The role of implicit learning in L2 morphosyntax: an artificial-grammar language learning experiment

Thesis

There is sustained interest in exploring implicit learning in second language acquisition (SLA). Previously, Williams and his colleagues (2004, 2005, 2009) demonstrated the involvement of implicit learning in grammatical form-meaning mappings. However, Hama and Leow (2010) revisited Williams’ study, finding that participants failed to learn novel form-meaning connections implicitly. Additionally, Williams (2004) touched on the potential relationship between implicit learning and learners’ gendered language background, but his later studies found no evidence for this relationship. In view of these conflicting results, the present study identified several methodological flaws of existing studies and modified the experimental design accordingly to probe into this topic in a more refined manner. The emphasis of this research lies in the role of implicit learning in the acquisition of L2 morphosyntax. Drawing on an artificial-determiner paradigm proposed by Williams (2005), this study worked with forty Chinese native speakers with English as their second language. All instructions and experimental materials were in English. Participants were introduced to four artificial articles: gi, ro, ul, and ne, which were taken as determiners to modify nouns. They were told that the determiner usage depended on whether the modified nouns referred to animals or humans. However, what was unknown to the participants was that gi was used to modify nouns related to predators, while ul was used to modify those associated with prey. Similarly, ro was accompanied by nouns of powerful roles, yet ne was for those denoting powerless roles (i.e. the hidden rule). The present study adopted a mixed method design to investigate the occurrence of implicit learning, L2 learners’ consciousness of the hidden rule, and the relationship between consciousness and L2 learners’ gendered language background. The results suggested that L2 learners could learn the hidden rule of the artificial-determiner system implicitly. Thirty-one of forty participants remained unaware of the hidden rule at the end of the experiment, but their accuracy rate of target items was significantly above the chance level. In general, the aware group outperformed the unaware group in the test task. L2 learners’ consciousness of the hidden rule was found to be unrelated to their prior knowledge of gendered language. These findings supported the involvement of implicit learning in acquiring L2 morphosyntax, especially free grammatical morphemes. It would be advisable if L2 teachers could provide L2 learners with a supportive environment to promote implicit learning.

Authors: Yang, Y

• DOI: 10.5287/ora-6gk8q0q8d
• Type of award: MSc taught course
• Level of award: Masters
• Awarding institution: University of Oxford
• Language: English
• Keywords: implicit learning; mixed method with quantitative and qualitative data; ALSLA; ALSLA2024; artificial grammar; morphosyntax
• Subjects: Quantitative research; Qualitative research; Implicit learning; Second language acquisition; Grammar; Morphosyntax; Applied linguistics