help > Build-up of first-level model gPPI
Showing 1-2 of 2 posts
Sep 7, 2018 10:09 AM | Marlieke van Kesteren
Build-up of first-level model gPPI
Hi, I just started using gPPI whiile being familiar with PPI within
SPM. I wrote a wrapper to run all participants, but somehow the
models were constructed but not estimated. So I wrote my own code
to estimate the models and calculate the contrasts.
I have a question about these first-level models though. It seems that the initial model was copied and updated with the PPI factors, which are added behind the inital factors, so I assume the contrasts should be added to the latter. I accidentally did not do that the first time and noticed final group results are exactly the same when contrasting the factors taken from the initial model vs. the added factors that are called "PPI_...". I am a bit confused by this and am wondering what the factors in the model mean exactly. Hope you can elucidate that for me.
Thanks in advance for your answer!
Best regards,
Marlieke van Kesteren.
I have a question about these first-level models though. It seems that the initial model was copied and updated with the PPI factors, which are added behind the inital factors, so I assume the contrasts should be added to the latter. I accidentally did not do that the first time and noticed final group results are exactly the same when contrasting the factors taken from the initial model vs. the added factors that are called "PPI_...". I am a bit confused by this and am wondering what the factors in the model mean exactly. Hope you can elucidate that for me.
Thanks in advance for your answer!
Best regards,
Marlieke van Kesteren.
Sep 11, 2018 08:09 AM | Marlieke van Kesteren
RE: Build-up of first-level model gPPI
Ok I found out I made a mistake so contrasting the "old" factors is
not the same as contrasting the PPI factors. Still, I'm curious to
know whether I'm correct in assuming that I have to contrast the
PPI factors to get the PPI activity levels, and whether I can do
the same for the parametric modulators..?
