Vertical divergence and counterroll eye movements evoked by whole-body position steps about the roll axis of the head in humans

In healthy human subjects, a head tilt about its roll axis evokes a dynamic counterroll that is mediated by both semicircular canal and otolith stimul ation, and a static counterroll that is mediated by otolith stimulation onl y. The vertical ocular divergence associated with the static counterroll to o is otolith-mediated. A previous study has shown that, in humans, there is also a vertical divergence during dynamic head roll, but this report was n ot conclusive on whether this response was mediated by the semicircular can als only or whether the otoliths made a significant contribution. To clarif y this issue, we applied torsional whole-body position steps (amplitude 10 degrees, peak acceleration of 90 degrees /s(2), duration 650 ms) about the earth-vertical (supine body position) and earth-horizontal (upright body po sition) axis to healthy human subjects who were monocularly fixating a stra ight-ahead target. Eye movements were recorded binocularly with dual search coils in three dimensions. The dynamic parameters were determined 120 ms a fter the beginning of the turntable movement, i.e., before the first fast p hase of nystagmus. The static parameters were measured 4 s after the beginn ing of the turntable movement. The dynamic gain of the counterroll was larg er in upright (average gain: 0.48 +/- 0.10 SD) than in supine (0.36 +/- 0.1 0) position. The static gain of the counterroll in the upright position (0. 21 +/- 0.06) was smaller than the dynamic gain. Divergent eye movements (in torting eye hypertropic) evoked during the dynamic phase were not significa ntly different between supine (average vergence velocity: 0.87 +/- 0.51 deg rees /s) and upright (0.84 +/- 0.64 degrees /s) positions. The static verti cal divergence in upright position was 0.32 +/- 0.14 degrees. The results i ndicate that the dynamic vertical divergence in contrast to the dynamic ocu lar counterroll is not enhanced by otolith input. These results can be expl ained through the different patterns of connectivity between semicircular c anals and utricles to the eye muscles. Alternatively, we hypothesize that t he small dynamic vertical divergence represents the remaining vertical erro r necessary to drive an adaptive control mechanism that normally maintains a vertical eye alignment.