Article contents
STATISTICAL SENSITIVITY, COGNITIVE APTITUDES, AND PROCESSING OF COLLOCATIONS
Published online by Cambridge University Press: 24 July 2018
Abstract
Frequency and contingency (i.e., co-occurrence probability of words in multiword sequences [MWS]) are two driving forces of language acquisition and processing. Previous research has demonstrated that L1 and advanced L2 speakers are sensitive to phrasal frequency and contingency when processing larger-than-word units. However, it remains unclear whether such statistical sensitivity is robust across tasks and among subcategories of MWS. In addition, little is known about whether cognitive aptitudes can moderate such sensitivity. This study examined L1 and advanced L2 speakers’ statistical sensitivity to phrasal frequency and contingency as well as cognitive aptitudes’ moderating effects on such sensitivity when processing English adjective-noun collocations. Participants performed a phrasal acceptability judgment task (PJT). Meanwhile, their aptitude profiles were measured by six aptitude tests, which loaded separately onto implicit language aptitude, explicit language aptitude, and working memory capacity. Linear mixed-effects modeling revealed that both L1 and L2 English speakers were sensitive to phrasal frequency and contingency of collocations, although L2 speakers’ sensitivity was much stronger than that of L1 speakers. None of the aptitudes was found to moderate language users’ statistical sensitivity to either collocation frequency or contingency. Interestingly, disassociation patterns between the PJT performance and the involvement of implicit or explicit language aptitude among the L1 and L2 speakers were found. It was concluded that L1 and L2 speakers differed in terms of the way they processed the collocations, as well as the nature of their collocational knowledge.
- Type
- Research Article
- Information
- Copyright
- Copyright © Cambridge University Press 2018
Footnotes
This article is based on the author’s qualifying paper, supported by the PhD program in Second Language Acquisition at the University of Maryland, College Park. I would like to express my gratitude to my supervisor Dr. Michael Long for his tremendous support for this project. I would also like to thank Drs. Nan Jiang, Robert DeKeyser, and Steven Ross for their insightful suggestions that greatly improved the research. I am immensely grateful to Dr. Gisela Granena for her advice on the use of the LLAMA test, and to Dr. Scott Kaufman for giving me the access to the serial reaction time task. The data collection was generously supported by Lars Bokander, Anxin Bai, and Wenbo Li, and I would like to thank them. I also thank Dr. Stefano Rastelli for his constructive comments, as well as Dr. Michael Long, Nicco Cooper, Jason Struck, and Zhiyuan Deng for their proofreading of the manuscript.
The experiment in this article earned an Open Materials badge for transparent practices. The materials are available at https://www.iris-database.org/iris/app/home/detail?id=york%3a934519&ref=search.
References
REFERENCES
- 29
- Cited by