Description of the problem: I am designing an experiment where I show participants images to describe by video recording their answers. I created a loop to show them images. The problem is that after some items, the images start to upload later and later until it becomes a blank white screen. After about 30 seconds of a blank page, the experiment continues, but images sometimes are shown more briefly than they are supposed to be. To try to solve it, I start recording after the first second. I also added 1 second waiting time as a routine before the target description routine. But the problem endures. I uploaded images of the target_description routines below.
Do you have any suggestions?
Thank you for your help in advance!
Are the images set by a condition file in the loop? You might want to add the images to the “additional resources” in the “online” tab of the Experiment Settings, which should make Pavlovia pre-load all of the images before the experiment starts.
They are set by condition, but they are also uploaded to the online TAB. The thing is, sometimes the experiment runs without a problem, but sometimes it crashes. As there is no consistent crushing, I cannot understand where the problem is.
The inconsistency is interesting. Does it actually crash and give an error message or is it just slow? You can find error messages in your web browser’s developer console (look in the ‘view’ or ‘window’ menus).
I don’t know much about how webcam and audio uploading works for Pavlovia, in particular whether it’s trial-by-trial or all at the end, but it might also be a delay due to uploading. If so I would have no idea what to do about it.
No, it does not crush or give an error. At some point, there is nothing in the screen; it is all white. But the pictures seem to be still running because, after some time, it starts showing the pictures again. For example, the loop runs until the 5th picture, then the screen becomes only white for some time, and then it starts showing the 8th picture. I looked at the console; there are some warnings but no errors. Here are the pictures:
OK, no obvious problems in the logs. The behavior basically sounds like it’s not pre-loading all the images and instead trying to load them on the fly, and it’s just not loading them fast enough. I have seen that before with jsPsych if you don’t set up a preload system, but if you have the images in the “additional assets” list already then PsychoJS should automatically pre-load them. I would make sure ALL of the images are in the “assets” list.
You could also try making a copy of the experiment that’s the same except without the microphone and webcam responses, to see if is still happens without those. If not, it’s something to do with when it’s uploading data between trials, but I’m not sure how to address it.
Otherwise you could make your repository shareable and we can have a look at the experiment directly.
Hi, I re-uploaded all the items again and pushed the changes. I was able to complete the experiment without a problem, and I ran the experiment online with a few participants. Most participants either returned the experiment because the experiment glitched or did the experiment by waiting white screen to go away.
I am not sure if it is related to the camera. I have another experiment with a very similar design. That seems to run without a problem.
I have no idea why this is happening.