[FieldTrip] using double-dipole source model for reconstruction of auditory phase locking

[Sophie Herbst]

Dear Fieldtrip experts,

I am reconstructing the sources of EEG inter-trial phase consistency in an
auditory paradigm using DICS beamformers.

To achieve better separation of the bilateral auditory cortices, I would
like to use leadfields with pairs of dipoles, where each location is
modeled by a dipole from each hemisphere as suggested by Popov et al.
<https://www.frontiersin.org/articles/10.3389/fnins.2018.00711/full> (section
Frequency Domain Beamforming).

Before, when using the standard single dipole model, I used
ft_sourceanalysis with
cfg.dics.keepfilter = 'yes';
cfg.dics.fixedori = 'yes';
and then projected the single trial analytic time courses (obtained with
the Hilbert transform)  through the filter using matrix multiplication.
Now, I am not sure whether this is the correct thing to do with the double
dipole models, or whether I should get rid of the second dipoles before
projecting the data through the filter.

Could you advise me on how to do that, and how to select the dominant
orientation in that case?

(In the paper above, the ‘pcc’ method is suggested in combination with the
double-dipole models to "to separate at each grid point the ‘scanning
dipole’-of-interest’s from the ‘suppression dipole’-of-no-interest’s
activity, before the computation of the coherence coefficient between the
scanning dipole and the modulation signal". )

Thank you very much in advance,
Sophie

-- 
*Dr. Sophie K. Herbst*
CEA Researcher
NeuroSpin, CEA Saclay, DRF/Joliot
Inserm Cognitive Neuroimaging Unit
Bât 145, Gif s/ Yvette F-91190 FRANCE
Phone: +33 (0)1 6908 7938
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20210922/a46ffe4a/attachment.htm>