[FieldTrip] beamformer to localize spontaneous oscillation source

["Jörn M. Horschig"]

Dear Weiwei,

as an alternative to computing the NAI (as Eelke pointed to), you can 
normalize the leadfield by settings cfg.normalize = 'yes' prior to 
calling ft_compute_leadfield. I've heard from high places that this 
works pretty well, but haven't compared both methods myself.

Best,
Jörn

Eelke Spaak wrote:
> Dear Weiwei,
>
> After using cfg.dics.projectnoise = 'yes', did you also compute the
> "Neural Activity Index" by computing source.avg.pow = source.avg.pow
> ./ source.avg.noise? If not, the noise estimate remains unused.
>
> Best,
> Eelke
>
> On 19 November 2013 13:50, PWW <ww.peng0923 at gmail.com> wrote:
>   
>> Dear Fieldtrippers,
>>
>> In the beamformer tutorial
>> (http://fieldtrip.fcdonders.nl/tutorial/beamformer), the contrast of
>> (post-pre)/pre was calculated to circumvent the  noise bias towards the
>> center of the head.
>>
>> However, I wonder whether it is possible to only localize the pre-stimulus
>> oscillation source, such as pre-stimulus alpha power for example, just like
>> localizing the sources of spontaneous oscillatory activity. Actually, I
>> tried to do it, and use neural activity index to remove the center of the
>> head bias, by setting setting cfg.dics.projectnoise='yes'. But it doesn't
>> work so well.
>>
>> Does anyone have any suggestions about localizing the cortical sources of
>> spontaneous oscillatory activity (e.g. pre-stimulus interval only)? is it
>> any difference with the procedures described in the tutorial?
>>
>> Any pointers would be much appreciated.
>>
>> Thanks.
>>
>> Regards,
>>
>> Weiwei
>>
>>
>>
>> _______________________________________________
>> fieldtrip mailing list
>> fieldtrip at donders.ru.nl
>> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>>     
> _______________________________________________
> fieldtrip mailing list
> fieldtrip at donders.ru.nl
> http://mailman.science.ru.nl/mailman/listinfo/fieldtrip
>