search

dcnieho / Titta

Matlab and PsychToolbox interface to Tobii eye trackers using Tobii Pro SDK
Other
42 stars 15 forks source link

Understanding the output file #41

Closed ksgfan closed 2 years ago

ksgfan commented 2 years ago

Hi,

is there any Wiki about the meaning of variables in the output .mat file?

I am having trouble to understand, why there are 3 variables in dat.data.gaze.left.gazePoint.inUserCoords and what is their meaning (why the correlation between 2nd and 3rd is equal to 1)? What is the difference between dat.data.gaze.left.gazeOriginand dat.data.gaze.left.gazePoint? How would you compute fixations and saccades from the data?

Best, Dawid

dcnieho commented 2 years ago

Hi Dawid,

What are the 2nd and 3rd variables you are referring to?

The various data points are defined in the tobii documentation here: https://developer.tobiipro.com/commonconcepts.html. I'll add a note to the documentation.

As for how to do fixation classification with these output files, see /demos/readme_analysis_example in the titta distribution.

Cheers, Dee

On Mon, Aug 15, 2022 at 11:38 AM ksgfan @.***> wrote:

Hi,

is there any Wiki about the meaning of variables in the output .mat file?

I am having trouble to understand, why there are 3 variables in dat.data.gaze.left.gazePoint.inUserCoords and what is their meaning (why the correlation between 2nd and 3rd is equal to 1)? What is the difference between dat.data.gaze.left.gazeOriginand dat.data.gaze.left.gazePoint? How would you compute fixations and saccades from the data?

Best, Dawid

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>

ksgfan commented 2 years ago

many thanks! :)

I meant the dat.data.gaze.left.gazePoint.inUserCoords(2, :) and dat.data.gaze.left.gazePoint.inUserCoords(3, :), which are perfectly correlated.

dcnieho commented 2 years ago

Hmm, i assume that they are correlated but not the same, correct? That makes sense when the screen plane is slanted with respect to the coordinate system of the eye tracker, then any change in vertical gaze angle (so Y coordinate of gazePoint on the screen in the user coordinate system) will also yield a change in Z coordinate, viewing distance)

ksgfan commented 2 years ago

Their correlation is 1. corr(dat.data.gaze.left.gazePoint.inUserCoords(2, :)', dat.data.gaze.left.gazePoint.inUserCoords(3, :)', 'rows', 'complete')

ans =

    1.0000
dcnieho commented 2 years ago

Are the values the same? If the values makes sense that's not weird

On Mon, Aug 15, 2022, 18:47 ksgfan @.***> wrote:

Their correlation is 1. corr(dat.data.gaze.left.gazePoint.inUserCoords(2, :)', dat.data.gaze.left.gazePoint.inUserCoords(3, :)', 'rows', 'complete')

ans =

1.0000

— Reply to this email directly, view it on GitHub https://github.com/dcnieho/Titta/issues/41#issuecomment-1215349267, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANUOGJAC6DEPH73CW4HINLVZJYCTANCNFSM56RVODXA . You are receiving this because you commented.Message ID: @.***>

ksgfan commented 2 years ago

No, the values are not the same and make sense. Thanks!