Rotman Research Institute of Baycrest, Toronto, Canada.
J Neurosci Methods. 2009 Oct 30;184(1):152-60. doi: 10.1016/j.jneumeth.2009.07.014. Epub 2009 Jul 21.
Addressing the issue of effective connectivity, this study focuses on effects of indirect connections on inferring stable causal relations: partial transfer entropy. We introduce a Granger causality measure based on a multivariate version of transfer entropy. The statistic takes into account the influence of the rest of the network (environment) on observed coupling between two given nodes. This formalism allows us to quantify, for a specific pathway, the total amount of indirect coupling mediated by the environment. We show that partial transfer entropy is a more sensitive technique to identify robust causal relations than its bivariate equivalent. In addition, we demonstrate the confounding effects of the variation in indirect coupling on the detectability of robust causal links. Finally, we consider the problem of model misspecification and its effect on the robustness of the observed connectivity patterns, showing that misspecifying the model may be an issue even for model-free information-theoretic approach.
针对有效连通性的问题，本研究关注间接连接对推断稳定因果关系的影响：部分传递熵。我们引入了一种基于传递熵多元版本的格兰杰因果度量。该统计量考虑了网络（环境）的其余部分对观察到的两个给定节点之间耦合的影响。这种形式主义允许我们量化特定路径中环境介导的间接耦合的总量。我们表明，部分传递熵是一种比其双变量等价物更敏感的技术，可以识别稳健的因果关系。此外，我们证明了间接耦合的变化对稳健因果关系检测的混淆效应。最后，我们考虑了模型失拟及其对观察到的连通模式稳健性的影响，表明即使对于无模型的信息论方法，模型失拟也可能是一个问题。
