Adaptive modification of vestibularly perceived rotation

Abstract

Results from Bloomberg et al. (1991) led to the hypothesis that saccades which accompany the dark-tested vestibulo-ocular reflex (VOR) tend to move the eyes towards a vestibularly derived percept of an intended oculomotor goal: also that this is so even when that percept has been adaptively modified by suitably prolonged visual-vestibular conflict. The present experiments investigate these implications by comparing the combined VOR + saccade performance with a presumed "motor readout" of the normal and adaptively modified vestibular percept. The methods employed were similar to those of an earlier study Bloomberg et al. (1988) in which it was found that after cessation of a brief passive whole body rotation in the dark, a previously seen earth-fixed target can be accurately located by saccadic eye movements based on a vestibular memory of the preceding head rotation; the so-called "Vestibular Memory-Contingent Saccade" (VMCS) paradigm. The results showed that the vestibular perceptual response, as measured after rotation by means of the VMCS paradigm, was on average indistinguishable from the combined VOR + saccade response measured during rotation. Furthermore, this was so in both the normal and adapted states. We conclude that these findings substantiate the above hypothesis. The results incidentally reaffirm the adaptive modifiability of vestibular perception, emphasing the need for active maintenance of its proper calibration according to behavioural context.