Xu Yangwen, Lin Qixiang, Han Zaizhu, He Yong, Bi Yanchao
National Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing 100875, China.
National Key Laboratory of Cognitive Neuroscience and Learning & IDG/McGovern Institute for Brain Research, Beijing Normal University, Beijing 100875, China.
Neuroimage. 2016 May 15;132:542-555. doi: 10.1016/j.neuroimage.2016.03.004. Epub 2016 Mar 10.
Semantic processing entails the activation of widely distributed brain areas across the temporal, parietal, and frontal lobes. To understand the functional structure of this semantic system, we examined its intrinsic functional connectivity pattern using a database of 146 participants. Focusing on areas consistently activated during semantic processing generated from a meta-analysis of 120 neuroimaging studies (Binder et al., 2009), we found that these regions were organized into three stable modules corresponding to the default mode network (Module DMN), the left perisylvian network (Module PSN), and the left frontoparietal network (Module FPN). These three dissociable modules were integrated by multiple connector hubs-the left angular gyrus (AG) and the left superior/middle frontal gyrus linking all three modules, the left anterior temporal lobe linking Modules DMN and PSN, the left posterior portion of dorsal intraparietal sulcus (IPS) linking Modules DMN and FPN, and the left posterior middle temporal gyrus (MTG) linking Modules PSN and FPN. Provincial hubs, which converge local information within each system, were also identified: the bilateral posterior cingulate cortices/precuneus, the bilateral border area of the posterior AG and the superior lateral occipital gyrus for Module DMN; the left supramarginal gyrus, the middle part of the left MTG and the left orbital inferior frontal gyrus (IFG) for Module FPN; and the left triangular IFG and the left IPS for Module FPN. A neuro-functional model for semantic processing was derived based on these findings, incorporating the interactions of memory, language, and control.
语义处理需要激活颞叶、顶叶和额叶广泛分布的脑区。为了理解这个语义系统的功能结构，我们使用了一个包含146名参与者的数据库来研究其内在功能连接模式。基于对120项神经影像学研究的荟萃分析（Binder等人，2009年）所产生的语义处理过程中持续激活的区域，我们发现这些区域被组织成三个稳定的模块，分别对应默认模式网络（模块DMN）、左侧颞周网络（模块PSN）和左侧额顶网络（模块FPN）。这三个可分离的模块通过多个连接枢纽整合在一起——左侧角回（AG）和左侧额上/中回连接所有三个模块，左侧颞前叶连接模块DMN和PSN，背侧顶内沟（IPS）的左后部连接模块DMN和FPN，以及左侧颞中回后部（MTG）连接模块PSN和FPN。还确定了在每个系统内汇聚局部信息的省级枢纽：模块DMN的双侧后扣带回皮质/楔前叶、双侧后AG和枕外侧上回的边界区域；模块FPN的左侧缘上回；左侧MTG的中部和左侧眶额下回（IFG）；以及模块PSN的左侧三角IFG和左侧IPS。基于这些发现得出了一个语义处理的神经功能模型，该模型纳入了记忆、语言和控制的相互作用。
