how to use mne to compute non-phase lock power for time frequency analysis

Hi, Iā€™m new to MNE. Iā€™m using ā€˜tfr_array_morletā€™ to compute time frequency power. I am wondering how to compute the non-phase lock power, should I remove ERP as input data before use ā€˜tfr_array_morletā€™ ? And what is the mean of the ā€˜zero_meanā€™ parameter, how should I define its value ? Thanks in advance.

Hello @hillhillll and welcome to the forum!

Iā€™m tagging @cbrnr, who might be able to give you some advice here!

Good luck,
Richard

I usually do not remove phase-locked activity when computing time-frequency power. I recommend using mne.time_frequency.tfr_morlet(), which operates on an Epochs objects instead of an array. In particular, this function will return the inter-trial coherence itc in addition, which contains information on phase-locked activity.

Regarding the zero_mean parameter, I am not entirely sure what it means, but Iā€™d go with the default value of True (interestingly, the default is False for mne.time_frequency.tfr_array_morlet(), but I donā€™t recommend using that anyway).

1 Like