基于网络分析方法研究珠心算训练对脑功能网络的影响

发布时间：2018-02-06 07:46

本文关键词： 脑网络珠心算训练功能磁共振认知能力　出处：《浙江大学》2017年博士论文　论文类型：学位论文

【摘要】：大脑是人体最复杂的器官,是调节人类各种功能的中枢。研究人员利用各种方法认识脑、保护脑、开发脑。近年来,人们发现脑区之间功能的整合与分化使脑存在特殊的网络拓扑结构。而这种网络拓扑结构与日常认知能力息息相关,并且会随着发展发育、疾病、训练而发生一定的改变。珠心算是一种基于视空间表征的计算方法。研究发现,珠心算训练不仅会影响人的认知能力,并且会影响脑的功能和结构。然而,现有的研究缺乏从脑网络角度研究珠心算训练对脑神经机制的影响,从而无法从全脑整体角度和脑区之间的相互关系来了解珠心算的神经机制。本研究中,我们采用图论理论,基于静息态和任务态功能磁共振数据,探究珠心算训练对脑网络拓扑结构的影响。由于在儿童磁共振数据处理中,空间标准化步骤容易产生误差,影响实验结果。因此,在实验中我们首先优化磁共振处理步骤,希望能增强结果可靠性。在研究中:(1)我们构建了基于实验样本的儿童脑模板,检验儿童模板对于磁共振结果可靠性的影响。我们发现,即使实验样本量较小,构建的脑模板依然可以降低空间标准化带来的误差。同时,在统计分析中,将个体空间与模板空间的差异作为协变量进行相应的矫正,可以降低这种差异对统计检验的影响,达到增强统计检验的敏感性的目的。所以,作为整个研究的基础,在之后的研究中我们均构建了符合被试信息的模板。之后,我们设计了三个子实验研究珠心算训练对脑功能网络的影响,我们发现:(2)视空间策略的广泛运用会增强相关脑区与其它脑区之间的联系,增强这些脑区处理信息效率,并提升相应脑区在脑功能网络中的重要性。这些脑区包括右侧前扣带回、右侧顶下小叶和右侧眶内额上回。(3)不同脑区之间联系的紧密程度是不同的,并以不同的子网络形式存在,不同认知能力相关的子网络组成了全脑功能网络。本研究发现,珠心算训练会促进脑网络中子网络的功能分化,增加子网络的内部连接,降低子网络间的连接。同时,珠心算训练对不同子网络的拓扑结构也存在不同的影响,例如提升视觉网络平均局部效率,而降低运动感知觉网络的平均参与系数。我们推测这些网络结构的变化与视空间策略在珠心算训练中的广泛运用密切相关,训练促使这些网络能更独立、更有效地处理信息。(4)此前研究发现训练与高级认知能力之间存在广泛的迁移效应。基于执行功能任务,我们发现珠心算训练会增强训练者的行为绩效。脑网络分析发现,在执行功能相关任务中,珠心算训练者的额顶网络的功能连接强度显著的大于对照组。珠心算训练不仅影响静息态下脑网络结构,同样会影响特定任务态下的脑连接属性。我们推测,在珠心算训练中,视空间策略的广泛运用会增强相应脑区之间的联系,进而提高这些脑区信息处理的能力。而这些脑区信息处理效率的提升又会进一步影响与之相关的高级认知能力。因此,我们的研究从脑网络的角度为珠心算训练与认知能力之间的迁移效应提供了有力的理论支持,有助于我们进一步了解珠心算训练与认知能力之间的关系。
[Abstract]:The brain is the most complex organ in the body, is the central regulation of various functions of human beings. The researchers use a variety of methods to recognize the brain, protect brain, brain development. In recent years, people found that the integration and differentiation of brain regions between the brain network topology. This special network topology structure and cognitive ability is closely related to the daily. And with the development of disease, development, training and change. Abacus is a calculating method based on visual spatial representation. The study found that mental arithmetic training will not only affect people's cognitive ability, and will affect the structure and function of the brain. However, existing studies lack of mental abacus training on of brain mechanisms from the brain network angle, which can not be from the relationship between whole brain and brain regions to the whole view of the neural mechanism of solution. In this study, bead mental arithmetic, we use graph theory, based on static State information and task state fMRI data, explore the impact of abacus training on brain network topology. Because of the children's MRI data processing, spatial normalization error prone, affect the experimental results. Therefore, in the experiment we first optimize the magnetic resonance processing steps, hoping to enhance the reliability of the results in the study.: (1) we constructed the experimental sample of children brain template based on template for magnetic resonance inspection of children influence the reliability. We found that even if the sample size is small, the brain template can still reduce the error caused by spatial normalization. At the same time, in the statistical analysis, the difference between individual space and space template as a covariate corrected, can reduce the impact of this difference on the statistical test, to enhance the sensitivity of the statistical test purpose. Therefore, as a basis for the whole research In the study, we have constructed with information subjects template. After that, we design, three sub experiments of abacus mental calculation training on brain functional network we found: (2) extensive use of visual space strategy can enhance the correlation between brain regions and other brain areas, enhance the efficiency of information treatment of brain regions, and enhance the importance of the corresponding brain regions in the brain functional network. These brain regions including the right anterior cingulate gyrus, right inferior parietal lobule and right orbital frontal gyrus. (3) the connection degree between different brain regions is different, and existed in the sub network of different forms and different cognitive abilities the sub network composed of whole brain functional networks. The study found that the functional differentiation will promote brain network trained network, increase internal network connection, reduce the connection among sub networks. At the same time, the calculation of different training The topological structure of sub networks have different effects, such as enhancing the visual network average local efficiency, and reduce the average participation coefficient of movement perception network is widely used. We speculate that these changes in network structure and spatial strategy in the closely related mental arithmetic training, training the network to be more independent and more efficient information processing. (4) previously found extensive migration effect between the training and cognitive ability. Based on the executive function tasks, we found that the training will enhance the trained performance behavior. Brain network analysis found that in executive function tasks, the trained frontoparietal network connectivity strength was significantly greater than that of the control the trained group. Not only affect the resting state brain network structure, will also affect the specific tasks under the state of brain connectivity. We speculate that in bead mental arithmetic training In practice, extensive use of visual space strategy will enhance the corresponding brain regions, thus improving the ability of information processing in these brain regions. And improve the efficiency of information processing in these brain regions and will further affect the related cognitive ability. Therefore, we provide a strong theoretical support of the research from the angle of brain network the network transfer effect between the trained and cognitive ability, help us to further understand the relationship between mental training and cognitive ability.

【学位授予单位】：浙江大学
【学位级别】：博士
【学位授予年份】：2017
【分类号】：R338

【参考文献】