Dear Nela,

There is no straightforward way to restrict the solution to one
hemisphere. However:

1) Perhaps what you should restrict is not the solution but the
statistics you do afterwards i.e. you can use a mask or do a small
volume correction. If you are looking at early responses there
shouldn't be a problem to localize them correctly without restriction
if they are indeed unilateral. But leaving the sources in the other
hemisphere might help you absorb some activity of no interest.

2) Even if you do want to restrict the solution it doesn't make much
sense to restrict it to the whole hemisphere but you are more likely
to be interested in the early visual areas. For that the functionality
provided in 'Restrict solutions' is OK. You should just provide a
matrix Nx3 with MNI coordinates of these areas in the rows, save it in
mat file (the variable name can be anything), load that mat file when
SPM asks for it and define the radius of the spheres around those
locations. By including enough points and playing with the radius you
can restrict the solution to any area you like, but as I sad you
should first check whether there is any need for restriction in the
first place.

Best,

Vladimir

On Tue, Jan 12, 2010 at 12:35 PM, Nela Cicmil <[log in to unmask]> wrote:
> Thank you for your help on the previous problem, I hope that you could help me with another:
>
> For MEG 3D source reconstruction of an evoked visual response, during the inverse step I would like to functionally constrain the solutions to one hemisphere only (i.e. the contralateral hemisphere to the visual stimulus, since I expect an early visual response). I am using the Template, not subject's own MRI, for 3D reconstruction.
>
> In the SPM8 manual, it suggests to create a matlab .mat file with a K x 3 matrix containing MNI coordinates of the areas of interest. However, I don't know how to create this MNI .mat file and I was wondering whether there are any tips for doing it? If one obtains all the MNI coordinates (I assume 3D matrix) into the .mat file, then is there a simple way to mask out one hemisphere?
>
> I am using SPM8 r3408.
>
> Thank you for your help,
> Nela
>
> ________________________________________
> From: Vladimir Litvak [[log in to unmask]]
> Sent: 10 January 2010 05:16
> To: Nela Cicmil
> Cc: [log in to unmask]
> Subject: Re: [SPM] combine two datasets into one
>
> Dear Nela,
>
> You should use Other/Merge (spm_eeg_merge).
>
> Vladimir
>
> On 9 Jan 2010, at 22:07, Nela Cicmil wrote:
>
>> Dear all,
>>
>> I have a simple formatting question but as a newbie to SPM8 I would
>> be really grateful if someone could help me out.
>>
>> I have two datasets (fif files) of MEG averaged evoked responses,
>> which have been averaged outside of SPM8 (in xplotter). Each dataset
>> has one condition i.e. trial type - I would like to compare between
>> these conditions.
>>
>> These fif format datasets are easily converted into SPM8 format
>> datasets by using Convert: however, is there a simple way to combine
>> these two datasets which have one condition each into one dataset
>> that has two conditions in it?
>>
>> This would allow me to inverse the two conditions together during 3D
>> source reconstruction and statistics, which I don't think that I can
>> do when they are in separate datasets.
>>
>> Thank you,
>> Nela
>>
>> ----
>> Nela Cicmil
>> D.Phil Candidate, Neurophysiology
>> Dept. Physiology, Anatomy & Genetics
>> University of Oxford
>