# Z score power spectrum wrt resting phase **URL:** \ **Category:** Support & Discussions\ **Tags:** time-frequency\ **Created:** [November 13, 2022, 3:55am UTC](https://mne.discourse.group/t/z-score-power-spectrum-wrt-resting-phase/5911 "2022-11-13T03:55:41Z")\ **Posts on this page:** 2\ **Page:** 1
**Author:** ![WhosePolter](https://avatars.discourse-cdn.com/v4/letter/w/b4bc9f/32.png) [@WhosePolter](https://mne.discourse.group/u/WhosePolter)\ **Post date:** [November 13, 2022, 3:55am UTC](https://mne.discourse.group/t/z-score-power-spectrum-wrt-resting-phase/5911/1 "2022-11-13T03:55:41Z")
`tfr.plot()` supports a correction mode of ‘zscore’. This is more accurately defined as the `standard z score`. Most researchers actually find z-scoring the power spectrum of interest **wrt a pre-event baseline (gen. the resting phase)** to be a better option, which is known as `baseline z-score`. [This page explains the difference between the two z-scores decently](https://bookdown.org/anta8363/fluoR_bookdown/stand.html). How to execute the latter? In the experiment that I am analyzing, there exist two EEG files for each subject corresponding to two sessions. The first corresponded to rest while the second corresponded to the events of interest. All essential parameters (samp. freq, electrode location, etc.) remained the same across both sessions. Now, I want to z-score the power spectrum of the second session wrt the first session. I also wonder if applying baseline correction on the raw data of the second session will be roughly the same? ---
**Author:** ![WhosePolter](https://avatars.discourse-cdn.com/v4/letter/w/b4bc9f/32.png) [@WhosePolter](https://mne.discourse.group/u/WhosePolter)\ **Post date:** [November 13, 2022, 3:38pm UTC](https://mne.discourse.group/t/z-score-power-spectrum-wrt-resting-phase/5911/2 "2022-11-13T15:38:04Z")
@richard Maybe, you can suggest some way to approach the question?