| Popis: |
The status of human language as either a recycler of general cognition (Tomasello, 2003; Bybee, 2010; Goldberg, 2006) or an encapsulated module with processes unique to itself (Chomsky, 1980; Fodor, 1983), has been intensely contested. The debate is important because it speaks to what language is made of and how it is acquired. Although evidence is building of a significant interaction between language and cognition (e.g., Dąbrowska & Divjak, 2015, 2019), the extent of this interaction, the development periods in which it occurs, and the domains to which it corresponds, remain unclear. This thesis contributes to the understanding of these issues with four studies. Study 1 explores the directionality of cognitive transfer between working memory (WM) and syntax among L1 and L2-learning 7-year-old children in a randomised controlled trial. Study 2 analyses individual variation in WM performance among these same participants by examining speed/accuracy trade-offs in the context of gender. Using corpus data, study 3 correlates the development of English and Spanish long-distance subject-verb agreement (SVA) with that of non-verbal WM in children between 2 and 12 years of age. Study 4 examines the extent to which WM is implicated in processing long-distance SVA given the level of formulaic language used. The studies revealed that: [1] WM training improved L1 and L2 syntactic performance, but not vice versa; [2] the reaction times to the WM task were bimodally distributed, and boys tended to respond quicker but girls were more accurate; [3] the development of long-distance SVA was predicted by the development of non-verbal WM; [4] the degree of formulaicity observed in sentences with long-distance agreement was consistent with some form of WM involvement. Interpreted together, these results present a compelling and comprehensive case in favour of a deep integration between WM and syntax in the process of language acquisition. |
