Behavioural characteristics of the quick phase of vestibular nystagmus before and after unilateral labyrinthectomy in guinea pig

Abstract

The aim of this paper was to characterise the conditions under which the quick phase of vestibular nystagmus is generated in response to sinusoidal horizontal angular accelerations in guinea pig and to determine whether the characteristics of the quick phase are altered following unilateral vestibular deafferentation (UVD). In experiment 1, the quick-phase response to 2-Hz sinusoidal stimuli with different peak head velocities was measured before and after UVD. In experiment 2, lower-frequency stimuli with a fixed amplitude (±20°) were used to measure the eye-movement response. In experiment 1, at 2 Hz, the most noticeable difference between UVD animals and normal animals was a reduction in the number of quick phases generated, particularly when rotating towards the lesioned side: the onset of the quick phase was delayed and occurred at a lower value of peak head velocity compared to normals. However, both these measures probably reflect the depressed slow-phase eye-velocity gain rather than a change in the quick-phase mechanism itself, because if a quick phase was generated there was no difference between UVDs and normals on a variety of measures (duration, position, peak eye velocity) for a 2-Hz stimulus. For both UVD and normal animals there did appear to be a position threshold for the onset of a quick-phase eye movement (approximately ±7.5°), although it should be noted that the threshold is not an absolute value. In experiment 2, with lower-frequency stimuli, the characteristics of the quick phase itself were altered in UVD animals. At these lower accelerations, although the head velocity and average eye position for the onset of the quick phase did not differ between UVD and normal animals, there were significant differences between UVD and normal animals in the number of quick phases generated (fewer), the duration of the quick phase (longer) and the peak eye velocity of the quick phase (slower). For lower-frequency stimuli there was no evidence of a specific eye-position threshold for the generation of a quick phase, although the position at which a quick phase occurred rarely exceeded the ±7.5° value obtained in experiment 1. The behavioural data were used to produce a biologically based neural-network simulation of both the slow- and quick-phase components of the vestibulo-ocular reflex, the results of which are presented in a companion paper.