Visual cancellation of the torsional vestibulo-ocular reflex in humans

Using the eye-coil/magnetic field method, we measured the torsional vestibulo-ocular reflex (VOR) in ten subjects during active head rotations in roll at about 0.5 Hz. In the dark, regardless of instructions or mental effort, the gains (eye velocity/head velocity) had a mean value of around 0.61. When they viewed a visual display that was stationary, gains rose to 0.72. When viewing a visual display that moved in roll with their heads, subjects could decrease their gains to a mean of 0.46. Separate experiments showed that, as expected at this frequency, the optokinetic system made only a weak contribution. It has been proposed that the horizontal VOR is cancelled by the smooth pursuit system. Since there is no torsional pursuit system, some other mechanism must be used to augment or partially cancel the torsional VOR. Attempts to show that imagination could change this gain showed only weak effects. When asked to imagine an earth-fixed scene, gains were around 0.63; when asked to imagine a subject-fixed scene, gains decreased to only 0.60. When allowed to use a tactile contribution to aid the imagination in cancelling the VOR, the gain dropped further but only to 0.57. We conclude that mental effort in the dark has little influence on the torsional VOR but vision does by a mechanism that is not optokinetic or pursuit.