[FieldTrip] Merging runs with different grad.tra matrix

[Marisa Monika Amalie Birk]

Hi Konstantinos,

Thank you for your response!

I’ve double-checked, and it seems that the number and type of channels
between runs are identical, and there are no missing channels.

I believe that the problem with different grad.tra matrixes occurred
because I rejected different components in each run using
ft_rejectcomponent. For example, in the first run, I might have rejected
component 24, while in the second run, I rejected component 13 and 19.
Therefore the grad.tra matrix changes differently between runs, depending
on which component / how many components I have rejected.

Because of this, using ft_appenddata does not work for me because the grad
fields of the input datasets differ from one another.

So, would you recommend merging the runs before ICA? Does it pose a problem
if the runs that I merge are rank-deficient and have different ranks -
could this significantly impact the output of the ICA?

Thank you again for your time!

Best regards
Marisa


Marisa Birk
PhD Student in Cognitive and Brain Sciences,
Action Recognition and Concepts Group
Center for Mind and Brain Sciences (CIMeC),
University of Trento, Italy
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20241106/319263c5/attachment.htm>