Whole brain activates during source reconstruction

  • MNE version: 1.3.1
  • operating system: macOS Sonoma 14.0 (23A344)


We are trying to do source reconstruction on MEG data but get confusing results.

We generate a source estimate for an evoked object. At certain time points, the whole brain is activated, which seems odd. See below for an example:

At the sensor level, the evoked seems correct:

This is the coregistration, which also seems fine:

This problem is recurrent for several sessions and subjects.

Has anyone faced similar problem before, or would have any idea of where the problem would come from?

Thank you very much for your help.