Abstract
Purpose :
Fixational eye movements (FEMs), comprising microsaccades, drifts and tremors, provide valuable information about oculomotor control and visual perception, and have been proposed as biomarkers for certain neurological and visual disorders. However, do the characteristics of FEMs change with fixation duration; and if so, what is the minimum duration for recording FEMs such that measurements of the characteristics of FEMs are stable?
Methods :
Eye movements were recorded while subjects (n=9, aged 21-35) fixated binocularly on a high contrast 1° cross target for multiple trials of 70s, using three eye-trackers with different spatial and temporal resolution: Eyelink II (250Hz), Eyelink 1000 Plus (1000Hz) and a scanning laser ophthalmoscope (SLO, eye positions sampled at 540Hz). After blink removal and denoising, a velocity-based algorithm was used to detect microsaccades and drifts in the eye-position traces throughout each 70s-trial. Characteristics of FEMs, including the rate and amplitude of microsaccades; duration and amplitude of drifts, were then determined for epochs that grew in in steps of 0.5s (0-0.5s, 0-1s, … 0-70s).
Results :
Averaged across subjects, the rate of microsaccades was constant across epoch length, and ranged between 0.96/s and 1.44/s across the eye-trackers. The amplitudes of microsaccades, and drifts and drift duration changed with epoch length, especially over the first 25s. The duration of drifts increased systematically with epoch length until it reached saturation. An exponential fit to each subject’s data yielded an average time constants of 0.65, 7.05 and 6.61s respectively for EyeLinkII, EyeLink 1000 Plus and SLO. The saturation level of drift duration differed across eye-trackers (0.63–1.02s). For the amplitudes of microsaccades and drifts, they changed with epoch length non-systematically, but became stable for epoch length beyond ~25s. The SLO yielded the lowest estimates of drift durations, amplitudes of microsaccades and drifts once the values reached a plateau.
Conclusions :
The characteristics of FEMs that we examined all showed stabilized values for epoch length over ~25s, suggesting that a minimum recording duration of 25s is required for us to reliably evaluate the characteristics of FEMs. The finding that these measurements differ with eye-trackers imply that the eye-tracker used should be considered when interpreting or comparing the characteristics of FEMs across studies.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.