Hello, I would like to know why we are limited to 4 regressors in the design specification within the QDEC GUI. If I understand correctly, there is no such limitation in the "old" method (via command line). Thanks, - Jerry
Jerry,
It boils down to a GUI issue, that we are working to eliminate in the next qdec release (that is, eliminate the limitation of four regressors).
The problem is that qdec currently automatically generates all possible contrast matrices for the specified up-to 4 regressors. The questions associated with each contrast are also automatically generated. Beyond four regressors, you begin to have enormous numbers of possibilities (hundreds to thousands), which makes automatic generation, calculation, and display, of all possibilities too unwieldy.
The next qdec will include a Contrast tab, so that prior to running the design, you would select from a list the contrasts that you want to keep (and of those, only the small ones will have the human-readable questions associated with it). Until then, you would have to run via the command-line (ie create your own contrast matrices) to make use of more than four regressors.
Nick
On Tue, 2008-03-18 at 14:10 -0400, Jerry Yeou-Wei Chen wrote:
Hello, I would like to know why we are limited to 4 regressors in the design specification within the QDEC GUI. If I understand correctly, there is no such limitation in the "old" method (via command line). Thanks,
- Jerry
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
While on the topic of future qdec versions: Is there any plan to develop a qdec that would support simultaneous analysis of the left and right hemisphere in a repeated measure type framework? (e.g., so that you could easily create statistical maps of the main effect of group, the main effect of hemisphere, and group x hemisphere interaction).
thanks, Mike H.
On Tue, 2008-03-18 at 14:46 -0400, Nick Schmansky wrote:
Jerry,
It boils down to a GUI issue, that we are working to eliminate in the next qdec release (that is, eliminate the limitation of four regressors).
The problem is that qdec currently automatically generates all possible contrast matrices for the specified up-to 4 regressors. The questions associated with each contrast are also automatically generated. Beyond four regressors, you begin to have enormous numbers of possibilities (hundreds to thousands), which makes automatic generation, calculation, and display, of all possibilities too unwieldy.
The next qdec will include a Contrast tab, so that prior to running the design, you would select from a list the contrasts that you want to keep (and of those, only the small ones will have the human-readable questions associated with it). Until then, you would have to run via the command-line (ie create your own contrast matrices) to make use of more than four regressors.
Nick
On Tue, 2008-03-18 at 14:10 -0400, Jerry Yeou-Wei Chen wrote:
Hello, I would like to know why we are limited to 4 regressors in the design specification within the QDEC GUI. If I understand correctly, there is no such limitation in the "old" method (via command line). Thanks,
- Jerry
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
for the hemisphere interaction, we need an interhemispheric registration. We now have that but it is under test. Ready "soon" I hope.
doug
Michael Harms wrote:
While on the topic of future qdec versions: Is there any plan to develop a qdec that would support simultaneous analysis of the left and right hemisphere in a repeated measure type framework? (e.g., so that you could easily create statistical maps of the main effect of group, the main effect of hemisphere, and group x hemisphere interaction).
thanks, Mike H.
On Tue, 2008-03-18 at 14:46 -0400, Nick Schmansky wrote:
Jerry,
It boils down to a GUI issue, that we are working to eliminate in the next qdec release (that is, eliminate the limitation of four regressors).
The problem is that qdec currently automatically generates all possible contrast matrices for the specified up-to 4 regressors. The questions associated with each contrast are also automatically generated. Beyond four regressors, you begin to have enormous numbers of possibilities (hundreds to thousands), which makes automatic generation, calculation, and display, of all possibilities too unwieldy.
The next qdec will include a Contrast tab, so that prior to running the design, you would select from a list the contrasts that you want to keep (and of those, only the small ones will have the human-readable questions associated with it). Until then, you would have to run via the command-line (ie create your own contrast matrices) to make use of more than four regressors.
Nick
On Tue, 2008-03-18 at 14:10 -0400, Jerry Yeou-Wei Chen wrote:
Hello, I would like to know why we are limited to 4 regressors in the design specification within the QDEC GUI. If I understand correctly, there is no such limitation in the "old" method (via command line). Thanks,
- Jerry
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer
freesurfer@nmr.mgh.harvard.edu
-
Doug Greve -
Jerry Yeou-Wei Chen -
Michael Harms -
Nick Schmansky