Cross-axis adaptation of torsional components in the yaw-axis vestibulo-ocular reflex

P Trillenberg, M Shelhamer, D Roberts… - Experimental brain …, 2003 - Springer
P Trillenberg, M Shelhamer, D Roberts, D Zee
Experimental brain research, 2003Springer
The three pairs of semicircular canals within the labyrinth are not perfectly aligned with the
pulling directions of the six extraocular muscles. Therefore, for a given head movement, the
vestibulo-ocular reflex (VOR) depends upon central neural mechanisms that couple the
canals to the muscles with the appropriate functional gains in order to generate a response
that rotates the eye the correct amount and around the correct axis. A consequence of these
neural connections is a cross-axis adaptive capability, which can be stimulated …
Abstract
The three pairs of semicircular canals within the labyrinth are not perfectly aligned with the pulling directions of the six extraocular muscles. Therefore, for a given head movement, the vestibulo-ocular reflex (VOR) depends upon central neural mechanisms that couple the canals to the muscles with the appropriate functional gains in order to generate a response that rotates the eye the correct amount and around the correct axis. A consequence of these neural connections is a cross-axis adaptive capability, which can be stimulated experimentally when head rotation is around one axis and visual motion about another. From this visual-vestibular conflict the brain infers that the slow-phase eye movement is rotating around the wrong axis. We explored the capability of human cross-axis adaptation, using a short-term training paradigm, to determine if torsional eye movements could be elicited by yaw (horizontal) head rotation (where torsion is normally inappropriate). We applied yaw sinusoidal head rotation (±10°, 0.33 Hz) and measured eye movement responses in the dark, and before and after adaptation. The adaptation paradigm lasted 45–60 min, and consisted of the identical head motion, coupled with a moving visual scene that required one of several types of eye movements: (1) torsion alone (-Roll); (2) horizontal/torsional, head right/CW torsion (Yaw-Roll); (3) horizontal/torsional, head right/CCW torsion (Yaw+Roll); (4) horizontal, vertical, torsional combined (Yaw+Pitch-Roll); and (5) horizontal and vertical together (Yaw+Pitch). The largest and most significant changes in torsional amplitude occurred in the Yaw-Roll and Yaw+Roll conditions. We conclude that short-term, cross-axis adaptation of torsion is possible but constrained by the complexity of the adaptation task: smaller torsional components are produced if more than one cross-coupling component is required. In contrast, vertical cross-axis components can be easily trained to occur with yaw head movements.
Springer