Hi,
Ive 40 session1 datasets that i want to compare with 40 session2 datasets
using Group melodic. Ive the following questions:
1.both sessions have exactly same paradigm but i want to see if response in
the 2 sessions is the same. to do this i specified the 80 datasets in group
ICA and specified the appropriate time series model and group paired
comparison model for post hoc analysis. but what kind of analysis should i
use whether i should use temporal concat. or tensor ICA.
2. when i ran temporal concat.analysis, the individual prestats and
registration was successful for each dataset..but there was an error when
PCA starts, Data size: 3120 x 38978
starting PCA...terminate called after throwing an instance of 'std:bad_alloc'
what(): St9bad_alloc
earlier i had used a smaller dataset and i didnt get this error. so i was
wondering if this is another samplesize-related issue just like it was in
the case of 3rd level FEAT.
3. Once prestats and registration is done on individual datasets i.e it
creates .ica directories for each dataset, how can i re-run a group analysis
without having to rerun the individual pre-stats and registration again.
any help would be appreciated. Thanks!
-vish
|