6-month-olds' segmentation and representation of morphologically complex words
說明
142 p
附註
Source: Dissertation Abstracts International, Volume: 76-10(E), Section: A
Adviser: Megha Sundara
Thesis (Ph.D.)--University of California, Los Angeles, 2015
One of the issues in infants' language acquisition is - how do infants find word-like forms from fluent speech. Previous literature on infants word segmentation has mostly focused on understanding the bottom-up cues, i.e., cues in the input such as acoustic/prosodic cues, that infants utilize in pulling out nouns. This dissertation asks whether infants can use top-down cues in pulling out verbs. Verb segmentation has been reported to be delayed as compared to noun segmentation and these results have been used to explain the delay in its acquisition of verbs. This dissertation argues otherwise, demonstrating that in fact at the beginning of word segmentation, i.e., at 6-months, infants can pull out verbs with the help of a known word mommy (a paradigm used in Bortfeld, Morgan, Golinkoff, & Rathbun, 2005)
The current dissertation goes further and asks how these verbs are represented. To be specific, this dissertation looks at 6-month-olds' segmentation of morphologically complex verbs, such as walking, walks, and walked, and asks whether preverbal infants can relate these forms to the root form walk. The main focus of this research is to understand how prelexical infants, who cannot rely on semantics, relate complex forms to the root forms
This dissertation expands our understanding of the role of the functional morphemes (such as -ing, -ed, -s) in this process by conducting a corpus analysis as well as behavioral experiments. In this dissertation, I locate the beginning stage of this complex form acquisition and show that at 6-months, infants start segmenting complex verbs, and based on the frequency and the characteristics of the functional morphemes, infants begin to relate complex forms to root forms. The findings of this dissertation highlight the importance of top-down cues in early language development and have crucial implications for verb acquisition. Also, these results provide evidence for morpheme-based processing models and acquisition models such as prosody-functor models, arguing for early representation of functional elements and their facilitatory influence on word segmentation and representation
