EEG Frequency

[nadoona]

Hi,

I was trying to measure the peak frequencies of the EEG waves recorded from the optical area (O1, O2, and Oz). By using FFT, the peak frequency seems to be 48 Hz. Is that figure reasonable? I guess it's so high and I dont know the reason for that. I tried some SSVEP experiments by letting the user focus on some flashing buttons having multiple frequencies (such as 9, 10, 11, 12, 15... Hz). All of them gave 48 Hz as the peak frequency.

Please advise.
Thanks a lot.

[mellinger]

Nadoona,

power line frequency is either 60Hz (Americas) or 50Hz (remaining world).
So I guess that your reported signal maximum is 50Hz power line noise (mains interference), which is a major source of EEG artifacts.

To minimize the effect of mains interference on your experiments, you should
- keep electrode impedances below 10kOhms,
- activate the power line notch filter provided by practically all EEG amplifiers,
- perform experiments in a well-designed, electrically shielded EEG lab,
- use proper grounding of subject and equipment (note that this can be dangerous to the subject if not done properly),
- ignore remaining signals at the power line frequency and its integer multiples.

Dealing with mains interference can be quite difficult and requires some training, so it might be a good idea to ask someone for help who is experienced in measuring EEG.

HTH,
Juergen

[saimrasheed]

Hi,
Being new at BCI, i have couple of questions.
How can we ensure, electrically shielded EEG lab? I can ground my device using an extra electrode but how can we ensure grounding of subject?

thanks

[gschalk]

Nadoona,

You typically do not need to shield/ground the subject. EEG recordings are a very common procedure and as such are described in many books on the subject. I suggest that you either consult somebody with EEG recording experience or to study relevant books.

Gerv