Thank you for your quick answer.
So to summarize, in my case I have a contrast condition A > condiiton B made in fMRI to localize the V1 area and a contrast condition B > condition A in fRMI to localize the V2 area. These areas will be then used as priors in the source reconstruction.
I make my source reconstruction independently on these two conditions but in the same inversion which means that in the structure D.other.inv{1, 1}.contrast for example, I will have two cells in the GW field (one corresponding to the source reconstruction with the condition A and the other corresponding to the source reconstruction with the condition B).
Then I can make a contrast on the evoked responses and power (GW) results of the reconstruction sources, like for example GW(condition A) - GW(condition B) in order to have the same contrast as in mri but in the cortical surface (only instead of being done before the source reconstruction it is made after).
In my case I don't make groupe analysis, I work on each subject separately for the moment.
Is the reasoning ok ?
Thank you,
Gaelle Nicolas
----- Mail original -----
De: "Vladimir Litvak" <[log in to unmask]>
À: "Gaelle Nicolas" <[log in to unmask]>
Cc: "spm" <[log in to unmask]>
Envoyé: Jeudi 3 Février 2022 14:37:12
Objet: Re: [SPM] Question about EEG contrast over epochs
Dear Nicolas,
I suggest you take your contrast at the source level i.e. reconstruct
the two conditions separately (but include them in the same inversion)
and then subject your source images to SPM across subjects.
Best,
Vladimir
On Thu, Feb 3, 2022 at 1:27 PM Gaelle Nicolas
<[log in to unmask]> wrote:
>
> Dear everyone,
>
> I have a question about the EEG contrast over epoch. I have 2 conditions, condition A and condition B.
> For my study I need to make 2 contrasts over epochs, the first one :
> - Condition A > Condition B [1 -1] --> to highlight hypotheses 1
> - Condition B > Condition A [1 -1] --> to highlight hypotheses 2
>
> However these two contrast created the same single differential evoked response (only the signe change, which is normal) but in my case it is a problem because then I compute EEG source reconstruction on this single differential evoked response and I have the same source reconstruction results for the two hypotheses (which is not good for me).
>
> To overcome this I could do a contrast like [1 0] but it's not a good solution because later I will compared these results with another source algorithm for which I make contrasts between condition A and condition B like explained before.
>
> Is there a way (other than doing a contrast over epochs [1 -1]) to maximize the signal in condition A compared to condition B and conversely ?
> In order to have different single differential evoked response for each hypotheses and then different source reconstruction results.
>
> Thank you in advance for your response,
> Gaëlle Nicolas
>
> --
> Gaëlle Nicolas
> PhD Student
> GIPSA LAB/Equipe ViBS
> 11 rue des Mathématiques, 38402 Saint Martin D'Hères
> GIN/Equipe Neuroimagerie fonctionnelle et perfusion cérébrale
> Bâtiment E.J. Safran, Chemin Fortuné Ferrini, 38700 La Tronche
>
--
Gaëlle Nicolas
PhD Student
GIPSA LAB/Equipe ViBS
11 rue des Mathématiques, 38402 Saint Martin D'Hères
GIN/Equipe Neuroimagerie fonctionnelle et perfusion cérébrale
Bâtiment E.J. Safran, Chemin Fortuné Ferrini, 38700 La Tronche
|