Large gaze shift generation while standing: the role of the vestibular system

Abstract

The functional significance of vestibular information for the generation of gaze shifts is controversial and less well established than the vestibular contribution to gaze stability. In this study, we asked seven bilaterally avestibular patients to execute voluntary, whole body pivot turns to visual targets up to 180° while standing. In these conditions, not only are the demands imposed on gaze transfer mechanisms more challenging, but also neck proprioceptive input represents an inadequate source of head-in-space motion information. Patients' body segment was slower and jerky. In the absence of visual feedback, gaze advanced in small steps, closely resembling normal multiple-step gaze-shift patterns, but as a consequence of the slow head motion, target acquisition was delayed. In ~25% of trials, however, patients moved faster but the velocity of prematurely emerging slow-phase compensatory eye movements remained lower than head-in-space velocity due to vestibuloocular failure. During these trials, therefore, gaze advanced toward the target without interruption but, again, taking longer than when normal controls use single-step gaze transfers. That is, even when patients attempted faster gaze shifts, exposing themselves to gaze instability, they acquired distant targets significantly later than controls. Thus, while patients are upright, loss of vestibular information disrupts not only gaze stability but also gaze transfers. The slow and ataxic head and trunk movements introduce significant foveation delays. These deficits explain patients' symptoms during upright activities and show, for the first time, the clinical significance of losing the so-called "anticompensatory" (gaze shifting) function of the vestibuloocular reflex.NEW & NOTEWORTHY Previous studies in sitting avestibular patients concluded that gaze transfers are not substantially compromised. Still, clinicians know that patients are impeded (e.g., looking side to side before crossing a road). We show that during large gaze transfers while standing, vestibularly derived head velocity signals are critical for the mechanisms governing reorientation to distant targets and multisegmental coordination. Our findings go beyond the traditional role of the vestibular system in gaze stability, extending it to gaze transfers, as well.

Keywords: anticompensatory; bilateral vestibular loss; coordination; gaze; multisegmental; turns.

Fig. 1.

Experimental setup. At the beginning of the trial, the subject had aligned himself with the central light-emitting diode (LED) at 0°. After a delay of 10 s, an eccentric LED (shown at −90°) was lit while the central LED was turned off, and the subject started to turn to align his whole body with it (outbound trial; A). Note that the lower extremities started to move later than any other segment and are shown in the schematic to be still directed toward the central target at 0°. After the subject had assumed the new orientation in space, the eccentric LED was turned off while the central LED was turned on, thus cueing the subject to return to the initial, now predictable direction in space (inbound or “return” trial; B). The in-house-made octagonal wooden frame, which was installed within the LED array to prevent subjects from falling, is not depicted in the schematic.

Fig. 2.

Representative examples of rightward inbound multiple-step (A) and single-step (B) gaze shifts to the central target at 135° eccentricity by a patient (red traces) and a control subject (blue traces). Position and velocity traces are shown at top and bottom, respectively. Vertical dashed lines indicate the termination of the primary gaze shift by a leftward (downward) slow eye movement. Note the shorter head-on-trunk and trunk contribution to gaze displacement in the patient compared with those in the control subject. A: gaze continued to shift toward the target by the sum of fast phases and head-in-space displacement. Horizontal dashed arrows indicate acquisition time (considerably longer in the patient). Slow-phase eye velocity in the patient is approximately equal and opposite to head-in-space velocity. B: the velocity of the first premature and a subsequent slow-phase eye movement in the patient is considerably lower than head-in-space velocity so that gaze continues to advance toward the target. In the control, the emergence of the slow phase indicates the termination of gaze displacement and target acquisition. Oblique arrows (bottom trace) indicate corrective saccades for the gaze drift after target acquisition in the patient.

Fig. 3.

Initial (primary) gaze shift amplitude. Distribution of data from all individual return trials in patients (red circles) and controls (blue squares) when target location was either visible (45°) or predictable (90°, 135°, and 180°). Note that in patients, the saccadic gaze transfer always terminated far before covering the target eccentricity.

Fig. 4.

Prolonged acquisition time of 90°, 135°, and 180° targets in the patient group. Shaded areas represent interquartile range of normal values, whereas values for patients with bilateral vestibular loss are shown by quadrangles and error bars (median and interquartile ranges). Visual targets at 45° were acquired by patients as quickly as by normal subjects.