论文总字数：45501字

摘 要

1. Introduction 1

2. Literature review 3

2.1 Theoretical issues 3

2.1.1 Levelt’s Model of Oral Production 3

2.1.2 Working Memory 4

2.1.3 Planning 4

2.1.4 The interplay of working memory and planning time on oral production 5

2.2 Empirical studies 5

2.2.1 Studies on the effects of working memory on oral production 5

2.2.2 Studies on the effects of planning on oral production 6

2.2.3 Studies on the effect of WM and planning on oral production 6

2.2.4 Limitations of previous studies 7

3. Methodology 8

3.1 Research questions 8

3.2 Participants 8

3.3 Data collection 8

3.3.1 Computerized reading span test 8

3.3.2 Picture-narrative task 9

3.4 Procedure 9

3.5 Data analysis 10

3.5.1 working memory 10

3.5.2 accuracy 10

4. Results and discussion 12

4.1 Results of Two-Way ANOVA 12

4.1.1 Results of ANOVA of accuracy in terms of the percentage of correct verb form 12

4.1.2 Results of ANOVA of accuracy in terms of the percentage of error-free clause 13

4.2 Discussion 14

4.2.1 Effects of WM on accuracy 15

4.2.2 Effects of planning time on accuracy 15

4.3 Effects of WM and planning time on accuracy 15

5. Conclusion 17

5.1 Major findings of the study 17

5.2 Theoretical implications 17

5.3 Pedagogical implications 17

5.4 Limitations and suggestions for future research 18

References 19

Appendix 1 21

Appendix 2 22

Acknowledgements

My heartfelt thanks firstly go to Professor Han Yawen, my supervisor, who has been paying close attention to my study in the course of my thesis formulation. For me, he has been more than a supervisor. Here I want to express my deepest gratitude for his insightful criticism, valuable suggestions and continuous support during the progress of thesis writing, without which this dissertation would not be possible. Without his enlightenment and encouragement, I would not have finished this thesis in such a smooth way.

My appreciation also goes to all my classmates and friends. I enjoyed their company, their brilliancy and their friendship. They offered generous help by searching valuable references and information for me when this thesis was still in the conceptual phase. I know I could learn a lot from them through my four-year campus life.

All the students in Class English 1301 and English 1302 at Nanjing Tech University have helped me a lot in the process of the investigating survey of the thesis. Their participating in the tests provided me with the first-hand information, which is the indispensable foundation of my thesis. Without their involvement, I cannot even finish my work. So thanks a lot for their valuable contributions.

Finally, I would like to extend my thanks to and deepest respect for my family and my best friends, who have been constantly concerned with the progress. It is their encouragement that gave me spiritual strength and confidence to overcome all the difficulties arising in the paper and other aspects of my life.

Abstract

This study investigates the effect of working memory capacity (hereafter WMC) and planning time on China’s EFL learners’ oral accuracy. A total of 51 English major sophomores in a university of Technology in China were tested. Working memory was measured by the reading span test. Then follows a picture-narrative oral task in English under the conditions of with or without planning time. Two factors were found to increase the accuracy of learners’ oral production – planning time and WMC; significant interaction between WMC and planning for correct verb form was also revealed by the research. The researcher discussed the possibility of alleviating the burden of working memory by adjusting task complexity.

Key words: working memory capacity, planning time, oral accuracy, oral production

中文摘要

本研究考察了工作记忆容量和准备水平对中国英语学习者口语准确度的影响。中国某工业大学51名大二英语专业学生参加了本次实验。研究者通过电脑版阅读广度测试对受试者的工作记忆容量进行测量，并在有准备时间和无准备时间情况下，让受试者完成英语图片描述口语测试。通过双因素方差分析，研究者发现，工作记忆容量和准备时间对中国英语学习口语准确度均有有显著效应，工作记忆容量和准备水平对动词的准确形式存在显著交互效应。研究者讨论了通过调整任务复杂度削减工作记忆容量负荷的可能性。

关键词：工作记忆容量 准备水平 口语流利度 口语产出

1. Introduction

There has been a growing interest in how psycholinguistic factors such as working memory (hereafter WM) may affect an EFL learner’s competence of speech. Researches focusing on learner-internal cognitive factors relating to speech have suggested that WM capacity can affect L2 speech production (Kormos, 2006). Oral accuracy, as one of the most important criteria, has been frequently adopted to measure EFL learners’ speech performance. This article addresses the question of how far working memory and planning time may influence China’s EFL learners’ improvement in spoken language.

A variety of studies have been carried out to find out the connection between planning time, WM and the oral performance of EFL learners. The significance of working memory capacity is originally suggested by Baddeley and later on received great emphasis in a lot of studies concerning oral competence. In spite of the attention in working memory, the possible contribution of planning time and WM to oral performance cited above has not been addressed in any detail.

