[solved] Randomize variable lists

Andrea_sct

Hi Sebastiaan,
I think I found an answer to my question here (http://www.cogsci.nl/forum/index.php?p=/discussion/389/open-random-order-to-variables-without-returning/p1) but I don't fully understand it.

In my experiment, I try to use a single question:
"If you have [P] probability to win [A] of money...."
where [P] and [A] are two variables in the loop. The former is a list of 45 probabilities, the latter is a list of 45 amounts of money.
If I run the experiment, the first probability is associated with the first amount the second P with the second A and so on.

I have to build an experiment in which there are 45 items with a randomized [P] order for every subject and random ordered [A] not repeated and not in fixed pairs with [P].

I hope to have explained properly my question.

eduard

Hi Andrea,

Unfortunately, like it was said in the thread you cited, you need to do some scripting in order to design your experiment in the way you want it. However, it is not very hard, so no need to worry. Here you can find the basic structure of how you can solve your problem. I will explain it briefly:

1. In the first inline_script, called "define_Amount", you have to create the list of the 45 options of how much money each subject can get per trial. You want it to be randomized, so you have to shuffle the list.

import random

exp.amount_list = [1,2,3,4....45] # or whatever it is in your case
random.shuffle(exp.amount_list)

2. After having created that list, you can enter the trial loop. Here, you have to add your other variable (i.e. probability) with each of the 45 levels included (just like that ). If you do it this way, the sequence, which is repeated in the loop takes exactly one of the probabilities as "P" variable for a trial. As you might have noticed, at this point the sequencedoesn't "know" yet what the other variable (amount) is, because it isn't included in the variable assistant. So, this is what we have to do in step 3.

3. In the inline_script "set_amount_as_variable", you have to set the value of the 2nd variable.

exp.set('amount', exp.amount_list.pop())

That's it. Now the sequence has one value for the probability and one for the amount. Each value is used exactly once and also in a random order for each participant. From this point, you can use them, in the way you described (" If [prob] equals [amount], ...").

Two small remarks:

1) Obviously, it doesn't matter which of your variables you define in the variable assistant and which you define yourself. It is totally up to you.

2) When you define one variable yourself in the very first inline_script, you shouldn't forget to put "exp." in front of the variable name, because if you don't, you won't be able to access it outside this inline_script. By adding "exp." you make it part of your experiment and hence, accessible at any point of it.

Do you understand now better?

Eduard

Andrea_sct

Thanks for your help,
when I try to begin an exp the software says:
"Variable 'amount' is not set in item 'rest_of_trial'.
You are trying to use a variable that does not exist. Make sure that you have spelled and capitalized the variable name correctly. You may wish to use the variable inspector (Control + I) to find the intended variable."

When I open variable inspector and I search for amount's variable I find:
"amount" in variable column
[Set to 'exp.amount_list.pop('] in the second column.

I put the script you post under the "run" label, is it correct?
Will it work if I change:

• numeric with character;

• labels "amount" with "D";

in this way:

first inline_script:

import random

exp.D_list = ["1","2","3","4","5"] # or whatever it is in your case
random.shuffle(exp.D_list)

and the second inline_script:

exp.set('D', exp.D_list.pop())

eduard

Hi Andrea,

when I try to begin an exp the software says: "Variable 'amount' is not set in item 'rest_of_trial'.

You defined your variable as part of the experiment, that is, if you want to access the variable, you have to tell Opensesame that it has to look for it in the experiment. The way this works is either by using brackets, [amount], or like this: exp.amount.

Will it work if I change:

numeric with character;

labels "amount" with "D";

Sure, the way you label your variables is pretty much your own decision.

Eduard

Andrea_sct

Hi eduard,
with the first inline I create the list:

import random

exp.amount_list = [1,2,3,4....45] # or whatever it is in your case

random.shuffle(exp.amount_list);

in the second inline I include the variable in the variable assistant:

exp.set('amount', exp.amount_list.pop())

and so when in the slider's question I write " If you have [prob] of win [amount]..." it should work or I forget to do something?

eduard

Exactly, that sums it up very well. I don't see a reason why it shouldn't work now, but why don't you just try it out?

Eduard

Andrea_sct

I have tried it but the variable amount hasn't been identified in the slider

eduard

Does it work if you use "exp.amount" instead of [amount] ?

Andrea_sct

Nothing the software doesn't recall the list and put in the questions exactly the words "exp.amount"

I try to move the inline script from run label to prepare label and the first windows works but when I click to accept answer it says:

Error while executing inline script phase: prepare
item: set_Amount_as_Variable
line: 2
exception message: 'int' object has no attribute 'pop'
exception type: AttributeError

Traceback (also in debug window):
File "dist\libopensesame\inline_script.py", line 133, in prepare
File "", line 2, in
AttributeError: 'int' object has no attribute 'pop'

eduard

The error tells you that the element you're popping from is not a list, but an integer.
So something went wrong there. Did you make sure, that you create a list in define_amount? Maybe you can post the code you used.

Btw. you're right, you have to set the amount for the trial in the prepare phase of the inline_script.

Andrea_sct

I thought that amount_list was only the variable name and so I change it to "M" now I have changed it to "M_list" and the experiment work.
Thanks for your help.

eduard

Nice!

Good luck with it.

Eduard