Skin Conductance Response Related EEG Oscillations-Computational Modeling Evidence

Several lines of evidence suggest a feasibility to define the skin conductance response (SCR) related EEG activity. The problem of the precise time localization of the SCR-related oscillations (SCR-ROs) has been resolved through computational dealing with the SCR signal itself and taking into account the variability of latency of the SCR. The SCR-ROs have been identified as delta-frequency EEG activity with an initial deep negative deflection. The thesis about the existence of the SCR-ROs has been tested through a computational modeling study applying the system identification approach considering the SCR-ROs as the output of the system. The study involved 20 participants and the stimuli for the elicitation of the SCR were slide presentations of short stories from the contemporary literature. Single-trial SCR-ROs have shown to be convenient for the system identification modeling approach. The EEG epochs were 1.5 s long and delays between the SCR-ROs and the respective SCRs were 1.0-4.0 s. Obtained values of the system parameters in the whole sample of the single-trial based models of the SCR-ROs have a modest variability. Validity of the concept of the SCR-ROs was tested by a Monte Carlo analysis at the level of each participant. The difference in average, median, and maximum value of the model fit in the system identification modeling between the true SCR-ROs and fake EEG epochs was significant at the level p < 0.001 in the sense of better fit in the true SCR-related epochs in 17 participants and in three subjects the significance reached the level of p < 0.05. The finding represents a proof of concept for the SCR-ROs and suggests the possibility for a psychophysiological observability of the neural system which is shared by processes which generate both the SCR and the SCR-ROs.