This dissertation investigates the interaction of developing extralinguistic cognitive systems with early language learning and processing through the case study of verb argument structure. The interaction of these systems with the linguistic system underpins fundamental theories of language learning and use: language does not exist in isolation. The extralinguistic cognitive systems sup- porting language learning and use are themselves in their earliest stages of development, potentially placing perceptual and computational constraints on learning. At later stages of development, when children have already learned a significant amount about their target language(s), the use of this acquired knowledge can further be influenced and constrained by still-developing cognitive systems.

We approach this complex interaction through two case studies examining the role of extralinguistic cognitive systems in verb argument structure (i) acquisition and (ii) processing. First, we examine how perceptual systems like visual working memory, which are limited throughout development, can support the representation of high-adicity event concepts. In two behavioral studies, we investigate whether the visual working memory system of adults and preschool-aged children is capable of yielding a 4-participant event concept. We show that adults are capable of represent- ing a trading scene under a single, 4-place concept, despite the typical limitations of the visual working memory system. We also show that preschool-aged children are capable of representing all 4 characters in the same trading scene as participants, raising questions of how young learners may map between their conceptual and linguistic representations. Second, we computationally investigate how immature extralinguistic cognitive systems, such as working memory and cognitive control, can impact the deployment of verb argument structure knowledge. To do so, we develop and implement a generalized left-corner parser with independent parameters correspond- ing to these systems. We successfully model parsing performance in a well-known example where children’s parsing differs from that of adults’, generating testable empirical predictions that could further adjudicate between these two systems.

By investigating independently developing cognitive systems through the lens of verb argument structure, this dissertation explores two ways that these systems can influence verb learning and use. In doing so, we provide an example of how to isolate the specific contributions of extralinguistic cognition to the acquisition and deployment of one type of linguistic knowledge, which can generalize beyond the realm of verb argument structure to other aspects of language development and acquisition.