Doctoral Dissertation Research: The Effects of Language Experience on Statistical Learning in Infants and Adults

University Of Southern California, Los Angeles CA

Investigators

Abstract

This doctoral dissertation project examines how the particular language a person knows influences how they learn an additional language. The project also examines whether knowing more than one language affects how a person learns an additional language. To address these questions, the researchers carry out experiments with infants who are exposed to only one language, infants who are bilingually exposed to two languages, and with monolingual English-speaking college students who are in the process of learning a second language. The experiments for all the participants involve testing their ability to learn artificial languages that have been designed to have important similarities to real human languages. By investigating what properties of the artificial languages are easier or harder to learn for various groups of participants, the researchers understand how a person's overall experience with languages influences their learning of new ones. The results of this project generate insights into the process of normal language learning that can be used to design language interventions for people with language learning difficulties. This project provides training opportunities for undergraduate students, to gain expertise in psycholinguistic research. The language properties we investigate involve dependencies between words that arise because of the grammatical rules of the language, in particular non-adjacent dependencies (NADs) that involve words that are not immediately adjacent. For example, in typical sentences in English, the form of the verb and the following noun depend on each other, but can be separated. Discovering these dependency patterns is essential for language acquisition. While recent research suggests that learners’ prior experience within a laboratory testing session shapes their expectations and influences the types of patterns they subsequently track, there has been little attention to the influence of learners’ broader language experience on subsequent learning. This project uses behavioral measures to examine the interactions between someone’s prior language experience and the ability to learn and generalize from dependency patterns in a new language. The first experiment compares the performances of monolingual and bilingual 12-month-old infants learning different languages on a NAD learning task. It tests two hypotheses: 1) that the structures in 12-month-old infants’ exposure to language influence their sensitivity to NADs, and 2) that their bilingual status influences this sensitivity. The second experiment compares the learning of different dependency patterns in native English speakers learning two different languages, and monolingual English speakers. In this experiment, the researchers test the effects of learning one language on the ease of learning an additional language, examining how the particular language that is learned make certain properties of the additional language easier to learn. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.