In this case, the event will be 7 sec. The way it is set up, it assumes that you will use an FIR model to analyze the results (ie, not assuming a shape to the HRF), making it difficult to separate the various subevents to the event (though you can assume a shape to the HRF). It is appropriate to use a non-zero value for psdmin (eg, -4 sec) when analyzing using an FIR so that you can see the level of signal before stimulus onset. This makes for nice graphs and allows you to see if the subject is somehow anticipating the stimulus. doug

On 05/14/2012 02:31 PM, Andrew Valdespino wrote:

(1) Version: optseq2.c,v 2.21.2.1

(2) Computer operating system: Mac OS X Version 10.7.3

(3) Command line: Not relavant here

(4) Description of the problem: A trial in my experiment involves a countdown (duration 2 seconds) followed by the event of interest (0.5 seconds) followed by a blank screen (3.5 seconds). I am having trouble with the definition of 'event duration'. That is, for the argument --ev x *y* z, what do I specify for *y *(event duration)? the entire trial is 7 seconds in length, but the duration of the event of interest is only 0.5 seconds (embedded between a countdown, and a blank screen).

As a second question, is psdmin always set at '0'? Would '0' be appropriate for the trial structure described above?

(5) Log and/or summary files: Not relavant here

---

---

Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer