Benny

Den onsdagen den 2:e maj 2012 skrev Jeanette Mumford:

Hi,

Depends on if you're publishing it or not. I prefer using percent signal change in papers, since it has more meaning. If you're just looking to see if there's an effect, then it doesn't matter and it might be faster to skip the conversion.

What difficulty were you having with using fslmaths and fslmeants to convert to percent signal change? If you have your ROI's in standard space, this is just 2 lines of code for each ROI. You can email me off this list if you need help choosing the parameters.

Cheers,
Jeanette

On Wed, May 2, 2012 at 10:59 AM, David Soto <[log in to unmask]');" target="_blank">[log in to unmask]> wrote:

Hi,

I was wondering whether it'd be fine to correlate individual copes
to individual performance without converting to % signal change...
following Jeannette M.'s tip for deriving individual cope values using
fslmeants on the filtered_func_data that appears in the higher level analyses.

I am finding it tricky to derive the individual % signal change
without using featquery, which takes a lot of time, and so
wondered this could be a reasonable approach?

guess no?

thanks, David