1 Neurology of Memory and Language Unit, Department of Neurology, Centre Hospitalier Sainte-Anne, 75014, Paris, France 2 INSERM, UMR_S1127, ICM, F-75013, Paris, France.
2 INSERM, UMR_S1127, ICM, F-75013, Paris, France 3 Centre de NeuroImagerie de Recherche (CENIR), Groupe Hospitalier Pitié-Salpêtrière, 75013, Paris, France 4 Sorbonne Universités, UPMC Univ Paris 06, UMR_S1127, ICM, F-75013, Paris, France 5 CNRS, UMR_7225, ICM, F-75005, 75013, Paris, France.
Brain. 2015 Feb;138(Pt 2):456-71. doi: 10.1093/brain/awu359. Epub 2014 Dec 16.
Concept formation is the ability to create an abstract link between dissimilar objects or thoughts and is crucial for abstract and creative thinking. This process is related to the integrity of the prefrontal cortex, given the altered performances reported in patients with frontal damage, particularly those suffering from the behavioural variant of frontotemporal dementia. However, the cognitive mechanisms and neural bases of verbal concept formation are not clearly understood. The present study was aimed at addressing the following unresolved issues regarding concept formation in the field of neurology and cognitive neuroscience: (i) Are alterations in concept formation specific to frontotemporal dementia or are they also present in other cortical neurodegenerative disorders such as Alzheimer's disease? (ii) Is impaired performance in concept formation due to cortical lesions specific to frontotemporal dementia or to a cortico-subcortical frontal syndrome? and (iii) What are the cognitive mechanisms and neural bases underlying concept formation? To address these questions, we designed the Verbal Concept Formation Task, an experimental paradigm based on the similarities test. Patients presenting with severe frontal dysfunction (frontotemporal dementia, n = 18, and the Richardson form of progressive supranuclear palsy, n = 21) or with medial temporal pathology (amnestic mild cognitive impairment or Alzheimer's disease, n = 14) and healthy participants (n = 18) were given the Verbal Concept Formation Task and a large battery of neuropsychological tests. In addition, all participants underwent 3D T1-weighted MRI to analyse grey matter volume using voxel-based morphometry. Frontal patients were significantly impaired on the Verbal Concept Formation Task as compared to non-frontal participants (P = 0.00001). Global performance score was positively correlated with scores in cognitive tasks assessing executive functions and with grey matter volume in several areas, mostly in the frontal-basal-ganglion network. Two types of errors were observed in frontal patients. The most frequent was discriminating instead of grouping items ('linking deficit'). Patients also linked items on a concrete instead of an abstract basis ('abstraction deficit'). Linking and abstraction deficits were related to partially different areas: the linking deficit to the dorsal anterior cingulate cortex, right middle frontal gyrus and both inferior parietal lobules and the abstraction deficit to the head of the caudate nuclei and the left superior frontal gyrus. These data suggest that verbal concept formation requires the integrity of the prefrontal-basal-ganglion functional network. In addition, it can be divided into two distinct cognitive processes, which are underlain by two partially different neural networks.
概念形成是在不同的物体或思想之间建立抽象联系的能力,对于抽象和创造性思维至关重要。这个过程与前额叶皮层的完整性有关,因为报告的前额损伤患者的表现改变,特别是那些患有额颞叶痴呆的行为变异型的患者。然而,言语概念形成的认知机制和神经基础尚不清楚。本研究旨在解决神经学和认知神经科学领域概念形成方面的以下未解决问题:(i)概念形成的改变是否仅存在于额颞叶痴呆,还是也存在于其他皮质神经退行性疾病如阿尔茨海默病?(ii)概念形成的受损表现是否是由于额颞叶痴呆的皮质病变特异性引起的,还是由于皮质下额前综合征引起的?(iii)概念形成的认知机制和神经基础是什么?为了解决这些问题,我们设计了言语概念形成任务,这是一种基于相似性测试的实验范式。患有严重额叶功能障碍的患者(额颞叶痴呆,n = 18,和进行性核上性麻痹的 Richardson 型,n = 21)或有内侧颞叶病变的患者(遗忘型轻度认知障碍或阿尔茨海默病,n = 14)和健康参与者(n = 18)接受了言语概念形成任务和大量神经心理学测试。此外,所有参与者都接受了 3D T1 加权 MRI 以使用基于体素的形态计量学分析灰质体积。与非额叶参与者相比,额叶患者在言语概念形成任务中表现明显受损(P = 0.00001)。总体表现评分与评估执行功能的认知任务的评分呈正相关,与多个区域的灰质体积呈正相关,主要在前额-基底节网络中。在额叶患者中观察到两种类型的错误。最常见的是将项目进行区分而不是分组(“连接缺陷”)。患者还将项目连接在具体而非抽象的基础上(“抽象缺陷”)。连接和抽象缺陷与部分不同的区域有关:连接缺陷与背侧前扣带回皮层、右额中回和两个下顶叶有关,而抽象缺陷与尾状核头部和左额上回有关。这些数据表明,言语概念形成需要前额-基底节功能网络的完整性。此外,它可以分为两个不同的认知过程,这两个过程由两个部分不同的神经网络支撑。