Dear Hector,
There are no fundamental differences and this is more of a consequence
of the generic implementation of second level models in SPM than
anything else. You can add covariates in either of the two entries and
the design matrices will be identical (therefore identical results). The
interaction entry is meaningless for one-sample or multiple regression
designs.
Best regards,
Guillaume.
On 20/04/17 15:48, Hector Espinos wrote:
> Dear SPM users,
> I have a little doubt about the use of covariates with multiple
> regression.
> In spm Batch Editor, in the SPM Statistics, in Factorial design module,
> when you choice a multiple regression design you put your scans and
> after there is a covariates label.
>
> Further ahead, there is another label with Covariates. The unique
> difference between the two options is that this last have and
> interactions options.
>
> I get the same result using the first covariates tag that the second
> one, when the latter I put "interations none".
>
> Although the result is the same, I would like to know what difference
> there is between the two and what would be the correct way to do it.
>
> Thank you very much in advance!!
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|