How can I decrease the preparation time?

stephaniekr

Hello!

I've just almost finished programming an experiment with auditory stimuli, but I've run into a small problem. In the experiment, a trial consists of an auditory stimulus, a question (text_display followed by keyboard_press), and then a message to the participant saying to click SPACEBAR when they are ready for the next item (again: text_display followed by keyboard_press). However, when running the experiment, after pressing the spacebar for the final message, the audio for the next trial takes between 5-10 seconds to load. This is far too long. Is there a way to shorten the prep time? Could this long of a delay have to do with how we've stored the audio files, or something about the sequence itself? Am I right in thinking the delay is likely due to the preparation part of the prepare&run strategy, or could this be happening for a different reason?

Any insight would be much appreciated!

Thanks so much,

Stephanie

DahmSF

Hi Stephanie,

I experienced something very similar.
I imagine that you have a loop with the sequence: auditory stim, vis stim, resp, vis stim, resp (space)

As far as i know this has to do with the loop. All the stimuli are loaded each time the loop starts again.
At least that would explain the delay.

What i did in order to avoid that was writing everything in inline_scripts.
Here you can decide manually what you want to put into the prepare phase e.g. the content of the stimuli.
Of course this requires some python skills. But you do not need to be a professional to write that. Just use older scripts (see forum), and change them to your purpose.
The manual is also very helpful to get startet http://osdoc.cogsci.nl/3.1/manual/python/about/

Hope this helps
Stephan

sebastiaan

Could this long of a delay have to do with how we've stored the audio files, or something about the sequence itself?

You'd have to upload the experiment for us to be able to tell, of course. But given that you use audio, one thing that comes to mind is pitch and pan: If these are adjusted, the audio is processed sample by sample, which can take very long for long audio files. (This could be done more efficiently, but that would add extra dependencies to OpenSesame, which I prefer to avoid.)

Could that be it?

stephaniekr

Thank you both for your helpful comments! I'll try both solutions and hopefully one of them will help solve the problem.

I also tried decreasing the amount of dependencies, which has definitely helped the preparation speed.

Thanks again!

Stephanie