Hi Hanah, I found another bug with the way that tNullMin was being implemented which causes this problem. You can get the new version from here: ftp://surfer.nmr.mgh.harvard.edu/transfer/outgoing/flat/greve/optseq2.linux doug

On 05/14/2012 08:47 AM, Hanah Chapman wrote:

Dear Doug,

Thank you for your response to my question about optseq and tnullmin/max on the freesurfer listserv. The new file you sent did fix the problem where some null trials exceeded the tnullmax, but I'm still having difficulty calculating the correct number of timepoints. I always get one very long (~60s) null event at the end. My starting approach was to compute the average length of the null trial within the range of tnullmin-tnullmax, and then allow that much time between each trial. I also tried to reduce the number of timepoints in steps, but I still get the problem of the long null at the end right up until the point that optseq returns an error (presumably because there are not enough timepoints). If you can offer any insight, that would be much appreciated. The starting code in which I used an average of 7s between each trial is this:

./optseq2.linux --ntp 360 --tr 2 --tnullmin 4 --tnullmax 10 --psdwin 0 20 --nkeep 6 --o ./attention/360tp --nsearch 1000 \ --ev DD 2 5 \ --ev DF 2 5 \ --ev DP 2 5 \ --ev DN 2 5 \ --ev FF 2 5 \ --ev FD 2 5 \ --ev FN 2 5 \ --ev FP 2 5 \ --ev PP 2 5 \ --ev PD 2 5 \ --ev PN 2 5 \ --ev PF 2 5 \ --ev NN 2 5 \ --ev ND 2 5 \ --ev NF 2 5 \ --ev NP 2 5

Thanks for your help,

Hanah

--

Hanah Chapman Postdoctoral Fellow Department of Psychology The Ohio State University