[open] Variable wizard

Michif

Hi

I really like the variable wizard, it makes stuff way easier. I come across a situation were I would be very happy to have one or two additional functions.

For instance, when I have the situation in which 2 or more variables can have the same values, and I dont want trials in which both of them have the same value, then I have to delete those trials (cycles) by hand. Which can be, when you have other variables as well, very time-consuming. For these situations, I would love to have the function to not fully cross all factors. Maybe after typing in, I could decide which factor and which levels of the factors should be crossed.

If that's not possible then I would be happy about a sorting function for all variables. Then you could easily spot the ones that you want to erase. So far the pattern in which they are listed after using the variable wizard seems rather inflexible.

Wouter

Hey Michif,

Not sure to what extent I can give a satisfying answer, but maybe I have a temporal partial solution

For instance, when I have the situation in which 2 or more variables can have the same values, and I dont want trials in which both of them have the same value, then I have to delete those trials (cycles) by hand.

Couldn't you use conditional run if statements in the trial sequence? like I did here . Basically, you could decide to run the trial sequence only if V1 != V2 . You could always extend this with an inline_script to make the condition to run a trial as complex as you wish.

I'm aware this is not the 'clean' solution you'd wish, as the loop object still has all unnecessary cycles in it, but I hope this helps.

Wouter

sebastiaan

Hi Michel,

I agree that it would be very nice if the loop variable wizard was more flexible (see feature request). What makes it tricky, though, is to decide how to implement this in a user friendly way. The things that you point out are specific to your experiment, other people will have different needs. Things like:

• making a Latin square design
• weighting certain combinations
• adding restraints, such that (for example) stimuli are not repeated within X trials
• etc.

So it would be a project in itself (!), and I'm afraid it will not be implemented for 0.27. For now, Wouter's suggestion makes sense. And, of course, you can also prepare your loop table in a spreadsheet and paste into the loop table.

Cheers!

Michif

Hi guys,

True, sebastiaan. In case I come up with one, I'll post it.

Thanks Wouter!