Revealing direction of coupling between neuronal oscillators from time series: phase dynamics modeling versus partial directed coherence.
Revealing direction of coupling between neuronal oscillators from time series: phase dynamics modeling versus partial directed coherence.
复制标题
从时间序列揭示神经元振荡器之间的耦合方向：相位动力学建模与部分定向相干。
作者：
D. Smirnov;B. Schelter;M. Winterhalder;J. Timmer
确定神经元振荡器之间的定向耦合从他们的时间序列的问题得到解决。我们比较性能的两个行之有效的方法：部分定向相干和相动力学建模。它们分别代表线性和非线性时间序列分析技术。在数值实验中，我们发现它们在适当的条件下都是适用的和上级的：如果观察到的行为“更接近”极限环动力学，后者是上级的技术，前者是更好的情况下，更接近线性随机过程。
The problem of determining directional coupling between neuronal oscillators from their time series is addressed. We compare performance of the two well-established approaches: partial directed coherence and phase dynamics modeling. They represent linear and nonlinear time series analysis techniques, respectively. In numerical experiments, we found each of them to be applicable and superior under appropriate conditions: The latter technique is superior if the observed behavior is "closer" to limit-cycle dynamics, the former is better in cases that are closer to linear stochastic processes.