[open] Randomization and memory across "experiments"

JSS

Hi,

I have a simple reading aloud task with 2 presentation phases for which I need to randomly select 12 items out of a pool of 60 items. So for the first presentation phase:

At the beginning of the experiment I specify the list of items
list = ["pain", "appel", "soupe", etc.]
global list

In the loop I select 12 cycles and in the prepare phase of the inline script I randomize the list and select one by one:
import random
random.shuffle(list)
global list
word1 = list.pop()
exp.set("word1", word1)

In the second presentation phase I need to do the same but now I have to select another 12 items from the pool of the remaining 48 items. In between phase 1 and 2, I run various other tasks with OpenSesameAndroid, so I need some kind of memory trace from phase 1 being carried over into phase 2 each time. Is that possible?

Thanks for ideas!

eduard

Hi,

Could you please specify what exactly you mean with "some kind of memory trace"?

If it is only about making your experiment remember which words were already chosen in the first presentation phase, your description sounds alright. You could also save all the lists as attributes of the experiment, so that you don't have to declare them explicitly as globals:

exp.list = ['pain','soupe', ...]
random.shuffle(exp.list)
# and later
word = exp.list.pop()

Good luck,

Eduard

JSS

Hi Eduard,

Some kind of memory trace refers to the second presentation phase in which I have to select another 12 items from the pool of the remaining 48 items after phase one (which has 60 items in total). This is each time a different random list. Can I save the end list of phase 1 (48 available items) and load it when I start phase 2 as a new experiment? Is that what you mean by saving the lists as attributes of the experiment?

Thanks for help!

eduard

Hi,

Attributes of the experiment basically describes variables that are stored as part of the experiment object instead of existing only locally in a certain part of your script.
So when you start your experiment, OpenSesame automatically initializes such an experiment object by calling experiment.init(), which contains by default some variables, such as screen width, screen heigth, subject number, etc...

In an inline_script you can access the experiment with self.experiment or simply exp and save new variables to it like so: exp.variable = value. Now the variable is accessible from everywhere in your experimental sequence. That is, if you define it in a first inline_script in the beginning of your experiment, you can retrieve its value in another item (e.g. a sketchpad) at a later stage.

Furthermore, as you can read read, variables defined in one inline_script are also accessible in other inline_scripts even without declaring them as globals or part of the experiment.
What this means for you is that you once popped elements from your 60-element list, there will be only 48 elements left, when you start popping again later. Your memory trace is therefore present by default.

Hope this clarifies things.

Good luck,
Eduard