Dear AV,
On Thu, 12 Aug 2021 23:54:04 +0100, AV <[log in to unmask]> wrote:
>Hello,
>
>I am interested in running VBM analyses using the TFCE toolbox/CAT12 and am new to using both. I understand from the others posts in this thread that we first:
>
>1. run our design in SPM
>2. estimate in SPM
>3. define contrasts using SPM results
>
>My first question is if we apply certain options for running the results in SPM before using the TFCE toolbox? For example, masking = none, then should we choose FWE or uncorrected? If uncorrected, should we go with the defaults p<.0001 and put in our value for extent threshold for voxels?
There is no difference in choosing the thresholds compared to the parametric analysis in SPM12. The same rules hold for the TFCE toolbox.
>
>I tried running the TFCE analyses and can see that we can choose FWE, FDR, or uncorrected when viewing results, but I wasn't sure if we should choose certain options in SPM-results before running TFCE estimation.
>
No, this is independent from the previous SPM12 analysis. Only the data, the design and the contrasts are used.
>Another question I had with the TFCE toolbox is that it states I can choose more than one contrast to estimate, however when I do so, I get an error message "Error using isempty
>Too many input arguments." Am I doing this the wrong way?
Please check that you have the newest TFCE version.
>
>Also, reading through the other posts it seems we can run repeated measures ANOVA with TFCE, but not using the flexible factorial design in CAT12. So can it be done through through SPM 2nd-level analyses instead?
I am not sure whether I fully understand that question. You can analyze (almost) all 2nd-level analysis design for bot flexible and full factorial designs. Only the interface for creating 2nd-level models in CAT12 has changed with the new CAT12.8 version. However, it's is still the same model as in SPM12, only the interface to build the model has changed.
Best,
Christian
>
>Thank you,
>AV
|