 |
 |
Dear Veronica There is no reason why the EBB should have any bias towards the centre of the head (source power estimates are normalized by weight magnitude). I would be excited by these findings. Especially if they are persistent (i.e. you get them in independent time windows and in different subjects). One possibility however is that the main signal generators are missed as they are highly correlated in time (eg auditory evoked responses in left and right cortex). In order to check out this (or which imaging assumptions best suit your data). I would compare model evidence values for MSP and IID or COH (i..e same data just different algorithms). Best Gareth -----Original Message----- From: SPM (Statistical Parametric Mapping) [mailto:[log in to unmask]] On Behalf Of Veronica Y Sent: 27 May 2015 23:10 To: [log in to unmask] Subject: [SPM] MEG SPM12 Beamformer Issue Dear Vladimir and SPM users, I am using the beamformer (EBB) provided by SPM12 as a source reconstruction method for my MEG results (sliding time window of 50ms from 100-600 post-stimulus onset). I am getting a strong, widespread, and long-lasting (from 100-600ms after stimulus presentation) activation (pcorr < 0.05) centred in the thalamus (see attached), which makes me think that beamforming is failing (spurious activation). However, for each individual source reconstruction, the variance explained is >85%, and the log evidence ranges from -9000 to -12000. Are these values expected in case of beamformer failure? If not, does anyone have any ideas what might be the issue? Thank you very much for your help, Veronica