Populations of tightly coupled neurons: the RGC/LGN system.
Populations of tightly coupled neurons: the RGC/LGN system.
复制标题
紧密耦合的神经元群：RGC/LGN 系统。
DOI：
10.1162/neco.2007.03-07-482
复制
发表时间：
2008
影响因子：
2.9
通讯作者：
Sirovich,Lawrence
中科院分区：
文献类型：
--
作者：
Sirovich,Lawrence
提出了一种描述耦合神经振子动态特性的通用数学模型。所采用的群体方法同样适用于偶联细胞以及这种偶联细胞的群体。该公式包括随机性，并保留了精确发射神经元的细节。基于普遍接受的观点，皮层布线，该配方适用于视网膜神经节细胞（RGC）/外侧膝状体核（LGN）中继细胞系统，早期哺乳动物的视觉系统。在视网膜水平的量子电压跳跃的小允许RGC贡献的福克-普朗克近似;然而，LGN描述需要使用有限的跳跃，快速突触动力学表现为膜电位的有限跳跃。分析了确定性和随机情况下的平衡尖峰行为。绿色的功能方法形成的基础上提出的渐近和精确的结果。这确定了尖峰比率（即，（即每一个本地生人数高峰期的研资局来港人数），即调迁比率的倒数。标准的扣球制度，在相对较少的参数的模型，提出。在合理的假设下，在没有其他地区输入的情况下，转移率≤1/2。因此，该模型表明，LGN/RGC系统可能是一个相对简单的尖峰编辑器。在没有其他输入的情况下，系统被设计为仅当两个或更多个RGC尖峰在相对短的时间内出现时才触发LGN尖峰。在实验室中已经记录了短暂超过1/2（但小于1）的转移比。包含脑干输入已被证明提供了一个信号，提高了传输率（Ozaki和Kaplan，2006年）。一个模型，包括这种贡献。
A mathematical model, of general character for the dynamic description of coupled neural oscillators is presented. The population approach that is employed applies equally to coupled cells as to populations of such coupled cells. The formulation includes stochasticity and preserves details of precisely firing neurons. Based on the generally accepted view of cortical wiring, this formulation is applied to the retinal ganglion cell (RGC)/lateral geniculate nucleus (LGN) relay cell system, of the early mammalian visual system. The smallness of quantal voltage jumps at the retinal level permits a Fokker-Planck approximation for the RGC contribution; however, the LGN description requires the use of finite jumps, which for fast synaptic dynamics appears as finite jumps in the membrane potential. Analyses of equilibrium spiking behavior for both the deterministic and stochastic cases are presented. Green's function methods form the basis for the asymptotic and exact results that are presented. This determines the spiking ratio (i.e., the number of RGC arrivals per LGN spike), which is the reciprocal of the transfer ratio, under wide circumstances. Criteria for spiking regimes, in terms of the relatively few parameters of the model, are presented. Under reasonable hypotheses, it is shown that the transfer ratio is ≤1/2, in the absence of input from other areas. Thus, the model suggests that the LGN/RGC system may be a relatively unsophisticated spike editor. In the absence of other input, the system is designed to fire an LGN spike only when two or more RGC spikes appear in a relatively short time. Transfer ratios that briefly exceed 1/2 (but are less than 1) have been recorded in the laboratory. Inclusion of brain stem input has been shown to provide a signal that elevates the transfer ratio (Ozaki & Kaplan, 2006). A model that includes this contribution is also presented.