1. 研究目的与意义（文献综述包含参考文献）

Effects of the Working Memory Capacity and Language Proficiency on L2 Learners#8217; Listening Comprehension 工作记忆容量和语言水平对二语习得中听力理解的影响 1. Introduction For decades, researchers and practitioners alike have been interested in predicting which learners are likely to succeed in acquiring a foreign language (L2). A variety of variables have been proposed and explored empirically (for a review, see Dornyei, 2006), with the goal of identifying separable components of language learning ability#8212;that is, language aptitude (Dornyei, 2005; Skehan, 2002). Working memory (WM) has been identified as being likely contributors to language aptitude (e.g., DeKeyser amp; Koeth, 2011; Hummel, 2009; Wen amp; Skehan, 2011). Listening is a key skill in second or foreign languages (L2/ FL); it plays a critical role in communication and in second language acquisition (Anderson and Lynch, 1988; Rost, 1990). Rost (1994) pointed out that listening is vital in the language classroom because it provides input for the learners. Therefore, exploring L2 learners#8217; listening comprehension is of much concern. A number of studies have reported relationships between WM and aspects of L2 learning (see Juffs amp; Harrington, 2011, for a comprehensive review). However, only Sagarra (2000) has examined the longitudinal effects of WM in classroom second language acquisition, and none has investigated the predictive utility of WM and language proficiency together within Two-Way ANOVA. Thus, the current study sought to provide preliminary evidence of the predictive validity of these two executive functions as a first step in a research program. 2.Literature Review 2.1. Working memory capacity First put forward by Baddeley and Hitch in 1974, working memory is defined as part of the human memory system that combines the temporary storage and manipulation of information in the service of cognition. The central executive, phonological loop and visuospatial sketchpad, the three relatively independent components, make working memory a complicated system. Working memory has a limited capacity and is a working space where human conducts cognitive activities such as understanding, learning and deducting. 2.2 Language proficiency Language proficiency is the ability of an individual to speak or perform in an acquired language. As theories among pedagogues as to what constitutes proficiency go (Archived April 8, 2005), there is little consistency as to how different organizations classify it. Additionally, fluency and language competence are generally recognized as being related, but separate controversial subjects. In predominant frameworks in the United States, proficient speakers demonstrate both accuracy and fluency, and use a variety of discourse strategies. Thus, native speakers of a language can be fluent without being considered proficient. Native-level fluency is estimated to be between 20,000#8211;40,000 words, but basic conversational fluency might only require as little as 3,000 words. 2.3 Empirical studies In recent years, listening comprehension is widely considered to be of great importance in second language learning and there is much research done on L2 listening comprehension from the perspective of working memory at home and abroad. Researchers have been attempting to track down evidence on the extent to which working memory capacity is related to learners#8217; performance and development of L2 listening comprehension (A. Was, 2007; Andringa et al., 2012 ; Wu, 2002; Wang, 2008; Chen, 2008; Gu, 2007). Was A. (2007) reexamines the relationship between working memory capacity and listening comprehension and emphasizes the important role of available long-term memory. Through the experiment, the analyses reveal that available long-term memory mediates the relationships of both working memory and background knowledge with listening Comprehension. In Andringa et al. (2012) study, they aim to find out which factors explain success in native and non-native listening comprehension. 121 native and 113 non-native speakers of Dutch are involved in the experiment. In their study, they adopt four digit span tasks and one non-word recognition task to measure the subjects#8217; verbal working memory capacity. Structural equation modeling is used to identify the predictors of individual differences in listening comprehension. It is manifested by some of the test#8217; s results that working memory does not explain unique variance in listening comprehension in either group, whereas knowledge is proved to be the determinant of success in listening comprehension by both native and non-native speakers. 2.4 Limitations of previous studies In recent years, research on working memory capacity has mainly focused on three dimensions including language input, output and the combination of both. The present study sets about investigating the effect of working memory capacity and language proficiency on L2 learners#8217; listening comprehension. On the whole, previous related research is featured by applying the most widely used working memory capacity tests, inconsistent results owing to the small quantity of studies and different research designs, and the lack of the combination of working memory capacity and language proficiency. Firstly, there is not enough statistical evidence to support the researchers#8217; findings. Although many scholars have conducted research on the effective factors of L2 learners#8217; listening comprehension and put forward a lot of valuable results, there are still many arguments about the findings. Thus, more systematic research is needed to shed light into the relationship between L2 listening comprehension and its effective factors. Secondly, the working memory capacity test in previous studies needs to be improved, especially in non-English speaking countries. Slight differences in L2 language proficiency may have an enormous impact on the different dimensions of L2 speech production. It is methodologically problematic to assume that in a group everyone has the same or an approximate proficiency level and it is necessary to control for the subject#8217;s language proficiency by a proficiency test.

