related to the underlying structure of these repre- sentations. This work will advance the field by establishing a framework by which to study morphosyntax in a comprehensive way that fully considers the complexity of tense and agreement and how it develops. Addition- ally, the technical and professional research skills I will acquire through completion of these aims will enhance my clinical SLP training to position me for an impactful career as an independent language scientist program- matically focused on a mechanist approach to language acquisition and treatment in cases of disorder.