Selecting a condition file for each participant and make sure it's not assigned twice across participants

[FJBorn]

Hello everyone. Thank you in advance for reading and thinking about my question.

The experiment I want to asking a question about is implemented as a builder experiment and it is an online experiment.
I am working on: MacOs Monterey, using PsychoPy version: v2021.2.3

Description of the problem:

I created an experiment that involves recruitment via Prolific, then a short questionnaire in Qualtrics, and then redirecting participants from there to my Psychopy experiment. As part of the Psychopy experiment, a condition file is selected based on the participant ID assigned by Qualtrics. This works fine with the following code:

# variable to determine conditions file
main_task_blocks = (("conditions/conditions_" + str(expInfo["participant"])) + ".xlsx")

It is essential to note, however, that some condition files will not result in a full set of experimental results if participants quit an experiment. Also, sometimes Qualtrics seems to assigned the same participant ID twice, when participants enter the experiment in the same moment. The balancing of experimental conditions in my experiment is based on the assumption that each condition file is used for one participant (I am balancing my exp. conditions across participants), so some condition files not being used potentially results in an experiment that will not be fully balanced. And here lies my problem.

Rather than using Qualtrics’ assigned IDs to determine the condition file, I was thinking about implementing a solution in which one condition file (from a set of condition files) is assigned randomly at the beginning of every experimental session and will then be removed from the list of possible condition files after the experiment has been completed.
Given the tools I know from the Builder, it seems difficult to update such a list of condition files and carry this information to the next participant (or exp. session).
Within one experiment, ofc this is quite straight forward to implement, however, when transferring information from one participant/ experimental session to another, it seems to become more challenging.

Therefore, I wanted to ask that question here. I am thankful for your ideas and I would very much appreciate any thoughts on how this could possibly be implemented in a builder experiment.

[wakecarter]

Can I recommend you look at this thread

Deal with condition files on psychojs/pavlovia

In psychopy on the local machine, a lovely feature is that we can create the condition file for all subjects. When I upload the experiment online, I cannot manually assign condition files before each subject does the experiment. Is there a way around it? My colleagues told me that what they do is whenever an online subject begins the experiment, the code will extract all available condition files, randomly choose one for the subject, and then remove the condition file to another fold so that fut…

I specifically wrote the VESPR Study Portal to deal with non-finishers (for licensed users).

[FJBorn]

Hi @wakecarter,

Thank you very much for this helpful tip. I think this is a very good solution.

However, I would prefer to implement a solution directly into my JS script, rather than requiring an additional instance. I see a relevant advantage in implementing it myself, as it’s making the procedure more transparent for me, and I have more control over how IDs are assigned, what I do with people who don’t finish half way through the experiment, etc.

[ajus]

Hi @FJBorn,

I coded a little procedure using the shelf that should be quite close to what you want to achieve. Maybe it’s a nice starting point to build off of.

Comprehensive deterministic counterbalancing using pavlovias "shelf"

Hi all, I coded a little procedure to have the ability to counterbalance deterministically using the shelf feature of pavlovia. While it’s obviously lacking all the other features of the VESPR study portal, it does a similar thing when it comes to counterbalancing. The pros are that it does not require you to set up anything outside of pavlovia and you have full control. What it does Whenever a participant starts your experiment on pavlovia, the code looks up how many participants already com…

All the best!

[FJBorn]

Hi @ajus,

Thank you! I appreciate your helpful comment and the reference to use your little procedure with Pavlovia’s “shelf”. My experiment will not use Pavlovia’s “shelf”., so this unfortunately won’t be an option for me. It seems that I will have to find a solution built around different experiments as, according to your post and wakecarter’s previous comment, there appears to be no way to directly implement it in the experimental code.

All the best!

[ajus]

Hi @FJBorn, just out of curiosity: How do you run your experiment online if not on pavlovia?

[FJBorn]

Hey again,

sorry, that was just a typo, I wanted to say that I am not planning on using the “shelf” in general. However, seems to be the only existing solution out there to manage transferring information from one participant to the other, so I will look into it a bit more. Many thanks in any case:-).