mdl-eyetracking 2019-04-11
  • Glossary
  • Install
  • Site
      • eyetracking
      • calibration
      • installation
      • pylink
      • eyelink
  • Page
      • How Eyelink works
        • heuristic filtering algorithm
        • velocity and acceleration thresholds

How Eyelink works¶

The EyeLink trackers use a velocity-based method for eye event parsing. Saccade onset is detected when the eye movement velocity or acceleration exceeds a pre-set threshold, whereas saccade offset is detected when the velocity and acceleration fall below the thresholds. The end of a saccade is the start of a fixation and the start of a saccade is the end of a previous fixation. Since this velocity-based method detects the onset and offset of saccades, it is usually known as a “saccade-picker” (Eyelink forum).

heuristic filtering algorithm¶

The eye movement data recorded in the data file, by default, has been filtered with a heuristic algorithm (Stampe, 1993). For more information on the heuristic filtering algorithm, see (Stampe, 1993).:

Stampe, D. (1993). Heuristic filtering and reliable calibration methods for video-based pupil-tracking
systems. Behavior Research Methods, Instruments, & Computers, 25(2), 137–142. doi:10.3758/BF03204486

velocity and acceleration thresholds¶

For the velocity and acceleration thresholds see the EyeLink 1000 Plus User Manual.:

EyeLink 1000 Plus User Manual (2013). EyeLink 1000 Plus. SR Research Ltd., Mississauga, Ontario, Canada.

Back to top

© Copyright 2019, Semeon Risom.
Created using Sphinx 1.8.5.