Hey Scott,
In the PPI approach this is taken care of by including the original
time series of your seed region (as well as the psychological
regressor) in your model as a covariate of no interest. Otherwise,
yes, in order to avoid the correlations being driven by task effects,
you would use time-series of the residuals for correlations between
regions, rather than the raw data.
Tobias.
On 2/28/06, Scott Fairhall <[log in to unmask]> wrote:
> Am I correct in thinking that, when assessing functional connectivity with
> the vanila approach, one should also remove the main effects (or 'driving
> effects') of the stimulus to reduce more coincidental (and non-interactive)
> aspects of the correlation a la Haynes et al., 2005; Macaluso et al, 2000)?
>
> Resulting regressions are assessed on the residuals of the main stimulus-
> effect. I am not completely sure how to achieve this (to remove the main
> effect). It may be a simple matter of having a separate regressor for
> the stimulus and the regressor-ROI but this seems a bit too easy.
>
> I would quite like to know how though....
>
> All the best
> Scott
>
--
Tobias Egner, PhD
fMRI Research Center
Columbia University
Neurological Institute, Box 108
710 West 168th Street
New York, NY 10032
Tel: (+1) 212 342 0121
Fax: (+1) 212 342 0855
Email: [log in to unmask]
|