Forum
Tobii Validity Codes
I have a question about possible redundancy of the Tobii validity codes.
First - I understand that (1,3) means that a valid x,y position was recorded for the left eye but not the right, and similarly, (3,1) means that a valid x,y position was recorded for the right but not the left. Do the (0,4) and (4,0) codes mean the same thing, or something different? If they differ, how?
-Daniel Egger
Duke University
Hi Daniel,
The validity code ranges from 0 to 4, with the following interpretations for each value:
0 The system is certain that it has recorded all relevant data for the particular eye, and that the data recorded belongs to the particular eye (no risk of confusing left eye with right eye by the system).
1 The system has only recorded one eye, and has made some assumptions and estimations regarding if the recorded eye is left or right. However, it is still highly probable that the estimations done are correct. The validity code on the other eye is in this case always set to 3.
2 The system has only recorded one eye, and has no way of determining if this is the left or the right eye.
3 The system is fairly confident that the actual gaze data is actually incorrect or corrupted. The other eye will always have validity code 1.
4 The actual gaze data is missing or definitely incorrect. A couple of gaze data with validity code 4 on both eyes, followed by a number of gaze data with validity code 0 on both eyes, are usually a sure sign of a blink.
It is recommended that the validity codes are always used for data filtering, to remove data points which are obviously incorrect. For most studies, we recommend removing all data points with a validity code of 2 or higher.
Questions like this might be more at home in the Tobii forums on http://www.tobiiforum.org/forums/default.aspx
Best Regards
Patrik Lindholm
Thanks so much Patrik - And I would be happy to post my questions elsewhere. I made up my own post because it was not clear to me which of the many default forum threads above would actually fit these type of questions best. Can you recommend a specific one?
In the mean time, we are unsure how to evaluate a large number of (0,0) validity code readings where the L eye x,y and R eye x,y pixel positions are far apart. We are getting many time stamps with a greater than 100 pixel distance between the two positions - some over 500 pixels - and this would seem to be larger than the expected measurement error range.
In your opinion, does this perhaps reflect some normal eye physiology -- lack of bifocal focus 5-15% of the time - or should be discard even (0,0) data when this distance reaches a certain maximum thresh-hold?
Hi Daniel,
Please post in this forum to get faster response on your questions.
http://www.tobiiforum.org/forums/default.aspx
Or send an email to support@tobii.com
Best Regards
Patrik Lindholm