Alina Kholodova

• This dissertation investigates the developmental trajectory of abstract and lexically dependent syntactic representations in monolingually and bilingually raised children. To examine the nature of these representations, the dissertation employs the structural priming paradigm—the tendency to reproduce previously encountered syntactic structures independent of specific lexical items (Bock, 1986). Structural priming provides a powerful methodological tool for investigating how syntactic knowledge develops and how structural representations are shaped by linguistic experience. The empirical focus of the dissertation is the dative alternation in German, which contrasts double-object constructions (e.g., Dora sent Boots the rabbit) with prepositional-object constructions (e.g., Dora sent the rabbit to Boots). Importantly, the prepositional-object construction is strongly dispreferred in German and rarely occurs in the linguistic input to young children. Evidence for this distribution is established through a corpus analysis of child-directed speech from the CHILDES database as well as an acceptability judgment and elicited production study with adult speakers. Building on these findings, Chapter 5 presents two structural priming experiments examining the production of the German dative alternation in monolingual and bilingual children across three age groups (3–4 years, 5–6 years, and 7–8 years) as well as in adults. The results reveal robust immediate and cumulative structural priming effects across all age groups. However, these effects are strongest in the youngest children and gradually decrease with age. This pattern suggests that syntactic representations in younger children are less stable than in adults and therefore more susceptible to adaptation both immediately and over time, potentially due to stronger surprisal when encountering less expected structures. In contrast, lexical boost effects were absent in the youngest group but emerged gradually with increasing age. This developmental pattern may reflect limitations in younger children’s working-memory capacity, which may constrain their ability to temporarily maintain lexical information that could strengthen priming effects. Alternatively, it may reflect still-developing links between verbs and their associated argument structures in the mental lexicon, which may become more firmly established with increasing age. Taken together, the findings support the Implicit Learning Account of structural priming, according to which syntactic representations are continuously updated through exposure to structural input via prediction-error mechanisms. The results suggest that structural learning mechanisms are particularly active early in development, whereas lexically dependent priming effects emerge only later as children’s cognitive and linguistic systems mature. More broadly, the dissertation contributes to our understanding of how syntactic representations develop in both monolingual and bilingual acquisition and how structural frequency and lexical information interact in shaping children’s syntactic choices.