Millisecond Forums

+x

 

AKrishna - 12/1/2022

Hi all,

a colleague is running a study using Inquisit Lab and has noticed odd behavior for some runthroughs. Specifically, on some screens, Inquisit doesn't respond to the keypress required by the trial. However, pressing Ctrl+Q to quit still works. No error message is shown; in all respects, the program behaves normally, except that the expected keypress does not cause the expected response. Data files sometimes record the final trial with the 'Ctrl+Q' response, sometimes they do not.

The problem does not appear systematic and cannot be consistently reproduced. Up to now, it has affected around 2-3% of runthroughs. My colleague reported running the test monkey for two hours nonstop and not reproducing the issue. I don't have direct access to the laptops running the studies or the script right now, but I thought I'd describe the problem in detail in case it fits any known issues. I'll try to get the script, Inquisit version, and precise laptop specs as soon as possible.

What we have observed so far:
- the problem appears to occur after the laptops have been running a while (after multiple sessions of the same experiment, which is fairly long)
- the laptops are generally running quite hot when the problem occurs
- up to now, it has only appeared on trials that used html-elements

What we have ruled out:
- the problem does NOT appear to be related to a background process (e.g., a notification from Windows) removing the focus from the Inquisit process, as apparently using Alt+Tab to ensure focus was on Inquisit did not fix the issue (although I could not observe this myself, as I was not present)

Does this sound in any way familiar or is there something in particular I should check while I wait for laptop specs and the script? Appreciate any input!

+x

 

Dave - 12/1/2022

+x

 

AKrishna - 12/1/2022

Hi all,

a colleague is running a study using Inquisit Lab and has noticed odd behavior for some runthroughs. Specifically, on some screens, Inquisit doesn't respond to the keypress required by the trial. However, pressing Ctrl+Q to quit still works. No error message is shown; in all respects, the program behaves normally, except that the expected keypress does not cause the expected response. Data files sometimes record the final trial with the 'Ctrl+Q' response, sometimes they do not.

The problem does not appear systematic and cannot be consistently reproduced. Up to now, it has affected around 2-3% of runthroughs. My colleague reported running the test monkey for two hours nonstop and not reproducing the issue. I don't have direct access to the laptops running the studies or the script right now, but I thought I'd describe the problem in detail in case it fits any known issues. I'll try to get the script, Inquisit version, and precise laptop specs as soon as possible.

What we have observed so far:
- the problem appears to occur after the laptops have been running a while (after multiple sessions of the same experiment, which is fairly long)
- the laptops are generally running quite hot when the problem occurs
- up to now, it has only appeared on trials that used html-elements

What we have ruled out:
- the problem does NOT appear to be related to a background process (e.g., a notification from Windows) removing the focus from the Inquisit process, as apparently using Alt+Tab to ensure focus was on Inquisit did not fix the issue (although I could not observe this myself, as I was not present)

Does this sound in any way familiar or is there something in particular I should check while I wait for laptop specs and the script? Appreciate any input!

Weird, but it might have something to do with the HTML. It's possible that the embedded browser frame takes input focus, thereby removing it from the app itself. Using the mouse to click the screen anywhere outside the embedded HTML should return input focus back to the app if that's the case.