2. 研究的基本内容、问题解决措施及方案

3. Methodology 3.1 Research Questions The study that I carried out aimed at exploring the effects of the working memory capacity and language proficiency on L2 learners#8217; listening comprehension. In line with such objective, the following questions are addressed： 1、Is there a significant effect of working memory capacity on EFL learners#8217; listening comprehension? 2、Is there a significant effect of language proficiency on EFL learners#8217; listening comprehension? 3、Does the effect of language proficiency on L2 learners' listening comprehension vary according to working memory capacity? 3.2 Participants 72 undergraduate students (6 males, 66 females) from a Chinese university were included in the study. All of them were between 18 and 24 years old with a mean age of 21.72. Most of them had been learning English for twelve years (4 years for primary school, 6 years for middle schools and 2 years for high schools). Thirty of the participants had high language proficiency, and other had low language proficiency. 3.3 Instrument and procedure WM Tests As the fact that Daneman and Carpenter's (1980) Reading Span Test only measures information storage ability and test in L1 can help avoid the interruption of the subjects' second language proficiency, the working memory capacity was measure by Dai's (2010) Chinese computerized reading span test in the present study, which was designed on the basis of Cui and Chen's(1996) Chinese reading span test. Composed of 70 Chinese complex sentences, each of which consists of 16 to 18 Chinese words, the computerized Chinese reading span test is arranged in five sets each of two, three, four and five sentences. The number of stimulus sentences in each set increased progressively from two to five and each level there are five similar groups. As soon as five similar groups at the same level were finished, the subjects were supposed to proceed to the next level. In the test, each stimulus sentence in the CCRST was followed by a judgment sentence to measure subjects' comprehension of the stimulus sentence while they had to retain the last word of it. Of the 70 judgment sentences, half were true, while the other were false. After the presentation of the stimulus sentences and judgment sentences in each group, the subjects were supposed to recall as many the final words as they could, regardless of order and time limit. The example in the CCRST was as follows: Stimulus sentence:在欧洲，儿童文学经典作品安徒生童话是除圣经之外发行最广的图书。

Last word: 图书 Judgment sentence:安徒生童话是欧洲发行最广的图书（F） The CCRST was operationalized on a PC via a pre-designed software package. The operating procedures of CCRST were just the same as those in Daneman and Carpenter's (1980) and Waters and Caplan's (1996) RSTs (Reading Span Tests). In order to help the subjects get a clear understanding of the procedures. They were provided with practices before formal test. Besides, during the practice, the average reacting time of their sentence judgment was calculated by the software. The subjects were remind that there was a limit on reacting time when they made judgment in formal test. Once the time was out, the software automatically jumped to the next stimulus sentence and the sentence judgment was regarded as a mistake. Reliability analysis suggested that a Based on Standardized Items was .878 (close to α= .862). The CCRST was very reliable (α gt; 0. 8). Listening Comprehension Tests The materials of listening comprehension test were chosen from a CET-6 model test of Spark in 2012. Spark is a media company's name, which is very famous for tutoring material of English test in China. Many colleges choose model tests of Spark as training materials before real exams. The listening materials are presumably valid and reliable. Before the test, subjects were asked to write down whether they had done the same listening comprehension test or not on the answer sheet, in order to make sure of the reliability of the test results. The listening materials include three short passages and a compound dictation passage, for task types of answering questions in multiple-choice format and a short passage of compound dictation. In this test, there are 10 multiple-choices questions based on three passages, and 10 words or short phrase based on compound dictation, both passage were spoken only once. The total sore of listening comprehension test was 20 and it lasted 15 minutes. 3.4 Data collection and analysis The subjects#8217; working memory capacity was measured by the reading span test. It was based on two aspects: information processing and information storage. For information processing (sentence judgment), one point was given to each correct answer, and no point for a wrong answer. For information storage (words recall), one point was given to each correctly recalled item regardless of its order within a set, and no point for an incorrectly recalled or not-recalled item(2007) suggested that Daneman and Carpenter#8217;s (1980) working memory test didn#8217;t view semantic and syntactic judgment as part of test. Mackey (2010) also discussed the scoring method of working memory. He pointed out that the limitation of absolute scoring method used by Daneman and Carpenter (1980). According to their scoring method, working memory capacity depended on the maximum set size for which subjects could recall all the final words, which might allow subjects to predict the size of the upcoming set and use some strategies to affect the absolute scores. He suggested that absolute scoring method should be combined with aggregate scoring method (based on both correctly processed sentences and correctly recalled final words) when measuring working memory capacity.