Quick presented components don't appear with indefinite components

[balbir-singh]

URL of experiment: SF_OD [PsychoPy]

Description of the problem: I am posting this as a new topic since the previous one did not have complete information.

Background: I am trying to have people do a categorization task. After a 200 ms ITI the trial starts with a fixation cross (500 ms) followed by an image (200 ms). Ideally the participants has an unlimited time to answer via a keyboard component. I built a very basic version of my experiment in builder v.2024.2.3.

In the basic version I built three loops. Each loop has the same ITI, fixation cross and image (total of 900 ms).

In the first loop there are only two components, the fixation cross and the image. Each loop automatically terminates after 900 ms. This loop works perfectly. So I don’t think it is a problem with the images, the frame rate, or the browser.

In the second loop there are three components, the two quick presented components and a keyboard component (indefinite duration). In this loop nothing appears on the screen. However, if I click spacebar 16 times (the number of loops), then I can advance.

In the third loop, I also included three components. The two quick presented components and a label that is presented indefinitely. Because of the indefinite duration, this loop will not advance. However, in this loop, the fixation cross and the image still don’t appear!

It runs well locally via python. When I run it locally via js or on pavlovia, neither work. I am not sure I understand why indefinite duration components would prevent the presentation of quick presented components?

Below are some screenshots of my flow so you better understand the structure.

My flow:

The first loop just shows the fixation cross and the image and terminates after 0.9 seconds automatically. This loop works perfectly.

The second loop has the same images and text but also has the keyboard component which is indefinite till the space bar is pressed. In this loop none of the quick stimuli are presented. If you press the spacebar 16 times you will be able to advance though.

[stanley1O1]

Dear Balbir-Singh,

Could you provide a copy of your experiment files for me to look at? Like the builder file and the excel file (and any other material needed to run it).
Issac

[balbir-singh]

Hi @stanley1O1,

Thanks! Here are the files you would need. Additionally, here is a link to the gitlab repository which should be public: https://gitlab.pavlovia.org/bsingh/sf_od

SF_OD.psyexp (38.6 KB)
prac_image_file.xlsx (8.8 KB)
Link to zip file for images

[stanley1O1]

Thanks for that.

After looking at your experiment, try giving this a look.

PsychoPy Python to Javascript crib sheet

Many people come to this forum because their experiment works fine locally but doesn’t work online. The reason is that the Python code needs to be translated to Javascript and the auto translate that was added to PsychoPy in January 2020 primarily translates the grammar rather than the individual functions. If you want to run PsychoPy online I highly recommend that you: Write your experiment in Builder using version 2023.1.3 (or later). Study my crib sheet to familiarise yourself with the…

As at some point some things from python didn’t always translate properly to java. I am not sure how that has changed in the more recent version of psychopy, but one of the “cheatsheet” versions you can see certain problems occur due to key responses on page 12:

PsychoPy Python to Javascript crib sheet 2023

Try some those fixes and maybe those problems will resolve?
Issac

[balbir-singh]

Unfortunately, the fixation cross and image are still not appearing. I read through the crib sheet and changed the response options on the keyboard component to not just include space. I also reduced the size of the images incase it was taking time to load (though since everything is being downloaded prior to the experiment this seems unlikely).

Since the first loop is working, I decided to change the flow so one routine would present the fixation cross and the image and another routine would gather the response.

I also recorded my screen during the loop. The recording shows that image is being pulled up, but weirdly nothing is showing up on the screen!

[wakecarter]

Hi. This is I think due to a bug in translating non-slip timing to PsychoJS in the current version of PsychoPy. There’s a fix in the works but for now your options are either to use an earlier version (e.g. 2024.1.4 – you might be able to just select this in Experiment Settings / Use version) or try removing the duration from prac_image_stim and instead adding a routine duration of 1 second in the routine settings. I have had some success with this solution but it doesn’t fix everything.

[balbir-singh]

Thanks! That did the trick. I changed the version in Experiment Settings and everything worked.