Figure 2.

Network activity and analysis approach. (A) The time-line for the EEG analysis comprised 500 s of EEG recording. One hundred seconds of EEG (prior to HFS) are used as baseline (−100 s time-point). The time-point “0’ represents the period during which HFS was applied. The 3 subsequent time-windows reflect EEG recordings 100 s, 200 s and 300 s after HFS 300 s. (B) Raw EEG, in which stimulus artifacts induced by test-pulse afferent stimualtion or HFS can is indicated by vertical lines. (C) From the raw EEG, 10 artifact-free 4.1 s epochs (C) were extracted from every 100 s time window (A). (D) The raw EEG epochs were analyzed as to their relative change relative to baseline in power levels for the frequencies in the delta, theta, alpha, beta and gamma ranges. (E) The principal of inter-frequency phase-to-amplitude-coupling (PAC) detected by envelope-to-signal correlation (ESC) is shown in 1 s of EEG. The resulting correlation score for one single 4.1 s epoch is influenced by the appearance of envelope-to-signal correlations. This means the shape of the high frequency filtered signal-envelope (bottom signal in F) and the shape of the low frequency filtered signal (middle signal in F) are approximately the same. The upper signal in (F) shows the raw EEG signal. The region at the dotted ellipse in the signal shows a typical envelope-to-signal correlation: The shape of the envelope of the fast signal shows approximately the same shape as the amplitude of the compared slow signal. The more often and pronounced this phenomenon appears within an analyzed EEG epoch, the higher will be the overall PAC-ESC score for this epoch. (F) The cosine phases of theta oscillations (black line in F) and the gamma oscillations during these theta cycles (dotted line in F) were extracted by using the Hilbert transformer. These data were then used to determine the peak-to-slope amplitudes of theta cycles and the mean gamma amplitudes during these theta cycles.