SR Research EyeLink Portable Duo User Manual page 72

within the bounds of the display or loss of tracking may occur because they
have looked too far outside of the trackable range of the eye tracker (not
because of a poor set-up). Watch for the warning signals on the tracker
screen to make sure that the pupil and CR signals are not lost when the
participant is doing so, and check the CR is not becoming distorted or
"smeared" when the participant looks at the top corners. Try moving the
monitor away to increase the viewing distance and raising the camera when
the CR smearing is seen (typically at the upper portion of the display).
• Participants who have never been calibrated before may require some
practice in stably and accurately fixating the calibration targets. If the result
of the first calibration is not optimal, try to perform at least two calibrations
per participant before beginning to collect data.
• For the Remote Mode, use the 13-point calibration for the best accuracy. For
the Head-Stabilized Mode, use the 9-point calibration type.
• Always check the pattern of the calibration grid. For a 9-point calibration,
the fixation crosses should form three parallel horizontal (or close-to-
horizontal) lines and three parallel vertical (or close-to-vertical) lines. Redo
the calibration or camera setup if you do not see this.
• If the current calibration looks good, press either the ENTER key to accept
the calibration or press V to go to the validation screen. Never press the ESC
key – doing so will discard the current calibration and thus revert to the
cached calibration results.
• Encourage participants to sit still! A participant that doesn't sit still probably
is not paying proper attention to the experimental task. Try to give the
participant a short break in the middle of the experiment and recalibrate
before resuming the experiment.
• When writing your own applications, try matching the background color of
the calibration and validation screen to that of the experimental displays.
Changes in pupil size caused by large changes in brightness between the
calibration and the experimental displays will degrade the system accuracy.
At the beginning of the experiment, let the participant adapt to the
environment and the ambient light levels before performing calibration and
data collection. If the illumination levels are altered (i.e. the lights are
dimmed) shortly before the experiment begins, the calibration accuracy will
be reduced as the participant adapts to the new illumination level and the
pupil dilates or constricts.
62 An EyeLink Portable Duo Tutorial: Running an Experiment
© 2016-2017 SR Research Ltd.