Findings from a variety of studies support the hypothesis that excellent working memory capacity would positively affect competence of speech. Such researches calls for studies that can go further and investigate more of the possible affecting factors of oral performance and, more precisely, oral accuracy. However, the understanding of how exactly adequate planning time may contribute to speakers’ performance remains unclear. Moreover, the interaction effect of planning time and working memory has not been investigated by any of the oral-training-relevant researches. This is probably due to the assumption that an advanced English speaker, regardless of how good they are at memorizing, must be prepared to talk whenever and wherever possible, even within pretty little time for preparation. This gap that the author addresses is believed to be important and significant to warrant attention; hopefully the result of the present study would guide EFL learners in how to most effectively use their time.

In the past several decades, a lot of studies have been carried out to investigate the effect on accuracy posed by planning time and working memory capacity separately, however much fewer studies have been conducted on the interrelationship between planning time and WM capacity. Specifically, a lot of researchers have found out that quantitative and qualitative analyses of oral accuracy play a critical role in the guidance of oral teaching, but there are fewer findings yielded in analysis of oral accuracy at different WM capacity levels and with different pre-task planning time given, which leads to the lack of pointed measures to different groups.

To solve the problem, we try every possible way to find out the relations between WM capacity and planning time and the oral performance of EFL learners, as well as possible ways to improve China’s English teaching standards. In the present work, the author intends to focus on how planning time and WMC may affect interpreters’ performance. More precisely, the author will try to find out whether there is a significant difference in oral accuracy among EFL learners at different working memory capacity and subjects given different planning time.

2. Literature review

Nowadays improving students` oral communication skills has been the primary task in China’s college English teaching. In this study the oral accuracy among EFL learners at different working memory levels and planning conditions will be analyzed. And in this chapter, literature is reviewed in two major parts with four sub-sections in each, first on the theoretical foundation based on which this research is conducted, second on empirical studies, both involving three variants in the present study, i.e., oral production (or oral accuracy), working memory, and planning. In addition, before the end of this chapter, the author conducted a brief discussion about the limitations in previous studies.

2.1 Theoretical issues

2.1.1 Levelt’s Model of Oral Production

Basically all studies concerning oral production or the notion of planning draw on the model of speech production proposed by Levelt (1989) as a reference. According to this model, the translation of speakers’ thoughts and intentions (or the production of spoken language) involves three major stages of processing:

• Conceptualization: in which the link between the desired concepts that are to be verbally expressed and the particular words are built and thus the preverbal intended messages are created;
• Formulation: in which the linguistic forms based on the preverbal messages are created and transformed into the expression of intended messages. This stage involves three sub-sections: grammatical encoding, morpho-phonological encoding, and phonetic encoding;
• Articulation: in which the actual speech is formulated through the involvement of different parts of vocal apparatus.

In addition to these three stages, there is another stage known as “self-monitoring”, through which the speakers test whether the overt speech is in accordance with the desired message and thus have a self-assessment of their utterance production.

According to the speech production model, planning could take place either prior to task performance or while performing a task (Ellis, 2005). This model and the related studies conducted by Levelt have shed light on both other studies concerning oral production and the present study. Though the speech production has been initially proposed to explain the oral production of native speakers, it has later on served as an explanation and reference of L2 oral production. Given that the oral production of native speakers (or L1 speakers) is automatic while that of L2 speakers is not, this model is more an explanation to L2 speech production from the perspective of psycholinguistic.

2.1.2 Working Memory

Working memory research has commonly been based on Baddeley’s (1986) influential multicomponent model of WM, in which WM has been devided into two categories: short-term storage and executive processing capacity. Ashcraft (1994) has introduced a theory that human beings possess a “multicomponent memory system” that consists of at least “three standard systems: sensory, short-term memory, and long-term memory” (Atkinson amp; Shiffrin, 1968).

Definitions of the term “working memory” can be found in a considerable body of related researches. It refers to the limited capacity cognitive system responsible for both storage and processing of information during the performance of complex cognitive tasks (McLaughlin; Heredia, 1996; Daneman; Carpenter, 1980, 1983; Baddeley, 1992a). It is also called “short-term store” or short-term memory, primary memory, immediate memory and refers to the temporary processing and storage of information. According to the findings of some studies in L2 oral production, greater WM capacity tends to facilitate speech formulation, or in other words, help produce more fluent, more accurate speech (Fortkamp,1999; Kormos, 2006).

