[Mne_analysis] Combine planar gradiometers after TFR

[Alexandre Gramfort]

        External Email - Use Caution        

hi Geoff,

if your TFR are contain power you have already only positive values. You sum
gradiometers you can have cancelations of the effect due to polarity? does it
answer your question? Now this suggests that if TFR is already power yes summing
is enough

Alex

On Fri, Jan 31, 2020 at 3:09 PM Geoff Brookshire
<G.Brookshire at bham.ac.uk> wrote:
>
>         External Email - Use Caution
>
> Hi all,
>
> I'm currently running a time-frequency analysis, and I'm having some trouble with power.plot_topomap. I'm trying to plot the difference between two conditions, but the plotting function only shows positive values when I try to plot gradiometers. According to the documentation, this function combines them using RMS:
> ch_type : 'mag' | 'grad' | 'planar1' | 'planar2' | 'eeg' | None
>     The channel type to plot. For 'grad', the gradiometers are
>     collected in pairs and the RMS for each pair is plotted.
>
> My understanding, however, is that this is not the correct behavior for TFRs. Since TFRs no longer include information about the direction of the dipole, wouldn't you want to combine planar gradiometers by taking their sum? Taking the RMS after computing TFRs causes me to lose the direction of the effect. This site explains a similar analysis in Fieldtrip: http://www.fieldtriptoolbox.org/example/combineplanar_pipelineorder/
>
> Is there a way to combine planar gradiometers by taking the sum of each channel instead of the RMS? This would be analogous to the Fieldtrip function ft_combineplanar.
>
> Alternatively, is there a different way to plot the difference between two TFRs over gradiometers, preserving decreases in power?
>
> Thanks!
> geoff
> _______________________________________________
> Mne_analysis mailing list
> Mne_analysis at nmr.mgh.harvard.edu
> https://mail.nmr.mgh.harvard.edu/mailman/listinfo/mne_analysis