基于阅读时间的二语习得者句子空隙处理研究

论文摘要

多年来，第一语言（以下简称一语）的句子处理问题一直是语言学和心理语言学的研究热点。第二语言习得研究（以下简称二语习得研究）则关注于个体的学习过程和结果，并致力于解释其达到某种语言水平的原因（Dussias,2003:531）。鉴于二语习得研究在侧重点上同一语句子处理研究有很大不同，加之研究者缺乏句子处理研究所需的技术资源和专业知识，因此，二语句子处理方面的研究一直未受到人们的关注。无论在一语还是在二语中，如何处理语言输入都是习得语法体系的关键，这是因为目标语语法的构建是由语言输入来驱动的。而二语习得者如何实时处理目标语语言输入对于研究二语语法的发展尤为重要。此外，许多研究者（Clahsen&Felser,2006a, b; Felser&Roberts,2007; Juffs,2005; Papadopoulou,2005）指出，成年人二语习得难以获得成功正是由于二语习得者不佳的句法分析机制所致。本研究针对近几年二语句子处理研究的现状，通过心理语言学的实验手段，考察中国英语学习者在线阅读两种填充词-空隙依赖结构（中间空隙和主语填充空隙）时所反映出的句法分析机制，比较其与英语本族人之间的区别，分析英语水平对句子处理过程的影响，结合现有相关理论进行讨论的同时提出作者自己的理论假设，并探讨研究结果对英语教学的启示。本研究的理论框架为建立在原则与参数理论基础上的Prichett的句法分析理论以及Clifton&Frazier的活跃填充词假设。研究中采用的实验句是基于这些理论框架进行设计，并进而展开分析和讨论的。实证研究由测试中间空隙效应和测试主语填充空隙效应两部分组成。为研究二语习得者处理句子空隙的能力，两部分实验句都遵循着将句法空隙处理（Syntactic Gap Processing）同词汇/语义联合（Lexical/Semantic Association）分开的原则。两部分实验步骤一致。被试者为上海某高校非英语专业二年级学生，根据大学英语四级成绩分为高分组和低分组，每组各35人。另设由32名英语本族人组成的对照组。实验前，每位被试者均完成了工作记忆水平的测试，用以研究工作记忆对实验结果的影响。本研究采用离线任务（offline task）和在线任务（online task）相结合的实验方法。离线任务要求被试者阅读由实验句改编而来的句子并回答问题，以考察被试是否具备理解这类语法结构的能力。在线阅读任务于离线任务后三星期进行，采用自控步速阅读法（Self-Paced Reading）的技术手段，收集被试者在线阅读实验句各部分所花时间及其对句子的理解情况。主要研究结果如下:（一）在线阅读过程中，二语习得者所使用的句法分析机制同英语本族人有着本质的区别。本族人能觉察到实验句子中传达句法信息的功能词语，并善于利用它们建立深层句法表征；而二语习得者无法像本族人那样利用句子空隙进行句法分析，主要依靠句中的词汇/语义信息来完成对句子的理解。（二）二语习得者的语法知识同其句法分析机制之间的关系不平衡。被试者在回答有关实验句的问题时准确率很高，表明他们具备了必要的词汇和语法知识，从而可以准确理解句子的含义。然而不同于本族人的是，他们在线阅读过程中，不能够利用实验句中的某些语法信息进行深入的句法分析。（三）两个水平组的被试者在中间空隙效应的实验中差异不显著，他们不能利用实验句中的中间空隙来完成填充词的整合。然而，两组被试者在主语填充空隙效应的实验中表现却不同。高分组的被试者同本族人一样进行了主语空隙的分析，而低分组的被试者则没能进行这种空隙分析。本研究对二语习得研究的贡献主要体现在以下几方面:（一）现有的一些相关理论，比如输入处理理论（Input Processing Theory: VanPatten,1996,2000,2002,2004），陈述性/程序性记忆模型（Declarative/ProceduralModel: Ullman,2001,2004,2006）以及浅结构假设（Shallow Structure Hypothesis:Clahsen&Felser,2006a, b）等理论，在探讨二语习得者处理目标语语言输入的过程时，都没有考虑到句法结构的不同为处理过程带来的影响。针对本研究的结果，作者提出了“空隙可及性等级”（GapAccessibility Hierarchy）的假设，解释了高分组被试者处理两种句法空隙时采用不同策略的原因，充实了二语句子处理理论研究的内容。（二）本研究在实验方法上有所改进。首先，不同于以往二语句子处理研究选取同一语言水平被试者的模式，将英语语言水平作为被试间因素融入实验设计中，有助于揭示句法分析机制的发展轨迹。其次，采用了经过特殊设计的实验句，成功地将句法空隙处理同词汇/语义联合区分开，更加准确地反映了二语句子空隙处理的真实情况。（三）本研究对英语教学有一定的启示。针对二语习得者语法知识同句法分析机制之间的不平衡关系，作者指出通过课堂教学帮助学习者改变先前不佳的句法处理策略的必要性，并探讨了“处理教学法”在我国开展的可行性。

论文目录

• Acknowledgments
• 摘要
• Abstract
• List of Tables
• List of Figures
• Abbreviations
• Table of Contents
• Chapter One Introduction
• 1.1 Statement of the Problem
• 1.1.1 Relevance of Sentence Processing to SLA
• 1.1.2 The Current Situation
• 1.2 The Need for the Present Study
• 1.2.1 The Processing of Filler-Gap Dependency Structures in L1
