# Source power in resting state eeg(eyes open/ eyes close)

**URL:** <https://mne.discourse.group/t/source-power-in-resting-state-eeg-eyes-open-eyes-close/7463>\
**Category:** Support & Discussions\
**Tags:** source-localization, beamformer, resting-state\
**Created:** [August 30, 2023, 9:26am UTC](https://mne.discourse.group/t/source-power-in-resting-state-eeg-eyes-open-eyes-close/7463 "2023-08-30T09:26:54Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![payam](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/payam/32/3155_2.png) [@payam](https://mne.discourse.group/u/payam)\
**Post date:** [August 30, 2023, 9:26am UTC](https://mne.discourse.group/t/source-power-in-resting-state-eeg-eyes-open-eyes-close/7463/1 "2023-08-30T09:26:55Z")

</div>

Hello,

Which source localisation method (minimum norm inverse or beamformer) is better for source power mapping for a 2 seconds duration epochs of eyes open/ close (not evokes) in resting state eeg recordings?

Best,  
Payam

---

<div class="post-metadata">

**Author:** ![britta-wstnr](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/britta-wstnr/32/3538_2.png) [@britta-wstnr](https://mne.discourse.group/u/britta-wstnr)\
**Post date:** [October 23, 2023, 1:33pm UTC](https://mne.discourse.group/t/source-power-in-resting-state-eeg-eyes-open-eyes-close/7463/2 "2023-10-23T13:33:22Z")

</div>

Hi @payam,

There is many things to take into consideration when deciding for one source reconstruction method over another. Let me list you some of them:

Beamforming is great if:

- you assume your activity to be focal
- you don’t have activity highly correlated in time across the brain (e.g. two auditory cortices)
- you want to make use of the properties of a spatial filter, e.g. the noise suppression properties

Beamforming is not so great if:

- you do not have individual MRI scans and individual sensor positions _especially_ with EEG

So you see, there is no definitive answer to your question.  
From a practical view, you could try both methods and you should see that the results converge. If they do not converge and especially if you have reason to believe that your forward model is not very accurate, I would trust the MNE over the beamformer.

HTH,  
Britta

---

<div class="post-metadata">

**Author:** ![payam](https://yyz2.discourse-cdn.com/free1/user_avatar/mne.discourse.group/payam/32/3155_2.png) [@payam](https://mne.discourse.group/u/payam)\
**Post date:** [October 23, 2023, 1:52pm UTC](https://mne.discourse.group/t/source-power-in-resting-state-eeg-eyes-open-eyes-close/7463/3 "2023-10-23T13:52:00Z")

</div>

Thank you very much, Britta.