Algorithm 1: Calculation of a causal matrix in the source space.

Input: A sample including EEG source signals $S^{\prime }=[s_1^{\prime }(t),s_2^{\prime }(t),\cdots ,s_c^{\prime }(t)]\in {ℝ}^{r\times c}$ Output: $G_s$ 1: for $i=1,\cdots ,c$, do 2:       for $j=1,\cdots ,c$, do 3:             Calculate the residual ${\xi }_1$ in autoregressive model by Equation (4). 4:             Calculate the residual ${\eta }_1$ in vector regression model by Equation (5). 5:             Calculate the Granger causality between the $i^{\mathrm{th}}$ and $j^{\mathrm{th}}$ EEG source sign by Equation (6). 6:       end for 7: end for