Hi Kris. I have experience with his type of scenario, and wanted to provide my input.
1) Your ratings will represent 4 separate regressors, since the difficulty of each condition will have a differential effect on the distinct output.
2) Do not know about about your contrasts, since I use AFNI. The amplitude modulated response can be built at the 1st level analysis (within-subject).
3) At the 1st level analysis (in AFNI), program 3dDeconvolve can output a b coefficient (unstandardized beta) for each regressor, not a correlation coefficient.
4) Do you really want your difficulty ratings to represent a regressor of no interest? If this is true, its effect can be accounted for at the within-subject level, so there is no need to include as a covariate in 2nd level analysis..
Hope this helps.
Christine
________________________________________
From: SPM (Statistical Parametric Mapping) [[log in to unmask]] on behalf of Kris Baetens [[log in to unmask]]
Sent: Wednesday, December 21, 2011 10:50 AM
To: [log in to unmask]
Subject: [SPM] Parametric Modulation - 1st level/ 2nd level
Dear SPM'ers,
I hope somebody can help. I plowed through the archives, but as none of the designs seemed equivalent to mine, I didn't find the right answers.
I have had participants performing four distinct tasks. After each trial, they rated the difficulty of the particular task they just performed for the particular stimulus they saw on a four-point scale. Apart from simple contrasts between the conditions (tasks) I would like to see in which areas the activity correlates (negatively or positively) with the parameter, e.g., "during task A, area X was significantly more active as the difficulty parameter increased/decreased". There's some confusion in our lab about how to proceed.
My questions are the following:
1) As the four tasks were qualitatively different, I assume I have to treat the difficulty ratings as four different parameters (one for each condition)?
2) At the first level, I specified four conditions, with one parametric modulator for each. Then in the contrast manager, I defined (apart from the contrasts between conditions) two contrasts for each parameter, a negative and a postive one (e.g., [0 1 0 0 0 0 0 0] and [0 -1 0 0 0 0 0 0]. Is this correct? Between what and what exactly is a contrast calculated in this step?
3) How to precisely interpret the output images of these contrasts? Do these reflect in some sense areas in which activity is positively/negatively correlated to the parameter?
4) How to proceed at the 2nd level? To demonstrate any difference between the conditions after "evening out" the influence of the parameter, I guess I would have to specify the parameters as covariates. But how to calculate, on a group level, "during task A, area X was significantly more active as the difficulty parameter increased/decreased"?
I would greatly appreciate any input.
Regards,
Kris
The information in this transmission is intended for official use of the Kessler Foundation. It is intended for the exclusive use of the persons or entities to which it is addressed. If you are not an intended recipient or the employee or agent responsible for delivering this transmission to an intended recipient, be aware that any disclosure, dissemination, distribution or copying of this communication, or the use of its contents, is strictly prohibited. If you received this transmission in error, please notify the sender by return e-mail and delete the material from any computer.
|