Advanced

Do pupil-based binocular video eye trackers reliably measure vergence?

Hooge, Ignace T.C. LU ; Hessels, Roy S. and Nyström, Marcus LU (2019) In Vision Research 156. p.1-9
Abstract

A binocular eye tracker needs to be accurate to enable the determination of vergence, distance to the binocular fixation point and fixation disparity. These measures are useful in e.g. the research fields of visual perception, binocular control in reading and attention in 3D. Are binocular pupil-based video eye trackers accurate enough to produce meaningful binocular measures? Recent research revealed potentially large idiosyncratic systematic errors due to pupil-size changes. With a top of the line eye tracker (SR Research EyeLink 1000 plus), we investigated whether the pupil-size artefact in the separate eyes may cause the eye tracker to report apparent vergence when the eyeballs do not rotate. Participants were asked to fixate a... (More)

A binocular eye tracker needs to be accurate to enable the determination of vergence, distance to the binocular fixation point and fixation disparity. These measures are useful in e.g. the research fields of visual perception, binocular control in reading and attention in 3D. Are binocular pupil-based video eye trackers accurate enough to produce meaningful binocular measures? Recent research revealed potentially large idiosyncratic systematic errors due to pupil-size changes. With a top of the line eye tracker (SR Research EyeLink 1000 plus), we investigated whether the pupil-size artefact in the separate eyes may cause the eye tracker to report apparent vergence when the eyeballs do not rotate. Participants were asked to fixate a target at a distance of 77 cm for 160 s. We evoked pupil-size changes by varying the light intensity. With increasing pupil size, horizontal vergence reported by the eye tracker decreased in most subjects, up to two degrees. However, this was not due to a rotation of the eyeballs, as identified from the absence of systematic movement in the corneal reflection (CR) signals. From this, we conclude that binocular pupil-CR or pupil-only video eye trackers using the dark pupil technique are not accurate enough to be used to determine vergence, distance to the binocular fixation point and fixation disparity.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
Accuracy, Binocular, Eye tracking, Pupil, Vergence
in
Vision Research
volume
156
pages
9 pages
publisher
Elsevier
external identifiers
  • scopus:85060252215
ISSN
0042-6989
DOI
10.1016/j.visres.2019.01.004
language
English
LU publication?
yes
id
7c155423-c92e-4077-8333-696f2f371569
date added to LUP
2019-01-28 12:23:41
date last changed
2019-02-27 05:11:14
@article{7c155423-c92e-4077-8333-696f2f371569,
  abstract     = {<p>A binocular eye tracker needs to be accurate to enable the determination of vergence, distance to the binocular fixation point and fixation disparity. These measures are useful in e.g. the research fields of visual perception, binocular control in reading and attention in 3D. Are binocular pupil-based video eye trackers accurate enough to produce meaningful binocular measures? Recent research revealed potentially large idiosyncratic systematic errors due to pupil-size changes. With a top of the line eye tracker (SR Research EyeLink 1000 plus), we investigated whether the pupil-size artefact in the separate eyes may cause the eye tracker to report apparent vergence when the eyeballs do not rotate. Participants were asked to fixate a target at a distance of 77 cm for 160 s. We evoked pupil-size changes by varying the light intensity. With increasing pupil size, horizontal vergence reported by the eye tracker decreased in most subjects, up to two degrees. However, this was not due to a rotation of the eyeballs, as identified from the absence of systematic movement in the corneal reflection (CR) signals. From this, we conclude that binocular pupil-CR or pupil-only video eye trackers using the dark pupil technique are not accurate enough to be used to determine vergence, distance to the binocular fixation point and fixation disparity.</p>},
  author       = {Hooge, Ignace T.C. and Hessels, Roy S. and Nyström, Marcus},
  issn         = {0042-6989},
  keyword      = {Accuracy,Binocular,Eye tracking,Pupil,Vergence},
  language     = {eng},
  pages        = {1--9},
  publisher    = {Elsevier},
  series       = {Vision Research},
  title        = {Do pupil-based binocular video eye trackers reliably measure vergence?},
  url          = {http://dx.doi.org/10.1016/j.visres.2019.01.004},
  volume       = {156},
  year         = {2019},
}