Abstract
Purpose:
To characterize the slow phase of the torsion vestibulo-ocular reflex (tVOR) with respect to quick phases using high-speed video-oculography (VOG).
Methods:
Four normal volunteers were tested. TVOR was stimulated by 10°-40° rotations in the roll plane using a 6-DOF motion platform (MOOG Inc., East Aurora, NY) or using active quasi-sinusoidal head rotations, while subjects were asked to look straight ahead at a laser-projected target on a distant detailed visual scene. 3D eye movements were recorded with custom VOG goggles (based on RealEyes xDVR, Micromedical Tech., Chatham, IL) driven at 130 Hz and scleral search coils at 1 kHz. Head movement was recorded using a 9-DOF motion sensor (MTx , Xsens, Los Angeles, CA). Recordings were analyzed using Matlab (Mathworks, Natick, MA) and Iris Tracker (Chronos Vision, Berlin, Germany). All slow phases with preceding and following quick phases were selected for analysis. Values represent mean +/- std. dev. unless otherwise indicated.
Results:
Subjects showed ocular torsion rotation contraversive to active head rotation, with peak slow phase tVOR velocity gain of 0.73 ± 0.11 for active head roll. Slow phase velocity and amplitude were proportional to head velocity. Mean slow phase velocity gain between 50 and 300 ms was stable (0.67 ± 0.02) despite large variation in head velocity (38.34 ± 7.66 deg/s) and in slow phase velocity (25.39 +/- 6.30 deg/sec). At higher head velocities, ipsiversive quick phases were generated with rate, amplitude, and peak velocity proportional to head velocity. Mean torsion eye position was closer to zero at the beginning of the quick phase, 0.50 ± 3.91 deg, than after the quick phase, 6.86 ± 3.98 deg.
Conclusions:
Slow phase gain peaks after the quick phase then remains stable on average during active head roll despite large variations in head velocity. The quick phase rotates the eye away from the zero torsion position so that the following slow phase rotates the eye back to the zero torsion position. These results suggest that 1) the eye-in-head velocity is modulated throughout the duration of the slow phase to achieve a stable gain; and 2) the quick phase anticipates the slow phase amplitude, rather than corrects for it. These results are valid for only the head velocities indicated. Correlation of slow phase and preceding quick phase amplitudes suggests that both movements are programmed simultaneously.
Keywords: 522 eye movements •
622 ocular motor control •
752 vestibulo-ocular reflex