I’m wondering if it is theoretically possible to generate an experiment in which some of the stimuli are provided by the participant. I’d like to have online participants upload 4 different images (personal photos). Their own photos would then be used as stimuli for two different trial types, so each of the two trial types has two possible images for display). Is this something I could do in PsychoPy?
Alternatively, if this is not feasible, is it possible for participants to provide two names, which are then called in subsequent trials where the stimulus is a text prompt (as an alternative to using personal photos as stimuli), e.g., they’d see text starting with “Think about [USER-PROVIDED NAME]…”
I’ve done this before with a dot probe task. I manually uploaded the images with specific filenames based on the participant number (e.g. 1.jpg and 1s.jpg) and then loaded them dynamically in PsychoPy based on expInfo[‘participant’]
Correct – it was for a daily training intervention so the participants had already taken part in a self reference study and a third of them were then asked to supply photos of themselves (smiling and sad).
I see. That’s good to know! For the Prolific/online recruitment version of this task I was hoping to have participants be able to upload the photos themselves and have those automatically called by the correct trial type, so I’m not involved in coordinating task administration with them. But I was wondering about the kind of scenario you describe as well, for a different version of the experiment with a clinical sample.
The other possibility (which may get easier in the near future) is to use the webcam to take a photo and then present that back to the participant later. I haven’t done this myself, but this demo might help.
Oh, very cool, that looks like an interesting option if they have a way to hold their photos up to the webcam…
The photos I want to use are actually photos not of the participant themself, but photos of loved ones that they have already (sorry if that wasn’t clear!)
So theoretically, it is possible to let the particpants “upload” their images on the fly. Keep in mind, that the images are just locally “present in the browser” and are not actually uploaded to the server. And it is important that the participants “do it the right way” (its not really error-proof).