Basque Center on Cognition Brain and Language, BCBL, Donostia-San Sebastián, Basque Country, Spain.
Basque Center on Cognition Brain and Language, BCBL, Donostia-San Sebastián, Basque Country, Spain.
Neuroimage. 2014 Mar;88:188-201. doi: 10.1016/j.neuroimage.2013.11.038. Epub 2013 Nov 27.
Language comprehension is incremental, involving the integration of information from different words together with the need to resolve conflicting cues when unexpected information occurs. The present fMRI design seeks to segregate the neuro-anatomical substrates of these two processes by comparing well-formed and ill-formed sentences during subject-verb agreement computation. Our experiment takes advantage of a particular Spanish feature, the Unagreement phenomenon: a subject-verb agreement mismatch that results in a grammatical sentence ("Los pintores trajimos…" [The painters3.pl (we)brought1.pl…]). Comprehension of this construction implies a shift in the semantic interpretation of the subject from 3rd-person to 1st-person, enabling the phrase "The painters" to be re-interpreted as "We painters". Our results include firstly a functional dissociation between well-formed and ill-formed sentences with Person Mismatches: while Person Mismatches recruited a fronto-parietal network associated to monitoring operations, grammatical sentences (both Unagreement and Default Agreement) recruited a fronto-temporal network related to syntactic-semantic integration. Secondly, there was activation in the posterior part of the left middle frontal gyrus for both Person Mismatches and Unagreement, reflecting the evaluation of the morpho-syntactic match between agreeing constituents. Thirdly, the left angular gyrus showed increased activation only for Unagreement, highlighting its crucial role in the comprehension of semantically complex but non-anomalous constructions. These findings point to a central role of the classic fronto-temporal network, plus two additional nodes: the posterior part of the left middle frontal gyrus and the left angular gyrus; opening new windows to the study of agreement computation and language comprehension.
语言理解是增量式的,涉及将来自不同单词的信息整合在一起,并且在出现意外信息时需要解决冲突线索。本 fMRI 设计旨在通过在主语-动词一致计算期间比较形成良好和形成不良的句子,来分离这两个过程的神经解剖学基础。我们的实验利用了西班牙语的一个特殊特征,即不一致现象:主语-动词不一致,导致语法句子(“Los pintores trajimos…”[画家们带来了……])。理解这种结构意味着主语的语义解释从第三人称到第一人称发生转变,使短语“Los pintores”被重新解释为“我们画家”。我们的结果包括首先在具有人物错误的形成良好和形成不良的句子之间进行功能分离:虽然人物错误招募了与监控操作相关的额顶网络,但语法句子(包括不一致和默认一致)招募了与句法-语义整合相关的额颞网络。其次,在左中额回的后部分对于人物错误和不一致都有激活,反映了对一致成分之间形态句法匹配的评估。第三,左角回仅在不一致时显示出增加的激活,突出了它在理解语义复杂但非异常结构中的关键作用。这些发现表明经典的额颞网络,加上两个额外节点:左中额回的后部分和左角回,在一致计算和语言理解研究中起着核心作用。