Regressors are overlapping even when formatted in local times

AFNI Message Board
1

AFNI version info (afni -ver): 25.0.08

Hello,

For some reason when I run my afni_proc.py script, and look at my X matrix, I have two regressors that look like they are overlapping. They shouldn't be overlapping, as the two timepoints that are similar are in different runs (i.e., different rows of the .txt regressor files).

Here's what the X Matrix looks like:

X
X1536×2048 214 KB

The two regressors in question are:

84.05 *
86.02
273.1
270.6

and

19.35 *
78.53
84.06
*

Any insight is appreciated. I wonder if afni is instead calculating my regressors in global format instead of local.

code text  # or delete if not needed
2

Howdy-

The stimulus event that starts at time = 86.02 s of the 2nd run in the first regressor, and the one that starts at time = 78.53 s of the 2nd run in the second regressor, are probably the events that overlap? That seems to make sense tha tthey would overlap, doesn't it? I believe the times count from the start of each run.

--pt

1 Like
3

Hi!

Thank you for your response -- seems like you were correct. When I was making my .txt regressor files, the run order was sometimes getting mixed up because of missing regressors (i.e., in the case of a block being filtered out due to a failed attention check).

-ssoulier