Switch to HD for best viewing experience.
In this clip Vasso and Cat demonstrate the common artefacts during EEG recordings. It is important to critically evaluate the data you've recorded. In the worst case it could have just been muscle activity. See the plot below which shows EEG from the forehead relaxed and smiling.
The positive electrode was placed at Fp1, the negative at A1 and GND at A2. Same setup as shown in the video clip.
To investigate further how much the Fp1 signal is polluted with EMG artefacts we recorded in addition between the chin and A1 which certainly should only contain EMG but no EEG. What you see on the screen are two channels: the first from Fp1 (EEG+EMG) and the second one from the chin (EMG only). Note that even at rest these amplitudes are roughly the same meaning that the EEG amplitude is probably much lower and buried in the EMG. To evaluate if it is possible to remove the EMG from the Fp1 channel we calculated the cross correlation between them. Final year project by Thomas Warrington. Many thanks to the very patient subject!
The first two references describe the EMG contamination of the EEG and then the following references discuss how this contamination affects brain computer interfaces and depth of anesthesia monitors (bispectral index). Both applications can be fooled into taking EMG for EEG with quite disturbing results.
Next clip: Using dry EEG electrodes