Home › Forums › Eye Tracking Devices › Pupil diameter correction based on gaze location

Tagged: , ,

Viewing 6 posts - 1 through 6 (of 6 total)
  • Author
    Posts
  • 19/11/2015 at 17:06 #3797
    milad
    Participant

    Moving the gaze location on the screen, induces fluctuations on the raw value of the pupil diameter. Does Tobii-X60 apply any correction on the diameter of the pupil based on the gaze location? If it does, is that based on fitting with the initial calibration (in my case a 5 point calibration) or an internal formula is used?

    03/12/2015 at 21:53 #3865
    Tobias Lindgren [Tobii]
    Participant

    Hi Milad,

    I assume you are using the Tobii Analysis SDK since you ask about the Tobii Pro X60 eye tracker and also about pupil diameter which is only available in the Analysis SDK. To give a relevant answer to your question I need to know a bit more about the circumstances. Could you please provide more information about:
    1) Is Tobii X60 the exact eye tracker model name (I ask since we have other products with similar names, e.g. the X2-60 eye tracker)?
    2) How much does the pupil diameter fluctuate and in what way (e.g. slowly drifting or random noise)?
    3) Do you experience the fluctuations also when moving your gaze around a screen with uniform brightness? The pupil diameter is affected by the brightness so looking at a bright spot will cause the pupil diameter to decrease.

    03/12/2015 at 22:19 #3866
    milad
    Participant

    Hi, many thanks for your answer, Yes, our Tobii, is X60, I always export the data file as .tsv and import it by matlab to get the pupil diameter. Our stimuli are two iso-luminance faces on the two sides of the screen, and it’s important for me to know if there is a correction on the pupil diameter when the participant (infants in our experiments), look off the center, from one side to the other. Because this problem has been reported in Tobii and other eye trackers (trying to find the paper I was reading a while ago), but I didn’t do a comprehensive analysis for this purpose on my data yet.

    Thanks again for your help!
    Milad

    04/12/2015 at 16:50 #3870
    Tobias Lindgren [Tobii]
    Participant

    Exactly how the pupil diameter is calculated differs between different eye tracker models but in general the pupil diameter calculation should compensate for different gaze angels and there should be no substantial fluctuations. But as always I suggest that you make some test recordings and make sure you are able to extract the data and conclusions you want before recording a large number of subjects.

    08/12/2015 at 17:07 #3878
    milad
    Participant

    Thank for your answer, but does this mean that Tobii, X60 applies corrections based on location of gaze or does not apply a correction? And if it does, is it based on a formula or factory pre-set calibrations?

    Thank you in advance!
    Milad

    09/12/2015 at 15:41 #3880
    Tobias Lindgren [Tobii]
    Participant

    The calibration process enables the eye tracker to calculate a physical model of the test subject’s eyes (one model per eye). This model is used to calculate the pupil diameter for each eye and takes into account both the gaze angle and the eye position.

  • Author
    Posts
Viewing 6 posts - 1 through 6 (of 6 total)
  • You must be logged in to reply to this topic.