0
$\begingroup$

I recorded EEG data of several subjects and for several sessions per subject. Some of the data looks flat as in figure below. But when I looked into the data, the data appears to be okay.I mean it looks like normal EEG data range as in file attached. And also when I plot the power spectrum, it also looks normal. I don't understand the visually flat thing as I haven't changed any settings or normalized it. This is raw data.

enter image description here enter image description here

enter image description here

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

If you plot anything with a large enough y-axis scale it's going to look flat.

Here are the same data, y = sin(x), plotted three times. All I have changed is the y-axis scale:

y=sin(x)

If you always use the same y-scale in whatever software you are using, then it seems like these data are lower amplitude than your normal data. This can be for many reasons, from amplifier settings to electrode impedance. It could also be that settings in your software have changed.

Also, please always plot your power spectra on a log-log scale, linear scale makes no sense for power spectrum.

Improve this answer
$\endgroup$
2
  • $\begingroup$ Thankyou for your answer, I investigated data the way you mentioned and that sort of showed me the signal on different scales. Actually, I am collecting data using emotive Pro, and some channels out of 32 show normal signal, some show flat (which can be visualize by changing the scale) but how come same recording have different behavior? Is that normal? Sorry I am new to this data, might be asking a naive question. Your insights will be really helpful. $\endgroup$
    – Hannan Naseem
    Commented May 29 at 6:11
  • $\begingroup$ @HannanNaseem The signals recorded depend on the properties of the electrodes and their connection with the scalp. If they have a poor connection, are corroded, etc, they're not all going to record the same. I don't think troubleshooting EEG is really something you can do with stack exchange in the long run, do you have an EEG expert you can learn from? $\endgroup$
    – Bryan Krause ♦
    Commented May 29 at 12:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.