Running experiments requiring precise timing

Online experiments
,
1

Hello, I am planning to run an experiment on touchscreen devices where precise timing is necessary. I was wondering whether this would be feasible to run it online on pavlovia. Bascially we are interested in participants’ perception on display lag. We will need to manipulate the lag between participants’ response (e.g. press a button on a touchscreen) and the point when the button starts to change shape. I understand that there is a lag which could not be completely eliminated. We want to do it online as we would need to collect a large amount of data within a limited time window. We could tolerate a certain amount of timing inaccuracies (probably less than 60 ms error) as long as they are not too variable (i.e. variability less than ± 10 ms). Would timing issues would a problem in this case? How would different devices, systems, and browsers make a difference?

Any suggestions would be much appreciated!

Best,

Mengxin

2

Device differences and differences in user mechanical responses would probably be the biggest thing. More so the former than the latter.

You would have to account for differences in response mechanics (are they using thumbs or index finger; handedness), user response latency (specifically how far away are the users’ fingers from the device), device response latency (how long before the device registers a touch when the user touches the screen), screen refresh rate (hz, how often pixels update), and screen latency time (time it takes for pixels to change).

To control or account for most these in a online experiment would be extremely difficult.