[Quantitative measures of cortical functional connectivity: a state-of-art brief survey]

This par reviews modern approaches to measuring cortical functional and effective connectivity in neurocognitive networks--the large-scale distributed systems of interacting neuronal populations which are thought to underlie the cognitive processing. Two broad classes of methods of connectivity estimation, linear and nonlinear, are discussed. In the class of linear methods, besides the coherence that is routinely used for measuring the strength of functional links, the vector autoregressive modeling of multichannel EEG is discussed in some details. The latter technique allows for estimating both functional and effective connectivity with such measures as directed transfer function (DTF) and direct partial coherence (PDC) which are commonly used in cognitive neuroscience. The impact of volume conduction onto the different estimates of connectivity is considered. The imaginary part of the complex-valued coherence as a way to reduce the artificial influence of volume conduction is also discussed. In the class of nonlinear methods, the Independent Component Analysis and the Transfer entropy as a method of estimation of directed influence are reviewed.