• 1.2.2 Weakness of Previous Filler-gap Processing Studies in L1 and L2
• 1.3 The Significance of the Present Study
• 1.3.1 Theoretical significance
• 1.3.2 Methodological Significance
• 1.3.3 Pedagogical Significance
• 1.4 Research Questions
• 1.5 Methodology
• 1.6 Organization of the Dissertation
• Chapter Two Review of Researches
• 2.1 Significance of Sentence Processing Research to SLA
• 2.2 Theoretical Models Accounting for L2 Sentence Processing
• 2.2.1 The Competition Model
• 2.2.2 Input Processing Model
• 2.2.3 The Declarative/Procedural Model
• 2.2.4 The Shallow Structure Hypothesis
• 2.3 Previous Studies on L2 Sentence Processing
• 2.3.1 The Processing of Structural Ambiguity in L2
• 2.3.2 The Processing of Filler-gap Dependencies in L2
• 2.3.3 Sentence Processing Research in China
• 2.4 Summary
• Chapter Three Theoretical Framework
• 3.1 Essential Issues in the Study of Sentence Processing
• 3.1.1 Principle-based vs. Constraint-based
• 3.1.1.1 Principle-based Approach
• 3.1.1.2 Constraint-based Approach
• 3.1.2 Grammar vs. Parser
• 3.2 The Processing of Filler-gap Dependency Structures
• 3.2.1 Syntactic Analysis: Gap vs. Gapless
• 3.2.2 Parsing Analysis: TRH vs. DAH
• 3.3 Theoretical Basis of the Present Study
• 3.3.1 Pritchett’s Parsing Theory: A Principle-Based Parsing Framework
• 3.3.2. Active Filler Hypothesis
• 3.4 Summary
• Chapter Four Research Methodology
• 4.1 The Two Syntactic Gaps Tested in the Empirical Study
• 4.1.1 The Intermediate Gap
• 4.1.1.1 Syntactic Analysis
• 4.1.1.2 The Designing Principle of the Experimental sentences
• 4.1.2 The Subject Filled Gap
• 4.1.2.1 Comparison of Subject Filled Gap with Object Filled Gap
• 4.1.2.2 The Designing Principle of the Experimental Sentences
• 4.2 General Procedures
• 4.2.1 Selection of Participants
• 4.2.1.1 Proficiency Control
• 4.2.1.2 Working Memory Control
• 4.2.2 Instruments
• 4.2.2.1 Offline Questionnaire
• 4.2.2.2 Online Reading Task
• 4.2.3 Data Collection and Analysis
• 4.3 Summary
• Chapter Five Experiment Testing Intermediate Gap Effect
• 5.1 Offline Questionnaire
• 5.2 Online Reading Task
• 5.2.1 Materials
• 5.2.2 Segmentation of the Experimental Sentences
• 5.3 Results from Online Comprehension Questions
• 5.3.1 Accuracy Score Figures
• 5.3.2 Statistical Analysis
• 5.4 Results from Online Reading Time
• 5.4.1 Reading Time Figures
• 5.4.2 Data Treatment and Reduction
• 5.4.3 Data Analysis at Segment 3
• 5.4.3.1 Reading Time Figures
• 5.4.3.2 Statistical Analysis
• 5.4.4 Data Analysis at Segment 5
• 5.4.4.1 Reading Time Figures
• 5.4.4.2 Statistical Analysis
• 5.5 Summary and Discussion
• 5.5.1 Summary of Test Results
• 5.5.2 The Native Speakers
• 5.5.3 The L2 Learners
• 5.5.4 The Role of Different Proficiency Level
• 5.5.5 The Role of Working Memory
• Chapter Six Experiment Testing Subject Filled-Gap Effect
• 6.1 Offline Questionnaire
• 6.2 Online Reading Task
• 6.2.1 Materials
• 6.2.2 Segmentation of the Experimental Sentences
• 6.3 Results from Online Comprehension Questions
• 6.3.1 Accuracy Score Figures
• 6.3.2 Statistical Analysis
• 6.4 Results from Online Reading Time
• 6.4.1 Reading Time Figures
• 6.4.2 Data Treatment and Reduction
• 6.4.3 Data Analysis at Segment 4
• 6.4.3.1 Reading Time Figures
• 6.4.3.2 Statistical Analysis
• 6.4.4 Data Analysis at Segment 7
• 6.4.4.1 Reading Time Figures
• 6.4.4.2 Statistical Analysis
• 6.5 Summary and Discussions
• 6.5.1 Summary of Test Results
• 6.5.2 The Subject Filled-Gap Effect at Segment
• 6.5.3 The “Filled Object Effect” at Segment
• 6.5.4 The Length Manipulation and Working Memory Level
• Chapter Seven General Discussions and Implications
• 7.1 Form and Meaning in L2 Gap Processing
• 7.2 Weakened Automaticity in L2 Processing
• 7.3 Possibility of L1 Processing Transfer
• 7.4 Gap Accessibility Hierarchy Hypothesis—A Tentative Model for L2Gap Processing
• 7.4.1 Comparing the Intermediate Gap Effect with the Subject Filled-GapEffect
• 7.4.2 Gap Accessibility Hierarchy Hypothesis
• 7.5 From Grammatical Competence to Parsing Mechanism
• 7.6 Pedagogic Implication – The Need for Processing Instruction
• 7.6.1 Basic Characteristics of PI
• 7.6.2 Comparison of PI with Other Instructional Methods
• 7.6.3 Feasibility of PI in China
• 7.7 Summary
• Chapter Eight Conclusion
• 8.1 Major Findings of the Present Study
• 8.1.1 L2 Learners vs. Native Speakers
• 8.1.2 Grammatical Competence vs. Parsing Performance
• 8.1.3 The Role of Proficiency Level
• 8.1.4 Implications of the Present Study
• 8.2 Limitations of the Present Study
• 8.3 Suggestions for Future Research
• References
• Appendices
• Appendix Ⅰ Sets of Sentences in the Working Memory Span Task
• Appendix Ⅱ Sentences in Offline Questionnaire for Testing Intermediate Gap Effect
• Appendix Ⅲ Sentences in Online Reading Task for Testing Intermediate Gap Effect
• Appendix Ⅳ Sentences in Offline Questionnaire for Testing Subject Filled-Gap Effect
• Appendix Ⅴ Sentences in Online Reading Task for Testing Subject Filled-Gap Effect
• 攻读学位期间发表的学术论文