beamformer source localization for EEG

[Robert Oostenveld]

Dear Hung

The differences between source reconstruction methods, or inverse
methods, are in the assumptions that they make on the sources. Some
make assumptions on the number of sources (dipole fitting),
distribution of activity (minimum norm linear estimation) and some
make assumptions on the correlatedness of the timecourses of the
sources (beamforming or music). A difference is also in wether and how
the noise in the data is incorporated in the estimate of the source
acitivity (ignored in plain old dipole fitting, explicitely suppressed
in beamforming).

The method that works the best is the method whose assumptions on the
data (true sources and noise) best corresponds to the data that you
have acquired.

best regards,
Robert


On 17 Jul 2009, at 17:56, Hung - Gmail wrote:

> Dear all,
> We have known that beamformers and other good source localization
> techniques have zero localization error for single source cases.
> However, in
> real applications there are many sources inside the brain then we will
> have non-zero localization error. How can we justify which inverse
> solutions is better for multiple source cases?
>
> Thanks a lot
> Hung
>
> ----------------------------------
> The aim of this list is to facilitate the discussion between users
> of the FieldTrip  toolbox, to share experiences and to discuss new
> ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html
>  and http://www.ru.nl/neuroimaging/fieldtrip.
>

----------------------------------
The aim of this list is to facilitate the discussion between users of the FieldTrip  toolbox, to share experiences and to discuss new ideas for MEG and EEG analysis. See also http://listserv.surfnet.nl/archives/fieldtrip.html and http://www.ru.nl/neuroimaging/fieldtrip.