Pavlovia Error - Data Not Saving When Piloting or Into Github Repository on Chrome AND Microsoft Edge

Online experiments
1

If this template helps then use it. If not then just delete and start from scratch.

OS Win11
PsychoPy version 2025.1.1

Pavlovia Error - Data Not Saving When Piloting or Into Github Repository on Chrome AND Microsoft Edge. I have tried with several experiments, several browsers, and several computers. Yesterday, I was able to get some data saved through Microsoft Edge, but today it is no longer working.

To clarify, when piloting, no file is being saved locally. When running, no file is being saved in the github repository. Any help would be greatly appreciated!

2

When piloting, do you see any error messages?

Can you test with a minimal experiment?

Does your experiment have any code in End Experiment tabs or try to do anything unusual with redirection or data saving?

3

Hi wakecarter,

I have tested with several experiments, including minimal experiments, previous experiments that I have already finished data collection with, etc. There are no specific redirection instructions for data saving.

The interesting part is that it seems to be browser and device dependent, and not experiment dependent. It seems to be working in the lab now on Microsoft Edge only, not on my personal laptop though. On my personal laptop, only firefox saves any data.

4

Also no error messages.

5

Update: It is also not dependent on the PsychoPy version used to coded the experiment. I tried an old experiment coded on 2023.1.3 Chrome appears to not be saving anything locally when piloting, or into gitlab when running. I am curious if others have the same issue as well. My thinking is that the recent chrome update may have affected something on the backend, I am not sure though.

6

PsychoPy version: 2025.2.3

When running experiments on Pavlovia on Google Chrome, no data is saving. Data for online participants that used Chrome is nowhere to be found, but I have files for participants that used Safari and Firefox.

7

@annakazatchkova @Maxim_Axentiev You both seem to have a very similar issue. Have you checked to see whether the data is downloadable from the experiment project page?

What version of Chrome do you have? I’ve just updated to Chrome version 155.0.8059.40 (Official Build) (64-bit) and my test experiment saved data both when piloting and running. I’m using a PC.

Please could you add me (Wake) to a minimal experiment that is failing? Is the failed data showing as submitted on the project page?

8

@annakazatchkova @Maxim_Axentiev You also seem to both be at the same institution in Ontario, Canada which leads me to wonder whether something odd is going on with your access. Is the issue ongoing, and is it dependent on your ISP?

9

I just tried with a new minimal experiment and it does seem to save on Chrome, but I am still having issues with the experiments I am actually trying to run, unfortunately. To answer your questions, 1) My Chrome version is also 155.0.8059.40 2) For any experiments I’ve tested, the data does not save to the project file either. 3) I have tested at home and at the lab and the University has a different ISP than I do at home.

I have tried with different experiments that have been coded on different versions of PsychoPy, and I have had other grad students from other labs test it on their experiments (hosted on separate Pavlovia accounts) that used to otherwise collect data normally, no longer collecting data. We have tried on different operating systems as well. Safari and Firefox are the only browsers collecting. The lab computers collected on Edge until the automatic update yesterday, and they all stopped collecting on Edge now.

It is interesting that it did work for the minimal experiment though this morning. I have added you to both the minimal and one that is not working for me that I would like to have up and running.

10

I think that the culprit is likely to be something in the bespoke JS code, especially if there is any code shared between you and @Maxim_Axentiev .

To my mind, the most likely culprits are:

disableAutoCorrect();
Your textbox functions
quitPsychoJS

Therefore, please could you try breaking the minimal test by adding the welcome routine and elements in the words routine that make use of the functions in Welcome? I would try this myself, but I have 2026.2.4 and I’d rather you confirmed you can replicate the error in your version before I try making edits (to the minimal test) myself.

11

I am also experiencing the same issue. I have an entirely different code than Anna (with no textboxes). When I run the experiment on chrome and microsoft edge the data is not saving. But it does work on firefox. I also have the latest psychopy update and no errors show.

12

I have made these changes to the minimal experiment, but it is still saving. Like I said, this issue works across the board for any normal experiment we have run in our lab in the past, ones with textboxes, ones without. The issue is also occurr other labs at our University. We all have different coding styles and needs, but the issue is occurring across the board on Chrome and Edge specifically. I don’t know if maybe it depends on the size of the experiment.

Wakecarter, I am wondering if you are able to test any of your own experiments that are more substantial? Even just partially to see if an incomplete file gets saved (because for me partial and full completions are both not saving).

13

I tried piloting the experiment you shared with me in Chrome, aborting after two mnemonics, and the pilot data saved (after pressing escape twice as normal).

14

Hi wakecarter,

I have found the solution:

  1. Originally when the issue came up, it was on a slightly different Chrome update, and is now resolved with the latest Chrome update. On a lab computer without the update, the issue still persists (even when completing fully or using escape twice). Will need to hope most participants’ Chrome will update automatically.

  2. I had never previously had the issue of needing to press escape twice for incomplete data to save, I have always just closed the browser tab. I see there was discourse in 2020 regarding this issue happening for others. Given that most online participants wouldn’t know to press escape twice, I will need to find a way to incorporate this into code - I am wondering if you have a solution to this. We use an undergraduate pool for most of our data collection, and we grant credit even if participants don’t fully complete the study (or dont penalize if they at least read through the consent form and declined), so it would be ideal to find a solution to this to avoid a slew of angry student emails.

  3. The “escape twice” rule is not needed on Firefox or Safari.

15
  1. If you know what version of Chrome was failing, please could you post that here.
  2. Here’s my time on incomplete data saving.

If you end the study early via quitPsychoJS then maybe that behaves differently. Closing the browser tab might also work, but I’m sure that there’s something that doesn’t normally work – maybe it’s navigating away from the page on the same tab.

16

I was not able to get the Chrome version as it had already updated before I had a chance to check.

I was able to find a solution to this. To save data from experiments that close the experiment tab, you can add the following code in any code component in the “Begin Experiment” tab on the JS side:

document.addEventListener('visibilitychange', () => {
  if (document.visibilityState === 'hidden') {
    psychoJS.experiment.save({ sync: true });
  }
});

I have tested this and it has resolved the issue. Adding this to your experiments will help keep record of incomplete attempts when participants close the tab.

17

Interesting. Thank you. I’ve just tested it and discovered that it will cause data to be saved if psychoJS._config.experiment.saveIncompleteResults is false.

I can also create a throttling error if I move away from the tab, and then back and then away again (which means that a single data file is saved).

It also successfully saves data when clicking Leave on closing the tab or navigating to a new site.

image
image425×131 4.15 KB

I’ll update my tip