Schmitt Bernadette M, Lamers Monique, Münte Thomas F
Department of Neurocognition, Faculty of Psychology, Maastricht University, PO Box 616, 6200 MD, Maastricht, The Netherlands.
Brain Res Cogn Brain Res. 2002 Nov;14(3):333-46. doi: 10.1016/s0926-6410(02)00136-2.
During comprehension, a personal pronoun (he, she, it) refers to a preceding referent (boy, girl, child). This co-reference could be established, among other ways, by using (i). conceptual/semantic information (biological gender agreement between the pronoun and its referent), (ii). syntactic information (syntactic gender agreement), or (iii). both. This event-related brain potential (ERP) study assesses this interplay of syntactic and semantic information. We focussed on the N400 component, related to semantic integration, and the SPS/P600 component, related to syntactic reanalysis. The study was conducted in German, because its rich syntactic gender system offers a means to dissociate between biological (MALE/FEMALE) and syntactic gender (masculine/feminine/neuter), especially in the case of diminutives (das(neuter) Bübchen(MALE) [the little boy]). German subjects read sentences in which a referent (Bübchen(MALE-neuter)/Bub(MALE-masculine) [boy]) was introduced. Later a personal pronoun was presented which either agreed with the referent in terms of syntactic gender, or in terms of biological gender, or both, or violated both agreements. Overall, results showed salient P600 effects for pronouns. This indicates that the establishment of reference involves syntactic reanalysis. Furthermore, we observed N400 effects in sentences with non-diminutives, but not with diminutives. This shows that conceptual/semantic integration is involved during non-diminutive but not during diminutive pronoun processing, or at least it could not be violated. The overall pattern of results supports the claim that for non-diminutives, both syntactic and conceptual information is used to establish co-reference, while for diminutives the process might be purely syntactically driven.
在理解过程中,人称代词(他、她、它)指代前面的所指对象(男孩、女孩、孩子)。这种共指关系可以通过多种方式建立,比如:(i)概念/语义信息(代词与其所指对象之间的生物性别一致性)、(ii)句法信息(句法性别一致性)或者(iii)两者皆有。本事件相关脑电位(ERP)研究评估了句法和语义信息之间的这种相互作用。我们聚焦于与语义整合相关的N400成分以及与句法重新分析相关的SPS/P600成分。该研究以德语进行,因为其丰富的句法性别系统提供了一种区分生物性别(男性/女性)和句法性别(阳性/阴性/中性)的方法,尤其是在指小词的情况下(das(中性)Bübchen(男性)[小男孩])。德国受试者阅读引入了所指对象(Bübchen(男性 - 中性)/Bub(男性 - 阳性)[男孩])的句子。随后呈现一个人称代词,该代词在句法性别方面、生物性别方面或者两者都与所指对象一致,或者两者都不一致。总体而言,结果显示代词存在显著的P600效应。这表明指代关系的建立涉及句法重新分析。此外,我们在非指小词的句子中观察到了N400效应,但在指小词的句子中未观察到。这表明在处理非指小词代词时涉及概念/语义整合,而处理指小词代词时则不涉及,或者至少不会被违反。结果的总体模式支持这样一种观点,即对于非指小词,句法和概念信息都用于建立共指关系,而对于指小词,这个过程可能纯粹由句法驱动。