Electrical stimulation + skin conductance + EEG

[Ryan P.]

Hi,
Our lab has started a new experiment that involves collecting skin conductance data as well as delivering an aversive electrical pulse train that simulates a shock (from biopac) while recording EEG with ActiveII 32 cap plus 8 external TP electrodes. I have attached a rough diagram of our setup below. When we deliver the shock every EEG channel starts a non-uniform slow wave drift that can last for up to 10 seconds (see attached) . 3 of 5 subjects we have tested so far have this issue. We use this setup for 2 out of the 6 tasks in the session. After the 2nd task is complete, we remove the skin conductance and shock electrodes and just record EEG. However, the problem will occur again for a more brief duration (~3-5 seconds) if the subject has a strong blink or a short jaw clench. During the slow wave drifting, the electrode offset bars are fluctuating up and down as well.

The last time this was happening I was able to turn the shock intensity all the way down on the BIOPAC during the shock task and noticed that problem went away when the next stimuli was delivered (a train of 20, 50ms port codes to the BIOPAC/BIOSEMI USB Box via parallel port). I have seen similar images on this forum viewtopic.php?f=1&t=758
but that post was never answered. Please help us understand how this happening or at least how we can rearrange our setup to prevent it. Thanks.

[Coen]

The subject is aroused by the shock and starts to sweat a little. Perspiration disturbs the skin-gel-electrode interface, and it takes some time (tens of seconds) before the interface has settled to a new equilibrium.

Usual recommendation is to keep the subjects cool and comfortable (although the last condition is probably not easy to achieve when the subject is shocked). High-pass filtering with a short time constant suppresses the artifact.

Best regards. Coen (BioSemi)