June 2015
Volume 56, Issue 7
ARVO Annual Meeting Abstract  |   June 2015
Validation of the King-Devick Test Using 500-Hz Binocular Infrared Oculography
Author Affiliations & Notes
  • M Ali Shariati
    Ophthalmology, Stanford School of Medicine, Palo Alto, CA
  • Ming-Hui Sun
    Ophthalmology, Stanford School of Medicine, Palo Alto, CA
    Ophthalmology, Chang Gung Memorial Hospital, Kweishan, Taoyuan, Taiwan
  • Yaping Joyce Liao
    Ophthalmology, Stanford School of Medicine, Palo Alto, CA
  • Footnotes
    Commercial Relationships M Ali Shariati, None; Ming-Hui Sun, None; Yaping Liao, None
  • Footnotes
    Support None
Investigative Ophthalmology & Visual Science June 2015, Vol.56, 2927. doi:
  • Views
  • Share
  • Tools
    • Alerts
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      M Ali Shariati, Ming-Hui Sun, Yaping Joyce Liao; Validation of the King-Devick Test Using 500-Hz Binocular Infrared Oculography. Invest. Ophthalmol. Vis. Sci. 2015;56(7 ):2927.

      Download citation file:

      © ARVO (1962-2015); The Authors (2016-present)

  • Supplements

Purpose: Eye movement abnormalities are common in different neuro-ophthalmic conditions. Many eye movement measures use saccades and smooth pursuit tasks are used as markers of head trauma, and the King-Devick Test has been well used in studying head trauma in football, hockey, and other sports. In this study, we used 500-Hz infrared eye tracker to validate the hand-held King-Devick Test.

Methods: We performed the hand-held or computer King-Devick test using 500-Hz binocular 2D infrared oculography and used time to perform each tests (1-3) as outcome measure. For eye movement study, we analysed the fixations, saccades, and blinks.

Results: In normal individuals, mean duration of the King-Devick Test was shorter using the hand-held test (test-1: 13.5±6.5 s; test-2: 12.9±4.1 s; test-3: 13.3±6.7 s; N=9) than the computer test (test-1: 15.9±0.8, P=0.04; test-2: 15.7±0.6 P=0.001; test-3: 16.5±0.7, P=0.008; N=10). In normals, there was no significant difference in the time from test-1 to test-3 (P=0.5-0.9), although test-1 was the easiest test (guiding lines, generous spacing) and the test-3 was the hardest test (no guiding lines, smallest spacing). Although each test only had 40 fixation targets (5 numbers per line, 8 lines total), eye tracking revealed that each subject made 104±5.9 (range 62-132) fixations, which is more than twice the number needed. The number of saccades was 161±38 (range 97-466), which is on average more than twice more fixations. The average saccade amplitudes were test-1: 4.8±0.1, test-2: 4.9±0.3, test-3 5.0±0.4 deg), with saccade latencies of test-1: 194.0±24.8 ms, test-2: 197.5±28.1 ms, test-3: 213.6±21.0 and average saccade velocities test-1: 98.7± deg/s, test-2: 99.6±6.6, test-3: 104.8±5.0. The subjects that took the longest time to perform the task had the most number of saccades, likely because of backtracking.

Conclusions: We used 500-Hz binocular 2D infrared oculography to analyze eye movement behaviour during the King-Devick Test. Although there were 40 fixation targets per test, all subjects made at least twice as many fixation attempts. Shorter test time correlated with fewer saccades, blinks, and higher saccade amplitudes. The King-Devick saccade-based task is a useful test to assess eye movement abnormality in head trauma and various neuro-ophthalmic conditions, in order to assess why different subject populations perform differently.


This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.