Department of Physics and Astronomy, Neuroscience Institute, Georgia State University, Atlanta, GA, USA.
School of Biomedical Engineering, Science & Health Systems, Drexel University, Philadelphia, PA, USA.
Neuroimage. 2018 Jul 15;175:460-463. doi: 10.1016/j.neuroimage.2018.04.043. Epub 2018 Apr 21.
In a recent PNAS article, Stokes and Purdon performed numerical simulations to argue that Granger-Geweke causality (GGC) estimation is severely biased, or of high variance, and GGC application to neuroscience is problematic because the GGC measure is independent of 'receiver' dynamics. Here, we use the same simulation examples to show that GGC measures, when properly estimated either via the spectral factorization-enabled nonparametric approach or the VAR-model based parametric approach, do not have the claimed bias and high variance problems. Further, the receiver-independence property of GGC does not present a problem for neuroscience applications. When the nature and context of experimental measurements are taken into consideration, GGC, along with other spectral quantities, yield neurophysiologically interpretable results.
在最近的《美国国家科学院院刊》（PNAS）文章中，Stokes 和 Purdon 通过数值模拟论证了格兰杰-盖维因果关系（GGC）估计存在严重的偏差或高度的方差，并且 GGC 在神经科学中的应用存在问题，因为 GGC 度量与“接收器”动力学无关。在这里，我们使用相同的模拟示例表明，当通过频谱分解启用的非参数方法或基于 VAR 模型的参数方法正确估计 GGC 度量时，GGC 度量不会存在所声称的偏差和高度方差问题。此外，GGC 的接收器独立性对于神经科学应用来说不是问题。当考虑实验测量的性质和背景时，GGC 与其他频谱量一起产生具有神经生理学可解释性的结果。
