The problem is that the dpsd is 0.5 which means that you want to perform an FIR analysis to reconstruct the hemodynamic response at a temporal resolution of 0.5 sec. This is probably not what you want to do. But you need 0.5 because your TR is 2.5 and your stimulus is 3.5. A way around this is to set your TR to 0.5 and ntp to 100*2.5/0.5. It won't be perfectly optimal, but in the end nothing is. I would also increase the psd window to about 24 sec.
On 06/30/2014 03:52 PM, Elise Cardinale wrote:
Hello!
I am trying to use optseq to optimize my study design and cannot get it to accept the parameters I need for my task and I have officially stumped those helping me at my institution so I need your help!
I am trying to run a very simple design where my stimuli are presented for 3.5 seconds and I have 60 stimuli:
10 Montreal Pain
10 Montreal Neutral
10 Ekman Neutral
10 Ekman Fear
10 Ekman Anger
5 Montreal Anger
5 Montreal Fear
I calculated that at 3.5 seconds each that equals a total of 210 seconds of task and I added 40 seconds of null (~20% of task). Totaling to 250 seconds. My TR is 2.5 seconds giving me 100 time points.
I tried to run the following Optseq command:
optseq2 **--ntp 100 --tr 2.5 --tprescan 0 --psdwin 0 16 .5 --ev angerE 3.5 10 --ev fearE 3.5 10 --ev angerM 3.5 5 --ev fearM 3.5 5 --ev neutralE 3.5 10 --ev neutralM 3.5 10 --ev pain 3.5 10 --nsearch 10000 --nkeep 6 --o EmotionalFaces_Donor_FINAL2
When I do I get the following:
INFO: Setting srand48() seed to 374226 optseq2 $Id: optseq2.c,v 2.15 2009/05/26 18:13:45 greve Exp $ NoSearch = 0 nSearch = 10000 nKeep = 6 PctUpdate = 10.000000 nCB1Opt = 0 seed = 374226 Ntp = 100 TR = 2.5 TPreScan = 0 PSD Window = 0 16 0.5 nEvTypes = 7 EvNo Label Duration nRepsNom 1 angerE 3.500 10 2 fearE 3.500 10 3 angerE 3.500 5 4 fearE 3.500 5 5 neutralE 3.500 10 6 neutralM 3.500 10 7 pain 3.500 10 PctVarEvReps = 0 VarEvRepsPerCond = 0 PolyOrder = -1 tNullMax = -1 tNullMin = 0 outstem = EmotionalFaces_Donor_FINAL2 AR1 = 0 No refractory penalty Cost = eff OutStem = EmotionalFaces_Donor_FINAL2 Summary File = EmotionalFaces_Donor_FINAL2.sum nTaskAvgs = 224
ERROR: DOF Constraint Violation: number of estimates (224) is greater than or equal to the number of time points (100). To fix: (1) increase the number of time points to 225, or (2) decrease the number of event types, or, (3) reduce the size of the PSD window, or (4) increase the dPSD.
If I decrease the PDS window it comes back with 10000/10000 being unusable. I can't change my dPSD interval as my TR is 2.5 and my event is 3.5 and even if I increase my time points (which is not an option- I cannot double the length of my task) it still comes back as 10000/10000 being unusable.
Any advice on how to fix this?
Thank you so much!
-Elise Cardinale
Laboratory on Social and Affective Neuroscience Georgetown University White Gravenor Hall 302 email: emc62@georgetown.edu mailto:emc62@georgetown.edu phone: 202-687-6853
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer