Millisecond Forums

+x

 

Dave - 7/30/2021

+x

 

M Gerathy - 7/30/2021

+x

 

M Gerathy - 7/29/2021

Thank you very much, David. Clearly I need to spend some more time learning the software.

The project is now completely ready to run.

I'm so sorry, David but I have instituted your changes and I am still experiencing the same problems (i.e. items presented less or more than the desired 9 times over the experiment, and being replaced into the list within each PEP block). I apologise, but at my level of understanding of the software I dont know where else to look to solve the problem.

Thank you very much again.

Matt.

Maybe you should provide some data and point out exactly what you mean. It's not clear because *item* selection has nothing whatsoever to do with lists in your script. Your lists perform *trial* selection.

Okay, having done some tests, the issue comes down to extraneous samples occuring from the true_probes, false_probes, and catch lists occurring *somewhere* (I don't have the time to go through the ~8000 lines making up the script with a fine-toothed comb to pin down exactly where / when right now.)

Fortunately there's an easy fix -- resetting the lists right before the actual sampling from then begins in the given block. Try changing the one_second trial that starts each block to

<trial one_second>
/ ontrialend = [
    list.true_probes.reset();
    list.false_probes.reset();
    list.catch.reset();
]
/ inputdevice = keyboard
/ stimulustimes = [0=blank, progress]
/ trialduration = 1000
</trial>

and that should ensure every trial contained in those lists is run exactly once during the given block.

In the future, if you have some question that you've diagnosed based on some test data, please provide the relevant data files. Generating test data takes forever with this script due to the sheer amount of trials -- and that exercise is not necessary if data showing an issue already exists.