Compensatory head and eye movements in the frog and their contribution to stabilization of gaze.

Compensatory head movements, recorded in unrestrained frogs, were compared to compensatory eye movements recorded from animals that had their head fixed. Movements were evoked by oscillating the animal in the dark (vestibular stimulation) or in the light in front of an earth-fixed, patterned visual background (combined stimulation) or by rotating vertical black and white bars (optokinetic stimulation) around the stationary animal. Oscillations occurred in the horizontal plane at frequencies between 0.025 and 0.5 Hz. Gain and phase values of head and eye movements, relative to stimulus movements were calculated. Evoked eye movements were limited in amplitude to +/- 3-6 degrees, increasing with the size of the animal. Head movements were limited to +/- 30-40 degrees. Resetting fast-phases of both head and eyes were very rarely observed during sinusoidal stimulation and no eye movements were recorded in the absence of intended head movements. Vestibularly evoked head movements exhibited a frequency-dependent threshold that was not observed for vestibulo-ocular responses. Above threshold, the gain of evoked head responses increased and reached a frequency-dependent plateau at which the system behaved approximately linearly. Within the linear range, gain of vestibularly evoked responses increased with frequency (from 0.04 at 0.025 Hz to 0.75 at 0.5 Hz) and phase lead decreased (from about 80 degrees to 0 degrees). Vestibularly evoked eye movements similarly increased in gain from 0.05 to 0.56 and decreased in phase lead from about 56 degrees to 10 degrees over the same frequency range. Optokinetically evoked head and eye movements had their highest gains (about 0.8 and 0.5) at low constant velocities (less than or equal to 1-4 degrees/S) or frequencies (less than or equal to 0.025 Hz). At higher constant velocities or frequencies the gain dropped. The phase lag increased from close to zero (at 0.025 Hz) to about 60 degrees for the head and to about 20 degrees for the eye movements (at 0.25 Hz). These phase lags are explained by reaction times of the evoked movements of about 600 ms (head) and 200 ms (eyes). Combined stimulation evoked compensatory head movements with gain and phase values that were frequency-independent in the linear range. Head movements compensated for about 80-90% of the imposed gaze shift with a small phase lag (0-10 degrees). Evoked eye movements were found to be large enough in amplitude and fast enough in time to enable a frog to stabilize its gaze exclusively with slow phase compensatory movements for a large variety of frequency and amplitude combinations. The two motor systems controlling movements of the head and the eye are matched in such a way that the non-linearities of the evoked eye movements can compensate for the non-linearities of the evoked head movements.