On 6/13/13 9:45 AM, A a wrote:
Hello all,

A question regarding the usage of Optseq2

I have this command line

optseq2 --ntp 200 --tr 2.5 --psdwin 0 15 2.5 --ev ev20 2.5 20 --ev ev40 2.5 20 --ev ev60 2.5 20 --tnullmin 5 --tnullmax 7.5 --nkeep 1 --o OptSeq_TRIALS/3_null9_6 --nsearch 10000

Now the efficiency tha I get is around 0.5. How I can tell this is a good or bad efficiency ? Is there a comparable level?
Not really. This is a question of statistical power, so it depends on how big your effect is. Eg, if your effect is 0, then it does not matter how what your efficieny is.

Also if I would like to test for the first order effect of using a parametric modulation, can I test this using OptSeq2?
No, this requires a response from your subject, which you don't have.

Lastly, if I would like to test the efficiency of a contrast, say ev60 > ev20, how I do this ?
You would have to create a contrast matrix and pass it with --C.
doug

Thanks


Adnan


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