Analysis of the neural circuit related to sensory discrimination based on multi-dipole estimation with high temporal resolution

尽管新的脑血流和代谢测量成像方法，如PET和fMRI，提供了在感觉辨别过程中活跃的大脑区域的详细解剖图像，但这些方法无法以足够高的时间分辨率揭示活跃区域的时间过程。另一方面，通过记录活跃的神经细胞群产生的神经磁场，可以以毫秒级的分辨率测量大脑活动模式的时间过程。此外，最近开发的全皮质MEG系统使我们能够一次测量所有大脑区域，而无需重新定位来记录每个区域。本研究的目的是建立具有足够高时间分辨率的多偶极子半自动化估计系统。此外，利用s…More分析系统研究了与感觉辨别相关的神经活动的时间结构。多偶极子估计算法基于以下三个基本要点：1)在向内极大值和向外极大值对的选择中，首先选择一个半球内最邻近的对作为估计的第一步。2) 4、3和2偶极子的估计误差分别小于10%、15%和20%。3)基于神经元活动的时空连续性和稳定性进行多偶极子估计。通过使用该系统，观察到以下与感觉辨别相关的神经元活动：1)其特定模态区域的活动。2)与注意力有关的区域，如前额叶区、扣带回等。3)背外侧前额叶皮层、海马、顶叶联合皮层和颞叶联合皮层等与记忆相关的区域。这些活动在开始时是按照上述顺序进行估计的，之后是并行的。这些脑区可能构建了感官辨别所需的工作记忆系统。少

Although new imaging methods for measuring regional cerebral blood flow and metabolism such as PET and fMRI provide a detailed anatomical picture of the brain regions that are active during an sensory discrimination process, these methods fail to reveal the time course of the active regions with sufficiently high temporal resolution. On the other hand, the time course of brain activity patterns can be measured at the millisecond level of resolution through recordings of the neuromagnetic fields that are generated by active populations of nerve cells. Furthermore, recently developed whole-cortex MEG systems enabled us to measure all the brain areas at a time, without repositioning to record each area. The purpose of the present study was to construct the system for semi-automatically estimating multiple dipoles with sufficiently high temporal resolution. Furthermore, the temporal structure of the neural activity related to sensory discrimination was investigated by using the analyzing s … More ystem.The algorithm for multi-dipole estimation is based on the following three basic points: 1) In selection of the pair of the inward maximum and the outward maximum, the most neighboring pair in one hemisphere is to be selected as the first step for estimation. 2) Estimation errors are less than 10%, 15% and 20% for 4, 3 and 2 dipoles respectively. 3) Multi-dipole estimation is carried out based on spatio-temporal continuity and stability of the neuronal activities.By using the system, the following neuronal activities were observed related to sensory discrimination: 1) activity in its modality-specific area. 2) the areas related to attention such as prefrontal area, cingulate gyrus, and so on. 3) the areas related to memory such as dorsolateral prefrontal cortex, hippocampus, parietal association cortex and temporal association cortex. These activities were estimated following the above order during the beginning, and in parallel after. These brain areas may construct the working memory system required for sensory discrimination. Less