psychopy.org | Reference | Downloads | Github

Too many participants simultaneously for online experiments?

Dear all,

Last week Thursday, 26.9.2021, I was trying to test 19 participants simultaneously on 3 different tasks. However, when everybody started opening the experiments, that is a part of the group task 1, another part of the group task 2 and the last part task 3, the participants got error messages when trying to start the experiments. At this point I was also unable to log-in to my experimenter account on pavlovia.
After waiting for a bit and people starting the tasks one by one everything went back to normal and the tasks worked as they should.

My question is whether this is a problem that I have to anticipate for in the future when doing the same procedure with a group of 20 again.

I found this post from January 2020 where Jon says that he never experiences such problems when running workshops. So I’m not sure what to make of it for the future.

URL of experiment:

(may only put two links in a post as new user)

Best,
Andreas

Bildschirmfoto 2021-08-26 um 14.02.45

Dear @AndreasB ,

I am sorry to read that you encountered an issue on this occasion.

What happened was that I deployed a fix to the main server on the 26th of September, which regrettably altered the encoding of the answers given in response to queries to the pavlovia back-end server. As a consequence, those answers could not be properly decoded by the front-end (which explains why you could not see your dashboard) and the experiments (which explains the specific error message your participants got).
We noticed the problem rapidly and corrected the issue (see Accessing Pavlovia: SyntaxError: Unexpected token & in JSON at position - #3 by apitiot). This explains why your participants could run the experiments later on.

In other words, this did not have anything to do with running your participants simultaneously. I can assure you that running 20 participants again will not be an issue. pavlovia.org routinely handles many, many more simultaneous experiments.

With my apologies for the mishap,

Alain

Dear @apitiot ,
Thanks for the reply and clearing out what happend!

Best,
Andreas