Amplitude Modulation GLT set up

AFNI Message Board
1

Hello!

I’m trying to use trial-by-trial behavior as a parametric/amplitude modulator and look at contrasts where I compare the modulation effects between conditions.

I saw the following post: https://afni.nimh.nih.gov/afni/community/board/read.php?1,78640,78703#msg-78703 which suggests that I should be doing the following:

-gltsym ‘SYM: myCondition1[1] -myCondition2[1]’

I just want to double check that I am conceptually understanding this correctly. If I were to do the following:
-gltsym ‘SYM: +Ant.Mean -Ant.Nice’ -glt_label 3 Ant.Mean.V.Nice

it would be equivalent to: -gltsym ‘SYM: +Ant.Mean[0] -Ant.Nice[0]’ -glt_label 3 Ant.Mean.V.Nice

and would NOT be actually capturing any of the amplitude modulation?

Whereas the following: -gltsym ‘SYM: +Ant.Mean[1] -Ant.Nice[1]’ -glt_label 3 Ant.Mean.V.Nice

Would be examining the difference in the modulation effect between the Mean and Nice conditions

Thank you!

2

Actually, without specifying an index, it should default to the sum. So:

-gltsym ‘SYM: +Ant.Mean -Ant.Nice’ -glt_label 3 Ant.Mean.V.Nice

would be (assuming just 1 AM param) the same as:

-gltsym ‘SYM: +Ant.Mean[0] +Ant.Mean[1] -Ant.Nice[0] -Ant.Nice[1]’ -glt_label 3 Ant.Mean.V.Nice

And you presumably do not want [0] added in.

  • rick
3

Hi Rick,

Glad I clarified! Thank you!