No, the normalisation is across the window(s) for which you generate
images. That's why if you want to put several time windows in the same
stats design you should specify them all together when generating
images. The scaling is actually slightly more complicated than that as
it only takes into account active sources. See lines 90-119 of
spm_eeg_inv_Mesh2Voxels if you want to know the details.
Best,
Vladimir
On Thu, Feb 7, 2013 at 3:54 PM, Robin Weiss <[log in to unmask]> wrote:
> Vladimir,
>
> Thanks for the info. Just wanted to confirm I'm understanding correctly.
> In viewing the results of an EEG source localization as an image, the data
> represented in red is source power normalized to the sum of power across the
> entire time period over which the source localization was performed. It is
> essentially an average of brain activity over the time period. Is this
> accurate?
>
> Thanks,
> Robin
>
> --
> Robin M. Weiss
> Research Programmer
> Research Computing Center
> The University of Chicago
> 6030 S. Ellis Ave., Suite 289C
> Chicago, IL 60637
> 773.702.9030
> [log in to unmask]
>
> From: Vladimir Litvak <[log in to unmask]>
> Reply-To: Vladimir Litvak <[log in to unmask]>
> Date: Sat, 8 Dec 2012 07:40:11 +0200
> To: <[log in to unmask]>
> Subject: Re: [SPM] Inversion result question
>
> Dear Robin,
>
> The data is source power normalised to the sum of power across conditions
> and vertices. It doesn't have meaningful physical units.
>
> During the inversion the data are reduced by projecting them to a number of
> spatial and temporal modes. Those mode numbers are given in the GUI table.
>
> Best,
>
> Vladimir
>
> On 7 Dec 2012, at 19:10, Robin Weiss <[log in to unmask]> wrote:
>
> Hello,
>
> I was hoping someone could clarify a couple of things about the results from
> EEG source localization in SPM.
>
> First, after performing an localization with the imaging technique and you
> view the results as an image, what is the scale/units associated with the
> range of colors? Please excuse the "technical" jargon, but when you view
> the image you get a picture of the head with red blobs. What are the units
> associated with the red blobs?
>
> Second, when looking at the inversion results from the Display->M/EEG tool,
> there is a table of information including fields such as "label" "date"
> "modality", etc. What does the field "#modes" contain?
>
> Thanks,
> Robin
>
> --
> Robin M. Weiss
> Research Programmer
> Research Computing Center
> The University of Chicago
> 6030 S. Ellis Ave., Suite 289C
> Chicago, IL 60637
> [log in to unmask]
> 773.702.9030
|