help
help > RE: Unequal RT
Jan 17, 2012 03:01 AM | Alfonso Nieto-Castanon - Boston University
RE: Unequal RT
Hi Sarah
From a 'practical' perspective, one simple way to go around this problem would be to first add only one set of subjects to your project and running the 'setup' and 'preprocessing' steps, then change the RT value and add the rest of the subjects to the project and run again the 'setup' and 'preprocessing' steps (selecting 'No' to the 'Overwrite existing subject results' question to avoid having the first set of subjects re-analyzed). After that, the rest of the analysis steps (first level and second-level analyses) are independent of the RT value so you can safely run those on the entire set of subjects.
Note, nevertheless, that differences in acquisition RT might have a significant effect on connectivity estimates, so there might be differences in connectivity between the two sets of subjects which could potentially confound your analysis results. One possible way to address this would be to add a second-level covariate labeling these two groups of subjects (e.g. dummy coded covariate taking the value 0 for one set of subjects and 1 for the second set of subjects) and entering this as a an additional covariate in some or all of your second-level analyses to correct for potential RT-related effects in your connectivity analyses. Another possible way to look into this issue is to check whether your analysis results are replicated within each subsample (e.g. define two dummy-coded second-level covariates, one for each group of subjects, and select each group separately when running your second-level analyses).
Hope this helps, and let me know if you would like me to further clarify any of this.
Best
Alfonso
Originally posted by Sarah Buetof:
From a 'practical' perspective, one simple way to go around this problem would be to first add only one set of subjects to your project and running the 'setup' and 'preprocessing' steps, then change the RT value and add the rest of the subjects to the project and run again the 'setup' and 'preprocessing' steps (selecting 'No' to the 'Overwrite existing subject results' question to avoid having the first set of subjects re-analyzed). After that, the rest of the analysis steps (first level and second-level analyses) are independent of the RT value so you can safely run those on the entire set of subjects.
Note, nevertheless, that differences in acquisition RT might have a significant effect on connectivity estimates, so there might be differences in connectivity between the two sets of subjects which could potentially confound your analysis results. One possible way to address this would be to add a second-level covariate labeling these two groups of subjects (e.g. dummy coded covariate taking the value 0 for one set of subjects and 1 for the second set of subjects) and entering this as a an additional covariate in some or all of your second-level analyses to correct for potential RT-related effects in your connectivity analyses. Another possible way to look into this issue is to check whether your analysis results are replicated within each subsample (e.g. define two dummy-coded second-level covariates, one for each group of subjects, and select each group separately when running your second-level analyses).
Hope this helps, and let me know if you would like me to further clarify any of this.
Best
Alfonso
Originally posted by Sarah Buetof:
Hey,
Due to various reasons we have unfortunately data acquired with unequal RT-time in a resting state experiment. That means that one part of our sample has been tested with one RT-time and another part with a different one.
Since it is not possible to run the toolbox with two different RTs for one sample I was wondering if you have any suggestions for how to handle that problem.
Greetings!
Due to various reasons we have unfortunately data acquired with unequal RT-time in a resting state experiment. That means that one part of our sample has been tested with one RT-time and another part with a different one.
Since it is not possible to run the toolbox with two different RTs for one sample I was wondering if you have any suggestions for how to handle that problem.
Greetings!
Threaded View
| Title | Author | Date |
|---|---|---|
| Sarah Buetof | Jan 16, 2012 | |
| Alfonso Nieto-Castanon | Jan 17, 2012 | |
| Renzo Torrecuso | Apr 30, 2021 | |
| Sarah Buetof | Jan 18, 2012 | |
| Sarah Buetof | Jan 18, 2012 | |
| Alfonso Nieto-Castanon | Jan 18, 2012 | |
| Sarah Buetof | Jan 19, 2012 | |
| Sarah Buetof | Jan 19, 2012 | |
| Michael Parrish | Mar 14, 2014 | |