Re: [Freesurfer] Event-Related design and jitter in stimulus onset

2011-12-09 Thread Douglas N Greve
By default, optseq uses an internal clock synchronize to the TR (so events can only start on the TR). This can be changed by setting dPSD (eg, to 1). This would allow the NULL events to be multiples of 1 instead of 2. The PSD is specified in seconds. doug SHAHIN NASR wrote: > optseq2 --ntp 132

Re: [Freesurfer] Event-Related design and jitter in stimulus onset

2011-12-09 Thread Douglas N Greve
What is your command line? SHAHIN NASR wrote: > > Thanks Doug. > In my experiment, each trial (event) consists of 1 sec stimulus > presentation and 5 sec blank interval (TR=2). I have also added a > condition with just blank presentation which will be repeated as much > as other eve

Re: [Freesurfer] Event-Related design and jitter in stimulus onset

2011-12-09 Thread SHAHIN NASR
Thanks Doug. In my experiment, each trial (event) consists of 1 sec stimulus presentation and 5 sec blank interval (TR=2). I have also added a condition with just blank presentation which will be repeated as much as other events. I have used optseq2 commands to generate the trial seque

Re: [Freesurfer] Event-Related design and jitter in stimulus onset

2011-12-09 Thread Douglas N Greve
Hi Shahin, there are several ways to answer this question. If you are going to analyze assuming a shape to the HRF, then the actual delay between stimuli is technically not important (though 4s will be good to avoid attenuation in the following stimulus). If you're using an FIR model of the HRF

[Freesurfer] Event-Related design and jitter in stimulus onset

2011-12-08 Thread SHAHIN NASR
Hi Surfers, It is my first experience with Event-related paradigms. In my experiment, trials take minimum 6s (TR=2s). I wonder what is the optimum variable delay between trials to be able to analyze data without any confound. Regards -- Shahin Nasr PhD in Cognitive Neuroscience Martinos Ima