句子加工是心理语言学研究的热点之一。当前句子加工领域存在两大对立的理论：花园路径模型和基于制约的模型。前者认为句子加工是模块、串行的，最初只有句法信息起作用，帮助人们构建结构上最简单的分析；后者认为句子加工中各种信息交互作用，所有可能的结构分析同时激活。然而，面对句子加工研究揭示的复杂现象，两大模型均未能做出合理解释。魏行编著的《交互串行加工模型--来自汉语歧义构式V NP1de NP2的加工证据》指出，句子加工完全可能按交互、串行的方式进行：各种信息交互作用，帮助人们建构一种最合理的分析。鉴于此，本研究提出了交互串行加工模型，并以汉语歧义构式为对象，考察了该模型的可行性。

Acknowledgements

前言

Abstract

Chapter 1 Introduction

 1.1 Research Orientation

 1.2 Research Background

 1.3 The Object of Study

 1.4 Research Questions

 1.5 Organization of the Book

Chapter 2 Sentence Processing Models: A Critical Review

 2.1 Modular versus Interactive Models

2.1.1 Modular Models

2.1.2 Interactive Models

 2.2 Serial versus Ranked-Parallel Models

2.2.1 Serial Models

2.2.2 Parallel Models

2.2.3 Unrestricted Race Model--A Hybrid Model

 2.3 An Assessment of Current Sentence Processing Models

 2.4 Serial Interactive Processing: A Possible Model

 2.5 Summary

Chapter 3 Experimental Methods Used in This Study

 3.1 Self-paced Reading

3.1.1 Three Versions of Self-paced Reading

3.1.2 Potential Problems with Self-paced Reading

3.1.3 Investigating Semantic Effects Using Self-paced Reading

 3.2 Structural Priming

3.2.1 Structural Priming as a Research Paradigm

3.2.2 Structural Priming in Sentence Comprehension

3.2.3 Investigating the Activation of Analyses Using Structural Priming

 3.3 Summary

Chapter 4 Experimental Studies on the Chinese Ambiguous Construction V NP1 de NP2

 4.1 Zhang's (1998) Study

 4.2 Hsieh et all's (2009) Study

 4.3 Critique of the Two Studies

 4.4 Overview of the Present Study

Chapter 5 Experiment One: Effects of Semantic Plausibility Information

 5.1 Introduction

 5.2 Experiment 1A: Semantic Effects on Processing VP-biased Items

5.2.1 Research Question and Hypothesis

5.2.2 Method

5.2.3 Results

5.2.4 Discussion

 5.3 Experiment 1B: Semantic Effects on Processing NP-biased Items

5.3.1 Research Question and Hypothesis

5.3.2 Method

5.3.3 Results

5.3.4 Discussion

 5.4 General Discussion

 5.5 Summary

Chapter 6 Experiment Two: Analysis Activated during Online Processing--Self-paced Reading

 6.1 Introduction

 6.2 Research Question and Hypothesis

 6.3 Method

 6.4 Results

 6.5 Discussion

 6.6 Summary

Chapter 7 Experiment Three: Analysis Activated during Online Processing--Structural Priming

 7.1 Introduction

 7.2 Experiment 3A: Priming of the VP Analysis

7.2.1 Research Question, Hypothesis and Prediction

7.2.2 Method

7.2.3 Results

7.2.4 Discussion

 7.3 Experiment 3B: Priming of the NP Analysis

7.3.1 Research Question, Hypothesis and Prediction

7.3.2 Method

7.3.3 Results

7.3.4 Discussion

 7.4 General Discussion

 7.5 Summary

Chapter 8 Discussion and Implications

 8.1 A Summary of Empirical Findings

8.1.1 Immediate Use of Semantic Plausibility Information

8.1.2 NP Being the Preferred Analysis

8.1.3 Semantically Equibiased Items Being Represented as NP

8.1.4 The On-line Processing of V NP1 de NP2: A Snapshot

 8.2 Implications of the Findings

8.2.1 Immediate Semantic Effects: Evidence against Modularity

8.2.2 One Analysis at a Time: Evidence against Ranked Parallelism

 8.3 Interactive Serial Processing (ISP): A New Sentence Processing Model

 8.4 Limitations

 8.5 Suggestions for Future Research

References

Appendix A The Plausibility Rating and Preference Percentage of the Test Items

Appendix B Experimental Materials Employed in Experiments 1A and 1B

Appendix C Experimental Materials Employed in Experiment 2

Appendix D Experimental Materials Employed in Experiments 3A and 3B