B1238
Title: Spectral causation entropy
Authors: Paolo Victor Redondo - King Abdullah University of Science and Technology (Saudi Arabia) [presenting]
Raphael Huser - King Abdullah University of Science and Technology (Saudi Arabia)
Hernando Ombao - King Abdullah University of Science and Technology (KAUST) (Saudi Arabia)
Abstract: Given several nodes in a brain network, functional connectivity describes the causal relationship between processes recorded at different regions. However, existing methods construct such connectivity mapping by considering pairwise analysis, which neglects the contribution of other network components and is unable to differentiate direct from indirect causal structures. Causation entropy, an information-theoretic causal measure, quantifies the magnitude and direction of information flow between two processes after considering all other processes in the network. To associate a derived network with established findings in cognitive science, a new spectral causal metric, spectral causation entropy (SCE), is developed that measures the direct causal impact between network nodes in the frequency domain. An efficient estimation approach is proposed based on combining copula theory and dimension reduction techniques for time series. A novel contribution is a simple and straightforward assessment of uncertainty via a resampling scheme, which allows adjustments for multiple comparisons. Based on SCE, summarizing significant direct information flow from all node pairs in the network results in the derivation of the spectral functional connectivity graph. Lastly, the performance of the proposed measure is demonstrated through numerical experiments, and interesting findings are reported on the analysis of electroencephalogram (EEG) recordings linked to a motor task.
