| In the light of Baddeley& Hitch's (1974) multi-component working memory model, Kellogg (1996) proposed the model of working memory in writing which is empirically supported by a number of L1 researchers and provides a theoretical and empirical basis for the research of writing cognitive process. Results of related studies indicate that, in L1 condition, written production requires both verbal and visual working memory. Verbal working memory is required in the translating and reading stage of writing, while visual working memory is required in the planning stage of imagery conceptual content (Kellogg, 1999). Research on the implication of working memory in L2 writing is still draggling in China. Chinese scholars still stay in the stage of introducing and reviewing related foreign theories and studies, without performing empirical studies.In view of the insufficient empirical studies of L2 writing cognitive process in China, the present study investigated the implication of working memory in L2 sentence written output, probes into the role that verbal and visual working memory play in Chinese EFL learners'L2 sentence generation, and tentatively explores the effects of working memory components in L2 written production. This study is specifically designed to examine the following questions: the influences that the verbal and the visual working memory impose on L2 sentence generation respectively. We replicated the dual-task methodology from Kellogg et al. (2007). The experiments were implemented among English majors randomly selected from the third grade in Ningbo University and the actual number of participants of the two experiments was both 40.Experiment 1 assessed the effects of visual working memory in L2 written sentence output. The results of this experiment agree with the assertion that visual working memory supports selective processes when imagery representations are aroused in planning stage of concrete language (Kellogg et al., 2007). On the other hand, the present study suggests, when phonological encoding in abstract L2 production is involved in visual working memory, there will be greater competition for the limited resource of visual working memory.Experiment 2 examined the role of verbal working memory in L2 written sentence output. The results indicate that at least one stage of phonologically encoding the conceptual content of a planned message requires the transient storage of phonological representation in verbal working memory. Additionally, compared with L1 condition, the present study shows that the phonological processing plays a more supportive role in L2 written production, which suggests L2 writing process requires greater verbal working memory room for transient storage of conceptual content representations.In general, the results of the two experiments are consistent with the results obtained in L1 condition. Accordingly, this study reveals the role of working memory components in L2 written production, i.e., verbal and visual working memory support specific processes in L2 production. Moreover, compared with L1 condition, verbal working memory is evidently more overloaded in L2 written production, and aural processing of phonological information is more demanding than visual processing. The research findings amplify the current theories of L2 writing cognitive psychology. The dual-task paradigm adopted in the present study enriches the research methods in foreign language teaching research. In light of these findings, two pedagogical implications for L2 writing teaching are suggested: In L2 writing teaching, language teacher should pay more attention to learners'writing process and instruct them according to individual writing sub-processes; Literature indicates working memory is a crucial factor in such cognitive activities as writing etc., and working memory ability can be improved through training, therefore teachers can help learners train their working memory through appropriate ways to improve their L2 writing ability.
|
PDF Full Text Request