Hi all, I’m trying to calculate the phase locking value for some EEG data I have. I’m following this tutorial.
I’m running into an issue. My data is in .csv format, but so far I have been able to use it. I created a numpy array and a raw array to get it formatted correctly and it seems to be working. For this tutorial though, I need to use inverse operators and an inverse ‘.fif’ file. See below from the tutorial:
# Setup for reading the raw data
raw = io.read_raw_fif(filename)
events = mne.find_events(raw, stim_channel="STI014")
inverse_operator = read_inverse_operator(filename_inv)
To be honest, I don’t even know what an inverse operator is and why I need it. Looking at the tutorial writer’s code output, they have 2 different files being used here. Raw data in a .fif file and another .fif file that is “inverse”. Is there something I can use in python to create my own version of “inverse data” from my MNE raw array?
This tutorial addresses the phase lock in source space, which represents the sources of the EEG in the brain, thus not in the sensor space (EEG). If you are looking for examples of connectivity for EEG, you might want to have a look at the following examples: Examples — MNE-Connectivity 0.7.0 documentation
Thank you, the tutorials you linked are helpful. My main issue is that I would like to create a figure similar to what was in the link I posted, a sort of “heat map” average of trials. Similar to the the graph shown in one of the tutorials you linked, but again as a heat map instead of a line graph.
You can get the time-frequency PLV results using the spectral_connectivity_epochs() function with method="plv", mode="cwt_morlet", and whatever connectivity indices you are interested in.
This will return a SpectroTemporalConnectivity object with results of shape (connections x frequencies x times). The get_data() method will return this as an array.
You can then use imshow like in the example you linked to plot this.