History of AFNI updates  

|
September 21, 2009 08:11AM
GAM doesn't take a duration parameter. That's what BLOCK is for: BLOCK(d) is basically GAM(4,1) integrated over a duration of 1 second. BLOCK5(d) is the same with GAM(5,1) instead -- delayed a little bit from BLOCK.

Try these commands to see what the BLOCK(1) response function looks like:

3dDeconvolve -num_stimts 1 -stim_times 1 '1D: 1' 'BLOCK(1)' -nodata 200 0.1 -polort -1
1dplot nodata.xmat.1D

To compare all 3 at once, try this:

Create file q.1D with the single line below

1*0.2 1.25*0.2 1.5*0.2 1.75*0.2 2.0*0.2

Then run the command below

3dDeconvolve -num_stimts 3 -stim_times 1 '1D: 1' 'BLOCK(1)' \
-stim_times 2 '1D: 1' 'BLOCK5(1)' \
-stim_times_AM1 3 q.1D GAM \
-nodata 200 0.1 -polort -1 -x1D stdout: | 1dplot -stdin -one -dx 0.1 -thick -thick

This will plot BLOCK(1), BLOCK5(1), and the sum of 5 GAMs spread evenly across a 1 second interval.

Then you pays your money and you takes your chances.
Subject Author Posted

3dDeconvolve - stim_times option

Srikanth February 23, 2009 05:48PM

Re: 3dDeconvolve - stim_times option

rick reynolds February 23, 2009 06:08PM

Re: 3dDeconvolve - stim_times option

Srikanth February 23, 2009 06:20PM

Re: 3dDeconvolve - stim_times option

rick reynolds February 23, 2009 07:13PM

Re: 3dDeconvolve - stim_times option

Srikanth September 19, 2009 04:48PM

Re: 3dDeconvolve - stim_times option

rick reynolds September 19, 2009 07:20PM

Re: 3dDeconvolve - stim_times option

Srikanth September 20, 2009 03:52PM

Re: 3dDeconvolve - stim_times option

bob cox September 21, 2009 08:11AM

Re: 3dDeconvolve - stim_times option

Srikanth September 21, 2009 09:47AM

Re: 3dDeconvolve - stim_times option

Srikanth December 17, 2009 06:37PM

Re: 3dDeconvolve - stim_times option

rick reynolds December 18, 2009 09:59AM

Re: 3dDeconvolve - stim_times option

Srikanth December 18, 2009 10:45AM

Re: 3dDeconvolve - stim_times option

Marc Ettlinger October 21, 2009 03:06PM