Facultad de Psicología, Univesidad Autónoma de Madrid, 28049, Madrid, Spain.
Universidad Complutense de Madrid, Madrid, Spain.
Brain Struct Funct. 2021 Apr;226(3):845-859. doi: 10.1007/s00429-020-02213-4. Epub 2021 Jan 20.
Resting state functional connectivity research has shown that general cognitive ability (GCA) is associated with brain resilience to targeted and random attacks (TAs and RAs). However, it remains to be seen if the finding generalizes to structural connectivity. Furthermore, individuals showing performance levels at the very high area of the GCA distribution have not yet been analyzed in this regard. Here we study the relation between TAs and RAs to structural brain networks and GCA. Structural and diffusion-weighted MRI brain images were collected from 189 participants: 60 high cognitive ability (HCA) and 129 average cognitive ability (ACA) individuals. All participants completed a standardized fluid reasoning ability test and the results revealed an average HCA-ACA difference equivalent to 33 IQ points. Automated parcellation of cortical and subcortical nodes was combined with tractography to achieve an 82 × 82 connectivity matrix for each subject. Graph metrics were derived from the structural connectivity matrices. A simulation approach was used to evaluate the effects of recursively removing nodes according to their network centrality (TAs) versus eliminating nodes at random (RAs). HCA individuals showed greater network integrity at baseline and prior to network collapse than ACA individuals. These effects were more evident for TAs than RAs. The networks of HCA individuals were less degraded by the removal of nodes corresponding to more complex information processing stages of the PFIT network, and from removing nodes with larger empirically observed centrality values. Analyzed network features suggest quantitative instead of qualitative differences at different levels of the cognitive ability distribution.
静息态功能连接研究表明，一般认知能力（GCA）与大脑对靶向和随机攻击（TAs 和 RAs）的弹性有关。然而，这一发现是否适用于结构连接还有待观察。此外，在这方面，还没有分析表现出 GCA 分布极高水平的个体。在这里，我们研究了 TAs 和 RAs 与结构脑网络和 GCA 之间的关系。从 189 名参与者中收集了结构和弥散加权 MRI 脑图像：60 名高认知能力（HCA）和 129 名平均认知能力（ACA）个体。所有参与者都完成了一项标准化的流体推理能力测试，结果显示 HCA-ACA 的平均差异相当于 33 个智商点。皮质和皮质下节点的自动分割与追踪相结合，为每个受试者获得了 82×82 的连接矩阵。从结构连接矩阵中得出了图度量。采用模拟方法来评估根据网络中心性（TAs）递归地按顺序删除节点与随机删除节点（RAs）的效果。与 ACA 个体相比，HCA 个体在基线和网络崩溃之前显示出更大的网络完整性。与 RAs 相比，这些影响在 TAs 中更为明显。HCA 个体的网络在 PFIT 网络更复杂的信息处理阶段的节点以及具有更大经验观测中心性值的节点被删除时，其网络退化程度较小。分析的网络特征表明，在认知能力分布的不同水平上存在定量而不是定性的差异。
