How to set the stimuli size/visual angle when conduct online experiment

tong

Hi, I have a question.

When conduct an experiment online by opensesame, the screen of participant might not the same. In this case, how the visual angle could be fixed regardless their screen size.

If someone have any suggestion or solution, please tell me, thank you!

lvanderlinden

Hi @tong

The dimensions of your participants' displays are stored in the built-in variables width and height. By default, they are logged to the output file. If you use a customized logger item, you could add these variables manually.

Also, you can use the values of these variables in your experiment like so:

• In the interface, by using the square-bracket method:

[width]
[height]

• In an inline_javascript item, by using the vars object:

console.log("The width of the screen is " + vars.width)
console.log("The height of the screen is " + vars.height)

For more information about the vars object, see this article:

• https://osdoc.cogsci.nl/3.2/manual/osweb/

Does this help?

(Of course, in order to convert pixels to visual degrees accurately, you also need to know the distance from the participants' eyes to the monitor, of which you don't have control when running experiments online...)

Cheers,

Lotje

tong

Thank you for your answer. I check the Opensesame output. Is it the default you say is the resolution? But even the resolution is fixed, would the stimuli preseted size fixed either? I think it will change with the screen size.

Again, thank you very much for your answer.

eduard

Hi Tong,

To make sure that you stimulus size are the same for all participants, you would indeed need to define them (e.g. so their position and size, etc.), relative to local variables (e.g. users' screen width and height).

Furthemore, if visual angle is really important, you should also instruct participants to do the task in a certain distance to the screen. 60-90 cm can make quite a noticeable difference!

All that said, there will always be some residual risk that participants' settings won't be exactly those that you intend them to be, unfortunately, but by being able to collect a much larger sample size, outliers tend to have less influence.

Hope this helps,

Eduard