Eye-tracking and physio integartion

[qrtmnopa197]

Hi!

I was wondering if there’s any established method for integrating experiments with eye-tracking and/or physiological measures (e.g. for sending parallel port codes from the task computer to a computer connected to BIOPAC). I didn’t see any elements that seemed to serve that purpose, so I thought I’d check to see if there was any other way it could be done.

Thanks!
Daniel

P.S. I originally, accidentally, posted this in the FAQ section. I hope that can be easily removed; sorry about that!

[Jeremy]

Hi Daniel,

I am not familiar with BIOPAC, and I’m not sure I understand your question. There is an EyeTracker element in PennController, but it is not very accurate and I haven’t spent much time developing it, which is why I never documented it. The way it currently works, it sends the data it collects to a distant URL, so you’d need to set up some interface to retrieve your data on some server, in much the same way that collecting audio recordings works (though the method is not fully transposable). Everything else is saved in the results file of your experiment.

Feel free to elaborate on your questions

Jeremy

[qrtmnopa197]

Hi, Jeremy.

Thanks for the response. For some of the experiments we run, we collect data on event-locked physiological responses to stimuli (e.g., an increase in skin conductance when a particular image is presented). In our current in-lab set-up, the task computer sends a trigger to a nearby physio-data-collecting computer to tell it exactly when a stimulus is presented, so the event-related skin conductance response can be accurately measured. We’re trying to find software that can be used to run experiments online and in-lab, but, to run experiments in-lab, we’d need the software to be able to send signals to a nearby computer collecting physiological data, very rapidly, rather than just sending data to a distant URL. I was curious if there’s any way to do this through PennController, though it sounds like it might not be designed for that sort of thing.

Thanks
Daniel

[Jeremy]

I see. Because of the very nature of PCIbex, hardware communication seems complicated, at best, so the most promising implementation I see would still need to use URLs.

In any case, as you said, I don’t think PennController would be the right tool to collect this kind of data, in this kind of setup. Not only is it ill-designed for wired communication, it’s also javascript-based, which means a time resolution of tens of milliseconds at best.

I’m also curious as to how you measure an increase in skin conductance in an online setting: do you specifically recruit participants who have access to a machine that can collect this sort of physiological data? Are they students in a program?

Jeremy

[qrtmnopa197]

We wouldn’t be able to collect skin conductance online, we’d have to do it in the lab. We’re just looking for experiment software that could work for both online and in-lab experiments, so we could collect some data in-lab and then get a larger, purely behavioral, sample through an online experiment. Regardless, thank you for your help!

[Author]

Posts