help
help > RE: Scrubbing and movement regressor covariates
Jun 21, 2016 05:06 PM | Alfonso Nieto-Castanon - Boston University
RE: Scrubbing and movement regressor covariates
Dear Rob,
It is perfectly fine to have all confounding effects in a single file/covariate. The main reason why CONN's default pipeline keeps those as separate files/covariates is just to give users then the choice of separately dealing with movement vs. outlier covariates (e.g. users might want to add first-order derivatives terms for the movement parameters, but not necessarily for the outlier regressors). In your case you mention that you already have 12 movement regressors which probably means that you have already computed the exact terms that you wish to have there so I do not see any issue at all with entering those as a single first-level covariate into CONN.
Hope this helps
Alfonso
Originally posted by Rob McCutcheon:
It is perfectly fine to have all confounding effects in a single file/covariate. The main reason why CONN's default pipeline keeps those as separate files/covariates is just to give users then the choice of separately dealing with movement vs. outlier covariates (e.g. users might want to add first-order derivatives terms for the movement parameters, but not necessarily for the outlier regressors). In your case you mention that you already have 12 movement regressors which probably means that you have already computed the exact terms that you wish to have there so I do not see any issue at all with entering those as a single first-level covariate into CONN.
Hope this helps
Alfonso
Originally posted by Rob McCutcheon:
Dear Alfonso/CONN users,
I have a very basic beginner's question. If one uses the CONN preprocessing pipeline, movement regressors and scrubbing of individual volumes are entered as two separate 1st level covariates.
I have performed preprocessing separately in SPM and have generated a file that has 12 movement regressors, followed by columns indicating which volumes should be scrubbed. Is it all right to have this combined movement regressor+scrubbing file as a single 1st level covariate; or does the regressors file and scrubbing file need to be separate?
Many thanks,
Rob
I have a very basic beginner's question. If one uses the CONN preprocessing pipeline, movement regressors and scrubbing of individual volumes are entered as two separate 1st level covariates.
I have performed preprocessing separately in SPM and have generated a file that has 12 movement regressors, followed by columns indicating which volumes should be scrubbed. Is it all right to have this combined movement regressor+scrubbing file as a single 1st level covariate; or does the regressors file and scrubbing file need to be separate?
Many thanks,
Rob
Threaded View
| Title | Author | Date |
|---|---|---|
| Rob McCutcheon | Jun 21, 2016 | |
| Alfonso Nieto-Castanon | Jun 21, 2016 | |
| Rob McCutcheon | Jun 22, 2016 | |