Exploring The Effects Of Frequency And Language Proficiency On The Representation Of Formulaic Sequences In The L2 Mental Lexicon

For a long time, the research focus of formulaic sequences is a kind of description from the linguistic perspective, while little is done in the field of psycholinguistics and second language acquisition. However, recently more and more focus has been shifted to the psycholinguistic perspective, and some monolingual research results show formulaic sequences of high frequency can be stored in the mental lexicon holistically. The present study based on Sosa and MacFarlane's study (2002) is to investigate whether formulaic sequences'frequency and the language proficiency of EFL Chinese learners may affect the representation of formulaic sequences in the L2 mental lexicon.The first hypothesis of the study is that frequency would affect the status of formulaic sequences in the L2 mental lexicon, and high frequency formulaic sequences may be stored in the L2 mental lexicon holistically. Hypothesis Two is that learners'language proficiency would affect formulaic sequences'representation, and holistic representation is much easier to be achieved in the mental lexicon of learners with higher language proficiency level.Two groups of subjects with different levels of language proficiency were chosen. They were asked to complete a reaction time test and a L2 frequency checking task. In the reaction time test, subjects were required to listen for the word of in the 24 two-word formulaic sequences embedded in 24 independent sentences. In addition, the onset time of of in each sentence was obtained by using the speech analysis software"Praat 4304". Thus, based on L1 frequency grouping of the 24 formulaic sequences in addition to the L2 frequency checking results, two one-way ANOVA tests were performed on the mean reaction times among the four frequency categories in the two subject groups respectively, and main significant differences were found. Besides, four independent samples t tests were performed on the mean reaction times between freshmen and postgraduates for each of the four frequency categories, and no significant difference was found. Therefore, Hypothesis One,to a certain degree, is confirmed while Hypothesis Two is not in the present study. Finally, pedagogical implications, limitations of the study and suggestions for further research are discussed.