Dear Pinar,
Do you really need to create a condition (ie regressor) per trial? One
would typically group them per level of a factorial design and only
model their average response.
Could the error you observe due to the fact that the onset of the last
trial actually falls after the end of the acquisition so that its
corresponding regressor ends up being a column of 0s?
Best regards,
Guillaume.
On 25/07/18 18:49, Pinar Pnr wrote:
> Hello dear SPM users,
>
> I am running " fMRI 1st level analysis". There are 125 trials in the
> data. I have constructed a .mat file containing onsets, names and
> duration for using in "multiple conditions". At "SPM contrast defining
> step I am trying to define a contrast including all the trials i.e. ones
> (1, 125) but I face this error: "the contrast is invalid". When I change
> the contrast to ones(1,124), it works (right padded with 2 zeros). What
> is the problem with ones (1,125)? And why 2 zeros are added to the
> contrast vector?
>
> Best Regards,
> Pinar
--
Guillaume Flandin, PhD
Wellcome Centre for Human Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|