finding significancy from t-test by pwelche.

Hello, I have 11 participants in my experiment with 2 taskswhich the first one is resting state, and the second one is the main task. So, I have these data and I just want to reach a comparison between each participant’s main task and the resting state task in the t-test, and then with performing pwelche to gather some information for my master. (I am an architect and for that, I am bad at the theory of signal processing :grinning:). my question is: how can I do this with mne or/and Python?
thanks in advance

  • MNE version: 1.2.2
  • operating system: Windows 10

t-tests can be found in scipy.stats
Are you referring to the welch method of determining power spectra? If so I believe the command is mne.psd_welch()

1 Like

Hi,
thank you very much. I follow the t-test in scipy.stats.
is it defined in scipy to get an average from all subjects in one file?
yes, it is power spectra. thanks again!

I’m not sure what measure you’re looking at, but typically what you do for a paired (this is if you have resting state and main task for each participant) t-test is to pass it two sequences. One sequence will be the variable of interest for resting state for all subjects and the other will be the variable of interest for main state for all subjects (making sure they’re in the same order).

The power spectra themselves are hard to use, since it’s a signal array and not a single value. You could select either peak frequency or peak amplitude though.

1 Like

Honestly, the analysis is not my task, but I try to provide data for my instructor in neuroscience to analyze, and i am just trying to preprocess the data. thanks a lot. that was useful information.

1 Like