Howdy, Stranger!

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

Supported by

Eye-tracking - Setting physical screen size and distance to participant

In PyGaze, the physical display size in centimeters is given by the variable SCREENSIZE (e.g., SCREENSIZE = (31.0, 17.4)) and the distance from the participant to the display, also in centimeters, is given by SCREENDIST (e.g., SCREENDIST = 60.0).
I don't know exactly how relevant these variables are but, as far as I could determine, there is no direct way to configure their values using PyGaze's OpenSesame plugins.
I assume they can be set using an inline script but I haven't looked into that yet.

At the top of the eye-tracking output files (I'm using a Tobii eye-tracker), in the initiation and calibration reports, I have the information "display size in cm: 33.8x27.1" and "distance between participant and display: 57.0 cm".
I don't know if these values are being calculated somehow; it seems to me that these are just the default values; but their are not correct (my current display size is about 31.0x17.4 cm and the distance should be between 60 and 65 cm.

Are these variables relevant?
Should I set them to the correct values using an inline script and, if so, how should I go about it?

My thanks in advance!
Best regards,


  • An inline script as follows seems to do the trick.

    from pygaze import settings
    settings.SCREENSIZE = (31.0, 17.4)  # physical display size (cm)
    settings.SCREENDIST = 60.0          # participant to display distance (cm)
  • edited October 2017

    Hi @brusil ,

    Did you find any consequences of leaving these parameters as default on your eye-tracking data? Do you have to set these parameters correctly to get the correct coordinates? Also calling @sebastiaan here!


  • edited October 2017

    Hi @hanzh,

    Since I'm new at this and I haven't yet been able to analyse my test data, I can't tell if these have consequences, but I don't think they do.
    Still, since they're printed to the eye-tracking data files, I prefer their value to be as close as possible to the reality.
    Probably @sebastiaan and @Edwin can tell for sure if this is of any relevance.


  • The consequence of not changing them depends on how you analyze the data. For example, if you define regions of interest, based on degrees of visual angle, you will have incorrect regions when the screen parameters are off. Perhaps, it might also affect the parsing of saccades and fixations (particularly, for small saccades), but that depends on how these algorithms are implemented in Pygaze.

    Hope this helps.


Sign In or Register to comment.