The oculogyral illusion: retinal and oculomotor factors.

Subjects in a dark chamber exposed to angular acceleration while viewing a head-fixed target experience motion and displacement of the target relative to their body. Competing explanations of this phenomenon, known as the oculogyral illusion, have attributed it to the suppression of the vestibulo-ocular reflex (VOR) or to retinal slip. In the dark, the VOR evokes compensatory eye movements in the direction opposite to body acceleration. A head-fixed visual target will tend to suppress these eye movements. The VOR suppression hypothesis attributes the oculogyral illusion to the signals that prevent reflexive deviation of the eyes from the target thus resulting in apparent target displacement in the direction of acceleration. The retinal slip hypothesis attributes the illusion to inadequate fixation of the target with the eyes being involuntarily deviated in the direction opposite acceleration, the retinal slip being interpreted as target displacement in the direction of acceleration. Another possibility is that the illusion could arise from a change in the representation of the perceived head midline. To evaluate these three alternative hypotheses, we tested 8 subjects at 4 acceleration rates (2, 10, 20, 30°/s²) in each of three conditions: (a) fixate and point to a target light; (b) fixate to the target light and point to the head midline; (c) look straight ahead in the dark. The displacement magnitude of the oculogyral illusion was least at 2°/s² ≈ 2° and was ≈10° at the other acceleration rates. The presence of the target light significantly attenuated eye movements relative to the dark condition, but eye movements were still present at the 10, 20, and 30°/s² accelerations. The eye velocity profiles in the dark at different acceleration rates did not show a one-to-one inverse mapping to the magnitude of the oculogyral illusion at those rates. The perceived head midline was not significantly displaced at any of the acceleration rates. The oculogyral illusion thus has at least two contributing factors: the suppression of nystagmus at low acceleration rates and at higher acceleration rates, a partial suppression coupled with an integration of the drift of the eyes with respect to the fixation target.