psychopy.org | Reference | Downloads | Github

Is it possible to randomize per participant and maintain randomization across sessions?

Description of the problem: For example, let’s say that each participant must remember a list of words that are randomly taken from a word pool during session 1. Then, a week later during session 2, some of those presented words are mixed with new words for a recognition test.

My question is whether it’s possible for PsychoJS to somehow go back to session 1 and retrieve the list of words that were presented. I think one solution is to set seeds based on participant numbers, so that the randomization can be reproduced each session. Would this work? I’m open to any other suggestions

Hey, I do not know if this specifically answers your question but randomization across participants topic is covered in a youtube video. Memory Experiment in PsychoPy (Part 2) | Randomizing Stimuli Across Subjects | PsychoPy Help #6 - YouTube

Based on this, it should be possible outside of PsychoPy.

So then, I guess what I’m concerned about is how to test if the way PsychoPy randomizes things will function the same way as the above link, then how to test if PsychoJS is randomizing it properly too. Ideally without starting up the whole experiment, if there’s a clever way to do it.

I’m also wondering if it’s good enough to simply set the seed based on participant number (where participant numbers are not randomized), or if I should add some sort of exponential to the participant number. (i.e. n + 13^n = 1 + 13^1 = 14, for participant #1)

I think that this is a good place to start. It should be pretty easy to test.

You might find my code snippets on seeding useful