2.1.3 Planning

Planning has been a widely tracked construct within the study of L2 tasks. It can be inferred from previous studies that planning has developed into an area of inquiry and became “a burgeoning area of investigation with task-based learning” (Ortega, 2005). From the perspective of psychology, planning is one of the executive functions of the brain. Some certain parts of the brain have turned out to be playing an intrinsic role in cognitive planning and even some associated executive traits such as working memory.

In L2 studies and task-based studies, planning is conceptualized as the opportunity to prepare and design your task performance before the actual performance. Simply put, it is the cognitive process of thinking about or organizing what you will do in the event of something happening. This process involves both preverbal automation (Levelt, 1977; Baars, 1980) and long-term organization of utterance.

In general, it is a mental activity in which human-beings allocate their attention proportionally and regulate their cognitive process (Yuan, 2001). When associated with L2 development, it is expected to redirect learners’ attention to a certain focus and therefore improve oral production (Skehan, 1996).

2.1.4 The interplay of working memory and planning time on oral production

Few scholars have studied the interplay of working memory and planning on oral production. The possible facilitative role of planning time has been proposed by Nielson (2013), in whose study, however, no effect has been shown for the interaction of working memory capacity with pre-task planning time.

In the present study, the author intends to study accuracy as an indicator of oral production. Accuracy means the extent to which the information outputted agrees with the goal that is supposed to be achieved in the target language (Yuan amp; Ellis, 2003:2). Compared with the rich researches on oral fluency or oral production as a whole, studies on oral accuracy are relatively scanty and most traditional studies focused on the written language.

Investigating accuracy is mostly limited in the grammatical level and accuracy means the output is in accordance with the grammar of the target language. Foster amp; Skehan (1996) have defined accuracy as “the similarity of the language forms to the target language”. Ortega (1999) still uses the grammatical accuracy to judge the morphological coherence between the noun and adjunct. Foster, Tohkyn amp; Wigglesworth (2000) believe that accuracy still lies in the grammatical level.

2.2 Empirical studies

2.2.1 Studies on the effects of working memory on oral production

According to Baddeley, WM is the capacity-limited cognitive mechanism that “enables us to temporarily maintain several pieces of information in mind while comprehending, thinking, speaking, and doing”. We can reasonably infer that WM plays a quite significant, positive role in cognitive activities and thus leads to differences between EFL speakers with high and low WMC. Research findings listed and discussed about above have lend support to this hypothesis, i.e., individual difference on WM capacity correlate with how much the EFL learners would benefit from the time they are given to prepare for their speech and, in doing so, produce more accurate language.

Borges-Mota (2003), after testing 13 advanced EFL learners by a picture description task and a narrative task, finds in her research correlation between WMC and oral performance (or in her study, fluency, accuracy, and complexity). In addition, the speaking span test adopted in the study for measuring WMC reveals that WMC can accurately predict the accuracy, complexity, and fluency of participants’ oral production.

Fortcamp (1999) draws on Daneman’s (1991) study and applys a set of seven experiments, taking 16 advanced EFL learners as subjects. In his study, he conducts both the speaking span test and the reading span test. The results of the research confirms his conjecture that WMC is functional, “varying according to the individual’s efficiency in the processes specific to the cognitive task with which it is correlated”.

Daneman (1991) assesses working memory by speaking span test and reading span test in order to find out whether working memory can be a predictor of speakers’ verbal fluency. The results of his study indicates that reading span correlates with oral fluency on both the speech and reading tasks, while reading span was only correlated with individual differences in oral fluency on the reading task.

In general, the results of the former studies have been contradictory, which is to say, whether working memory has a positive influence on oral production still remains unclear. Possible explanations could be that (1) some researchers adopt reading span test while some other adopt speaking span test, (2) subjects of the tests come from different background and can hardly perform consistently, and (3) the numbers of participants are far from sufficient to be representative.

2.2.2 Studies on the effects of planning on oral production

Overall, the previous studies have shown a positive impact of planning on oral performance. Several studies have shown that planning leads to gains in fluency (Foster; Skehan, 1996; Mehnert, 1998; Ortega, 1999). Planning has also led to gains in accuracy, although results have been more mixed in this respect (Ellis, 1987; Mehnert, 1998; Ortega, 1999; Skehan; Foster, 1999). It has also been shown in some studies that sometimes, higher fluency and complexity come with lower accuracy (Mehnert, 1998). In addition to this, it has been found that planning time contributes to accuracy according to different planning conditions (Mehnert, 1998) and different task types (Foster; Skehan, 1996).

Yuan and Ellis (2004) find that whereas planning would lead to a gain in fluency and syntactic variety, an engagement in unpressured online planning resulted in greater accuracy, due to the reason that participants without planning would more or less be distracted by the need to formulate and monitor under pressure, which accounts for less satisfying writing performance in terms of accuracy, complexity, and fluency.

请支付后下载全文，论文总字数：45501字