汉语焦点信息影响代词回指的行为和脑电研究

Up to now, although there have been a great number of studies conducted to deal with the problem of pronoun resolution, most of these studies were concerned with the question of how syntactic and semantic constraints influence the process of pronoun resolution. The question of how pronoun resolution is influenced by pragmatically-based constraints, such as information structure (especially focus structure) has not received as much attention in the literature. In particular, we know little about the processes of how focus structures interact with other linguistic information, such as grammatical role, verb semantic, and pronoun-antecedent distance, to determine the interpretation of pronoun. The purpose of this project is to gain insight into the question of how information structure, i.e. focus structures, influences the interpretation of the pronoun. To this end, we choose two different types of focus structures, that is, a top-down focus structure (linguistic focus) and a bottom-up focus structure (non-linguistic focus). More specifically, by measuring the latency and amplitude of Event related potential (ERP) components, we want to reveal the process of how these two different types of focus structures exert their influence on pronoun resolution, and during this process whether/how different focus information interact with syntactic and semantic information. Moreover, we also want to reveal the correspondingly electrophysiological response underlying these different processing stages. The systematic studies of how focus structures exert their influence on the process of pronoun resolution and especially, how focus information interacts with other syntactic and semantic information during pronoun resolution, can tell us how linguistic constraints and non-linguistic constraints interact with each other to contribute to anaphor resolution and sentence representation. Specifically, there are three key questions to address here, which are: 1) how top-down focus information influences the pronoun resolution; 2) how bottom-up focus information influences the pronoun resolution; 3) how focus structure, grammatical role, and verb semantic interact with each other during pronoun resolution.

目前虽然有关代词回指的研究有很多，但这些研究主要关注句法和语义因素如何影响回指加工，没有考察话语（语用）层面的信息结构，尤其是焦点结构，如何影响回指解读，没有考察不同类型的焦点信息是如何通过与句法角色、动词语义以及距离远近等因素的相互作用来决定回指解读。为此，本项目将在以往研究和项目组前期成果的基础之上，选择两种不同的焦点结构，即：自上而下的焦点（语言焦点）和自下而上的焦点（非语言焦点），并借助高时间分辨率的ERP技术，考察这两种不同类型的焦点信息对代词回指影响的异同，以及它们与句法、语义和距离等因素相互作用于回指加工的认知和神经机制，以此揭示在汉语句子理解中来自语言和非语言的信息是通过何种关系模式来影响回指加工和句子表征。具体关注的研究问题包括：1）自上而下的焦点信息如何影响回指加工；2）自下而上的焦点信息如何影响回指加工；3）焦点结构、句法角色和动词语义对回指加工的综合影响。

代词回指问题一直是语言学和心理语言学领域的研究热点之一，但现有研究主要关注句法和语义因素如何影响回指加工，没有考察语用层面的信息结构，尤其是焦点结构，如何影响回指解读。借用高时间分辨率的脑电技术，本项目考察了不同的焦点信息、句法角色以及动词语义如何影响汉语的代词回指。研究显示：（1）无论是在代词加工的早期阶段还是晚期阶段，自上而下焦点信息都会对代词回指产生影响。在早期阶段焦点信息对代词回指的影响不受先行语句法角色的影响，但在晚期阶段焦点信息和句法角色共同决定代词的回指解读。（2）相对于非话题（如：主语、宾语）位置上的先行语，代词倾向于回指话题位置上的先行语，非话题回指会导致话题转移，从而诱发更大的P600效应。这一结果不受动词语义偏向的影响。（3）在电生理机制上，汉语焦点信息对代词回指的影响主要表现为P600效应的变化，P600成分的大小会受到焦点信息、句法结构以及动词语义等多种信息的综合影响。本研究中没有发现印欧语回指文献中经常报告的LAN（或者N400）。综合上述结果表明，汉语焦点信息对代词回指的影响在代词回指的两个加工阶段均会出现。相对于印欧语的研究结果，汉语回指加工中，焦点、句法和语义对代词回指的影响主要表现在晚期加工阶段。在这一阶段，焦点、句法、语义等信息产生相互作用，共同决定代词的回指解读。