So, I have continuous EEG data from a control group and a test group. The recording was done while the participants were at rest, so no task was performed.

I want to start with some basic analysis, but I am not really sure how to give the first step. The recordings are ~5hours long. I believe it is best to segment the data. Should I perform the analysis on each segment? How do I combine the data of the whole group and then compare it with the other group?

I just started to work with EEG and none of the papers show a clear "tutorial" on how to do these basic steps. Any papers or books you could recommend would be of great help!


closed as unclear what you're asking by Christian Hummeluhr, user7759, Arnon Weinberg, Josh de Leeuw, AliceD May 10 '15 at 5:23

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • $\begingroup$ just an update: by basic analysis I mean time-frequency, laterality, etc... $\endgroup$ – student1971 Apr 29 '15 at 23:27
  • 2
    $\begingroup$ please add updates as edits to your question $\endgroup$ – Seanny123 Apr 30 '15 at 0:05
  • 4
    $\begingroup$ What's the goal of the analysis? There are a lot of things you could do with EEG data, but what you should do depends on the question(s) you are trying to answer. $\endgroup$ – Josh de Leeuw Apr 30 '15 at 14:08
  • 1
    $\begingroup$ Info on the purpose is critical. If the test group has,eg, epilepsy you may want to look for paroxysmal activity. Otherwise, FFT is likely a good start. $\endgroup$ – AliceD Apr 30 '15 at 20:59
  • $\begingroup$ I agree with the above comments, what is your question? From a data analysis perspective, no, you dont necessarily need to segment your data. An interesting approach would be to use different data reduction methods to explore your data. For instance, global field power is one simple way of reducing your data to only one "channel" (which is also reference free). Another approach is to read the literature on EEG microstates and explore those in your data. $\endgroup$ – blackBerry Jun 30 '15 at 9:16