Is "Trial Variables" really randomizing?

SuperLab SuperLab 4 Support
1

I’ve created an experiment in which 2 stimuli are presented simultaneously. The location of each stimulus and the font color of each stimulus are supposed to be random. I tried to do this using Trial Variables. It appears that the randomization feature is not working. For example, one of the variable levels is always overrepresented (i.e., the one listed 2nd in the level list). It is not random. I can even determine which level is overrepresented by switching the order of the levels in the trial variable list. In the attached file, the positive stimulus list is presented as RED and as UP more often (every single time the experiment is run…way too often to not be systematic). Thus, I’m wondering if the randomization feature is not working properly or if I’m doing something wrong.
thanks!

Random.zip (8.04 KB)

2

Randomization Appears Faulty

Hi,
Sorry to be a pest, but it is very important to determine whether or not the randomize function is working properly in “Trial Variables.” Obviously, if it is not working, then this could seriously jeopardize the scientific integrity of many of the study using SuperLab. I’ve continued to run the program, and the randomization feature leads to about a 3 to 1 (and even 4 to 1) distribution of the 2 levels in the Trial Variable set (not ~50/50 as it should). The level listed seconded always receives a larger proportion of showings.
thanks,
G.

3

Developers are currently working on your experiment. We will get back to you shortly.

4

This issue has been fixed in the new SuperLab 4.5.2 release.