[open] Preparing a set of stimuli for further execution within the loop (possible?)

alebedev

Dear Colleagues,

We've just successfully started one of our behavioral pilots...
About a day ago I found that the task has substantial timing problems due to the fact that it has been set up via loops...
Real 1000ms took 1025ms in OpenSesame (V 2.9.4 // tested on Windows 7 and 8 // Legacy back-end; even longer in psycho).. So, I had to change stimuli duration accordingly... (Fortunately, precise timing was not super critical for our pilot, but it will definitely be for the proper study and, of course, for its fMRI part..)

I used to think that OpenSesame pre-loads the stimuli by default (even those that are within the loops), but then (unfortunately, too late) I read the documentation that is specifically focused on timing (http://osdoc.cogsci.nl/miscellaneous/timing/) and realized that this is not the case..

One simple (but not ideal) solution would be to prepare all stimuli and fixation points separately in one sequence (so, the loop would have only one iteration)... But then I don't know how to set up responses and feedback properly in this case..
I tried implementing preloading with inline scripts without any luck... It does not seem to be straightforward for loop items.. (http://osdoc.cogsci.nl/2.9/usage/prepare-run/).. Or am I missing something?

I am attaching a reproducible (shortened) example of my stimuli (takes two minutes to run):
https://www.dropbox.com/s/yj9ay87ykk4s0a8/timeCheck.opensesame.tar.gz?dl=0

Could you please have a brief look at this problem and, if possible, suggest a solution?
Thank you so much for your time! I feel completely stuck with this...

Best Wishes,

Alex

sebastiaan

Hi Alex,

Do I understand correctly that your main goal is to have fully predictable timing, so that also the intertrial intervals are always constant? This is indeed not trivial to do, for the reason that you already found out: Preparation occurs at the level of the sequence items, so the timing within a sequence should be predictable, but the timing between sequences is not.

The issue of 'non-slip timing' has come up quite a few times on this forum, as fMRI folk generally need it. Perhaps this discussion answers your question:

• http://forum.cogsci.nl/discussion/298/

Cheers,
Sebastiaan