hi @sotiri, the first list seems to be working (I’ll have to go through all the other 11 - but so far so good!). Maybe something for the other people in this thread: if your participants have already clicked on the link to your experiments, they’ll have to run the experiment again in a private window. I just tried in my regular window, but it doesn’t work there (probably because of the previous sessions). When I try it in a private window, the experiment runs fine. Thank you again @sotiri for your help!
Thank you for letting me know.
My auditory experiments seem to be working fine.
I am really thankful for your help @sotiri
No problem @esijye, I believe clearing out the cache in the current browser window should do the trick as well, x
Back up again and running! Yes clearing cache is key. Thanks so much @sotiri for this!
Working! I didn’t have to clear the cache. Thank you so much!
Thank you so much for this! Both experiments are working now.
@bbuniv, yes of course, would it be possible to send me a link to your repository so I can look at the code? Thanks, x
gitlab link:
Hi @bbuniv, I believe upgrading PsychoPy and regenerating HTML would help in this case. Is the following working as intended more or less? https://run.pavlovia.org/support/reaction-time-svs/html/
thanks a lot I will check that.
have nice evening
Ever since the issue concerning the invalid access error was fixed by @sotiri, we have encountered a separate issue with our online study concerning the use of the polygons used for feedback. This was NOT an issue before @sotiri fixed this issue (this is why we are adding this here and not in a separate post) on 12/4. In fact, we have been running this experiment using this feedback for over a year. We are happy to post this issue in a new forum post but wanted to let you know that an ancillary issue came up once the original issue was fixed. Here are more details:
Our experiment asks the participant to select one of the two images they see on the screen by pressing the Left arrow key or the Right arrow key. Their task is to match the audio stimuli to the picture by pressing the left or right arrow key.
After they respond, they see a green box (polygon component) appear around the correct image with a text that says either “Correct” or “Oops! That was wrong” above the pictures. After that, the trial moves on to the next image. This has been working perfectly before 12/4. See screenshots of the code components we use for this sweet experimental block (ignore code_13 which just serves as a counter)!
The issue occurs in both scenarios: either when the correct response is the left or right image. For example, if the participant correctly presses the Right arrow key because they believe the right image matches the audio, a green polygon appears around the left image even though the green polygon should (and previously was doing this perfectly!) around the right image. The text feedback (e.g., “oops that was wrong” vs. “correct”) on the top of the screen does not seem to be flawed. Therefore, in this situation, the participant sees “correct” on the top of the screen suggesting that the text feedback is acting as designed but the polygon feedback is not.
It seems that the green polygon always appears on the left image and never on the right image, no matter which key is pressed. This of course is not as designed. The polygon should appear around the right image when the right image is the correct answer. The text is not affected by this issue. The text will say “Oops! That was wrong” when the right image is wrong and the Right key is pressed. The green box never appears on the right image no matter which arrow keys are pressed.
As I stated above, the issue started after the previous issue was fixed Friday 12/4. Since then we have had 6 participants who went through the experiment and experienced this issue.
It may be that we have to update the version of Psychopy was used to create this experiment on Pavlovia. This experiment is running on Pavlovia using Psychopy version 2020.2.4. We have reproduced this issue on both Mac and Windows machines. Let me know if I should do this before doing anything else. Also let me know if you want me to create a totally new post.
Hi @rosem09, sorry to hear PsychoJS is giving you trouble like this. I would think the issue to be maybe related to other changes introduced last week, but would be very surprised if the 12/4 fix has caused any unwanted side effects, because it only deals with interfacing with the Tone.js library. The problem you describe seems related to ShapeStim rendering. Would it be possible to share your code with me so that I can attempt to repair the script? Thanks, s.
Absolutely! Yeah after I wrote that I realized I was saying your fix caused my issue and that is totally not what I meant! Do you want my gitlab link? If so, here it is (I hope you can open it because my experiment is set to private and I cannot seem to set it to public): https://gitlab.pavlovia.org/maya.rose/tcall_100520
I am also happy to drop the .psyexp and .py files here if need be.
Note: this feedback appears in a comprehension block that is repeated twice throughout the experiment.
Ah cool, no worries, here to help, yes you can give the support team account developer access without turning the repo public, the following page has the details https://docs.gitlab.com/ee/user/project/members/index.html
Just added you!
OK great, cheers!
We also got errors a few days ago as a result of the Toneplayer changes. Now some of our participants are still getting errors but we can’t reproduce them and haven’t yet managed to get information from them about the errors. This is frustrating. Is there a way we can easily use an older version of PsychoJS for our experiment to protect ourselves from the introduction of new bugs?
Hi @sotiri, some crazy things append on my computer, lot of error and warning append when I open my experiment with new version, and impossible to generate new html, like you, another thing strange, your experiment works really well on safari and not with my experiment uploaded… too much strange things. the best will be to re-create all with new version… actually I just push your html updated, I will take time to change it in January.
Thanks for your help again
No problem @bbuniv, let me know when you are ready to stage your project again and I will be sure to take another look, x