Dear Pegah,
On Thu, Jun 9, 2011 at 10:34 PM, Tayaranian Hosseini P.
<[log in to unmask]> wrote:
>
> Dear SPMers,
>
> 1. When I coregister the data manually in GUI (using prepare function) and load the coregistered data in source reconstruction window, my source reconstruction results defer hugely from when I coregister the data inside the source reconstruction window. Do you have any idea why? Can it be because I am not asked about any template in former case? If so, how can I assign a template to my data in former case. The problem is I do not know which reconstruction I should pick as the correct one.
>
> I have attached the figures from the two different cases here.
There must be some difference between the two inversions but I doubt
this difference is in the way you did coregistration. Prepare and 3D
GUI call the same function and basically do the same thing. In
principle you should use the 3D GUI because coregistration in Prepare
is just to get sensors properly aligned for projection to 2D.
What you can do is load both kinds of solutions to the 3D window,
click the 'display' button under the 'Coregistration' button and check
for any differences. If you don't find any the difference must be
elsewhere.
>
> 2. Can you also explain what the colorful window at the bottom of "induced source reconstructed" figure is? I thought it should be a hanning window (my data is epoched and I choose a 0-120Hz frequency window and the whole time window).
>
>
If you want to get the Gaussian window leave the frequency band at
default (0). What you did generates wavelet projectors to summarize
the power of the response (basically like computing RMS) and it is not
the same as averaging over a window.
Best,
Vladimir
> Best,
> Pegah
>
>
|