Hi Miriam,
SPM supports either continuous or epoched datasets where the epochs
should be equal in length. So it doesn't make sense to specify time
windows of different length at conversion. The most straightforward
thing to do is to epoch at conversion. If you define your epochs in
such a way that they don't include the breaks it should be OK. The
only downside is you won't be able to filter continuous data.
A different way is to convert several times with different windows and
merge at a later stage after epoching. Then you should be careful to
specify a different output name every time so that the datasets won't
overwrite each other.
Best,
Vladimir
On Tue, Mar 13, 2012 at 2:01 PM, Miriam Klein-Fluegge
<[log in to unmask]> wrote:
> Hi Vladimir,
>
> We have some subjects in which the same recording of our EEG data contains
> several blocks/sessions, and we would like to convert the data to SPM using
> spm_eeg_convert but with a timewindow specifying the breaks between blocks:
> i.e. S.timewindow = [0 707;732 1426;1708 2377;2430 3024];
>
> At the moment, SPM only takes a vector with 2 elements as the timewindow. Is
> there a way to change that or is there some other way of importing the data
> without the block breaks?
>
> Best,
> Miriam & Katie
|