Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Supported by

regions-of-interest in opensesame

Hello everyone!

I'm new to OpenSesame, and I'm doing an eye-tracking study(use SMI).In my experiments, pictures appear in fixed positions on the left and right sides of the screen. I want to define regions-of-interest where the pictures appear and eventually count the number of gazes within the two regions of interest.

Thank you very much !


Comments

  • Hi,

    Sounds good. There are plenty ways of doing it. The easiest would probably be to do it offline after data collection. That is only an option if the gaze behaviour does not need to interact with the stimulation (e.g. no gaze-contingency). If you want to do it during the experiment, you need to have the coordinates of the images, and compare the current gaze position to these coordinates. For that you will need pygaze functions. If you don't care about some inaccuracies, the easiest would be to assume a circular ROI around the images. Then you can work with the distance to its middle to check which item was fixated.

    Does that help?

    Eduard

    Buy Me A Coffee

  • Thank you very much for your reply. Since I have not used opensame before, is it possible to use the data analysis software begaze that comes with SMI to analyze the data after using opensame for the experiment? Or do I need to use pygaze

    Thank you for helping,

    wx

  • Hi wx,

    I just mentioned Pygaze because it is the most common way to link eye tracking to Opensesame, but as long as a software has a Python API, you can also use that one. I am not familiar with begaze, but normally this should be possible (certainly if you do the ROI analysis offline). The only thing you should make sure to do is to link Opensesame's behavioral file with SMI's eye tracking logfile. So that you know which eye movements belong to which trial.

    I hope this clears things up

    Eduard

    Buy Me A Coffee

  • thanks for your help

    Best wishes

    wx

Sign In or Register to comment.