We created an experiment using PsychoPy and got it to run on Pavlovia. Our main goal is to have an experiment that participants can answer on mobile devices. The experiment consists of Pavlovia surveys and PsychoJS.
Before the most recent update the surveys looked perfect, adjusting ratios for mobile devices. However, after the most recent update, the ratios look as if they are ment for PC and not for mobile devices. I have tried checking if the function to enable a desktop version was enabled on mobile devices, but this was not the case. This problem occurs on both Android and iOS devices.
Can someone verify the same problem and can something be done about it?
Hi, I wanted to add to this because my issue may be related. I have noticed that on some mobile phones pavlovia surveys do not show up as intended (i.e. they do not take up the entire width of the screen) but only take a up a fraction of the screen. This makes it next to impossible for participants to read what is on the screen. The small display problem seems to occur on Android 12 phones, not on Android 13 or 14. It does not seem to be browser specific (both edge and chrome show the small display problem on an Android 12 phone). I have attached a screenshot that illustrates the problem
Sorry that you are experiencing this bug. We can replicate and are working on it. In the meantime I have a potential solution that you might try.
In your survey on pavlovia go to the “json editor” tab copy that text and paste it to a text editor on your local machine with the extension .json (to make usre it saves as a json file).
Thank you for your solution. This indeed seems to work!
The only problem we now have is that in our experiment setup, we first use a Pavlovia survey that is unlinked with PsychoPy in which we ask for informed consent. Then after giving certain answers, participants are then linked to a specific version of our experiment in PsychoPy. After this experiment, the participants are then redirected to a survey, which IS linked in PsychoPy.
Following your solution the latter seems to be completely fixed, but unfortunately the informed consent procedure still has the resolution issues (due to the fact that this survey is unlinked).
For the informed consent first survey to be fixed I think we unfortunately need to wait for a solution, or link the informed consent procedure to the PsychoPy experiment itself.
I am pleased that this partially fixed the issue (though apologies that it did not fully resolve it!). Is it possible to use the proposed approach for your informed consent form by creating a separate PsychoPy experiment that contains an embedded survey (with the json from your informed consent?)
I tried the demo and it looks great! I also changed the temporary fix back to the latest version of PsychoPy with the live ID of the Pavlovia Survey and this also looks perfect!
select “survey model file” as the survey type and select your json as the json file.
