regress_'BLOCK(d,p)'

AFNI Message Board
1

Hi afni experts,

I have one question regarding 3dDeconvolve. My experiment is slow event-related design, and duration of each event does not exceed one second. In the 3dDeconvolve help file, there are 2 variants of BLOCK: BLOCK4 and BLOCK5, and the latter is delayed by about 1 second from the former. However, I am not sure which ‘BLOCK’ (BLOCK4 or BLOCK5) is more suitable in my experiment.

Thanks a lot in advance!

-Xin

2

Xin, if you know the duration of the trials, you can directly specify the HDR predictor as BLOCK(d,1), where d is the duration.

3

Hi Gang,

In the 3dDeconvolve help files,
==> ** Despite the name, you can use ‘BLOCK’ for event-
related analyses just by setting the duration to
a small value; e.g., ‘BLOCK5(1,1)’.

Therefore, I am not sure which ‘BLOCK’ (BLOCK(1,1) or BLOCK5(1,1)) is more suitable in my experiment.

Thank you very much.

-Xin

4

Xin, I don’t think any of these, BLOCK(d,1), BLOCK(1,1) or BLOCK5(1,1), would make much difference.