Simultaneity Judgement Task
Hello everyone!
firstly i'd like to thank you for the development of your AI! from 0 i was able to build up the experiment with almost 0 knowledge about coding. It is super usefull as it is well designed in including good suggestions and comments!
but now i need human help :D well i'm building up a simultaneity judgement task. As soon as it is about multisensory integration i have two type of stimuli: an audio beep (60dBSL, 5 ms rise-and-fall time, 1200 Hz) and visual stimulus (5x5) of a yellow circle. both the stimuli need to be presented for 33ms with soa ranging from -300 to 300. Now the experiment works well but i have a problem: when, in the trials, there are negative soas (audio before visual) the stimuli always appear to be synchronous and the audio never precede the visual stimulus.
Do you kindly have any suggestion to fix this issue?
Kind regards,
Lorenzo
Comments
I forgot to say that Sigmund AI suggested me to ask you in the blog. I can provide the code of the experiment.
Hi @Lorenzon,
Can you provide your experiment? I see two options: co-routines and python coding in inline_scripts. I am not super familiar with co-routines, so maybe there are some restrictions that I don't see, but if you wrap your auditory and visual presentation in a co-routine and set the starting times programmatically, you should be able to have negative and positive soas.
If co-routines are no options, python scripting will be able to do the trick.
Eduard
Hi.
instead of managing negative value, can you delay your main figure to 300ms, and randomly generate your stimuli from 0 to 600ms ?
Hi @Lorenzon,
@eduard's suggestions should both be able to solve your issue. A third way would be to have a trial inclusing a sound, the visual stimulus and a sound again, and to condition the presentation of the first and the second sound (using the "Run if" parameter in your sequence) based on whether the SOA is negative (in which case you'd present the first sound and set the delay to the visual stimulus to the absolute value of the SOA) or null/positive (in which case you'd only present the second sound). This said, Eduard's solutions are better.
You might want to run some tests registering the time stamps of all your events to check how accurate the timing is, though. Depending on the operating system etc., the timing accuracy can be variable. If you require very precise timing, it'd be a good idea to check this.
Hope this helps,
Fabrice.