Because the PM are mean-centered for each run/condition, there is not
an easy way to adjust for the differences. You could use covariates at
the group level to account for variations. This is perhaps the easiest
as you don't have to worry about the estimate of the PM being the same
or different between runs.
The slope for each run might also be different, so that will be an
added complication to trying to combine the betas as well. You also
need to be cognizant of the stability of the PM term in each run and
how combining them might influence the results.
Best Regards, Donald McLaren
=================
D.G. McLaren, Ph.D.
Research Fellow, Department of Neurology, Massachusetts General Hospital and
Harvard Medical School
Postdoctoral Research Fellow, GRECC, Bedford VA
Website: http://www.martinos.org/~mclaren
Office: (773) 406-2464
=====================
This e-mail contains CONFIDENTIAL INFORMATION which may contain PROTECTED
HEALTHCARE INFORMATION and may also be LEGALLY PRIVILEGED and which is
intended only for the use of the individual or entity named above. If the
reader of the e-mail is not the intended recipient or the employee or agent
responsible for delivering it to the intended recipient, you are hereby
notified that you are in possession of confidential and privileged
information. Any unauthorized use, disclosure, copying or the taking of any
action in reliance on the contents of this information is strictly
prohibited and may be unlawful. If you have received this e-mail
unintentionally, please immediately notify the sender via telephone at (773)
406-2464 or email.
On Wed, May 1, 2013 at 9:51 AM, H. Nebl
<[log in to unmask]> wrote:
> Dear SPM experts,
>
>
> Basically a repost, but as I didn't get any answer so far...
>
> Assume several conditions, each of them parametrically modulated (same objective values across the whole experiment for all these conditions with a range of, say 0 - 10 and a mean of 5). Now some of the trials might have to be excluded due to wrong responses, so some of the values might be missing / underrepresentated. Thus the average value might be different between conditions, say a mean of 5 for condition 1 (as planned) and maybe only 4 for condition 2.
>
> One runs into the same problem when having several runs with the same conditions. The mean of condition 1 / run 1 might be 5, but for the 2nd run it might be something else.
>
> Now, when comparing the standard regressors (or averaging across runs) it seems to me that one should adjust them = bring them to "the same level" by adding the betas of the parametric modulator to the betas of the standard regressor till they correspond to the same mean value. What do you think?
>
>
> Best,
>
> Helmut
|