Hi Melissa
>hello spm users,
>
> I am working on a learn/recall paradigm with 5 subjects and have
>identified ten regions of interests and am trying to analyze the hemodynamic
>responses and compare the signal change curve between the tasks and the
>different regions. The way I thought about doing this was to first extract
>the time courses from spm_graph (I would do this after using the VOI
>function in SPM2) and plot them in excel. I have two questions
>concerning this first step to analyzing the hrf:
>
>1) is this an accurate way to go about extracting the time course?
>
>
this is a way .. SPMers would say a good one ..
but I would go for a bit more subjects ..
>2) and after the time course is extracted, is there a good model for
>averaging the time course to make a hemodynamic curve that can be used for
>comparison to other curves?
>
>
to compare/average data from diff subjects you need to normalize the
amplitudes ..you have many ways to do this - by a given value e.g. 100
or the min, the max .. the one I used and found good on raw data was the
average NRG as for each time point you divide by the square root of the
sum of each time point squared (xi = xi / SR(Sum xij`2) ..
Hope this helps
Cyril
|