Table 1 Overview of included EEG characteristics and their descriptions.
From: Reliability and state-dependency of EEG connectivity, complexity and network characteristics
EEG characteristics | Description |
|---|---|
Functional connectivity | The AECc was obtained by estimating the amplitude of the analytic signal for two signals, and subsequently computing the correlation between their envelopes. Normalized values range from 0 to 1. See for further AECc details Supplement S2. |
The PLI characterizes consistent phase differences between two signals, ignoring zero-lag correlations. Ranges vary from 0 (no synchronization, or perfect zero-lag correlation) to 1 (complete non-zero-lag phase locking). | |
The PLIMST and the AECcMST represent the average of a functional connectivity metric that is calculated over the edges of the MST instead of using a whole-brain average. | |
Inverted joint permutation entropy (JPEINV) | JPE quantifies the shared complexity between two multivariate time series by analyzing their joint symbolic dynamics, providing a normalized measure of synchronization or independence between the systems. Inverting the JPE means higher JPEINV values reflect higher functional connectivity. |
Permutation entropy (PE) | PE quantifies the complexity of neural activity by analyzing the order of consecutive data points in a time series. It measures the unpredictability of distinct patterns formed by ranking these points, reflecting the level of local brain activity complexity. PE was included since it is necessary to optimally interpret JPEINV results, and to provide a local measure of signal complexity. |
MST characteristics | |
Degree (k) | Measures the number of edges/links for each node divided by the maximum number of edges possible. The maximum degree (kmax), which is the highest degree in the MST, was used as a macroscale network characteristic. |
Leaf fraction (LF) | The ratio of the number of leaf nodes (L) divided by the maximum possible leaf number (which is equal to the number of nodes − 1). A leaf node is a node with only one edge. |
Diameter (D) | Refers to the largest distance between any two nodes in the MST backbone. It can be interpreted as a measure of efficiency, where a low diameter indicates an efficient information flow between brainregions. |
Betweenness centrality (BC) | Fraction of all shortest paths that pass through a node. A leaf node has a BC of zero. The central node in a star-like network, is characterized by BC = 1. For the MST global measure, the highest BC (BCmax) is used. |
Eccentricity (ECC) | The eccentricity of a node is defined as the length of the longest of the shortest paths from this node to any other node. Here, we used the mean ECC of all nodes. |
Tree hierarchy (Th) | Defines the hierarchy of the MST organization as optimal topology. Th is calculated as Th = L/ (2 m BCmax), where m = number of edges. |
MST overlap | To estimate not just MST-derived metric reliability, but also reliability of the global topology of the MST, MST overlap was analyzed. The MST overlap was computed by calculating the overlapping edges of individual MSTs, with the MSTs based on the average connectivity matrices for two recordings or conditions for each subject. Then, the mean overlap was analyzed across the different conditions. |
