William P. Huebner

Locomotor head-trunk coordination strategies following space flight

During locomotion, angular head movements act in a compensatory fashion to oppose the vertical trunk translation that occurs during each step in the gait cycle. This coordinated strategy between head and trunk motion serves to aid gaze stabilization and perhaps simplifies the sensory coordinate transformation between the head and trunk, allowing efficient descending motor control during locomotion. Following space flight, astronauts often experience oscillopsia during locomotion in addition to postural and gait instabilities, suggesting a possible breakdown in head-trunk coordination. The goal of the present investigation was to determine if exposure to the microgravity environment of space flight induces alteration in head-trunk coordination during locomotion. Astronaut subjects were asked to walk (6.4 km/h, 20 s trials) on a motorized treadmill while visually fixating on a centrally located earthfixed target positioned either 2 m (FAR) or 30 cm (NEAR) from the eyes. In addition, some trials were also performed during periodic visual occlusion. Head and trunk kinematics during locomotion were determined with the aid of a video-based motion analyzing system. We report data collected preflight (10 days prior to launch) and postflight (2 to 4 hours after landing). The coherence between pitch head and vertical trunk movements during gaze fixation of both FAR and NEAR targets was significantly reduced following space flight indicating decreased coordination between the head and trunk during postflight locomotion. Astronauts flying on their first mission showed greater alterations in the frequency spectra of pitch head movements as compared to their more experienced counterparts. These modifications in the efficacy of head movement control may account for the reported disruption in gaze performance during locomotion and may contribute to postflight postural and gait dysfunction.

Behavior of human horizontal vestibulo-ocular reflex in response to high-acceleration stimuli

We studied the horizontal vestibulo-ocular reflex (VOR) during transient, high-acceleration (1900-7100 degrees/s2) head rotations in 4 human subjects. Such stimuli perturbed the angle of gaze and caused illusory movement of a viewed target (oscillopsia). The disturbance of gaze could be attributed to the latency of the VOR (which ranged from 6-15 ms) and inadequate compensatory eye rotations (median VOR gain ranged from 0.61-0.83).

An investigation of horizontal combined eye-head tracking in patients with abnormal vestibular and smooth pursuit eye movements.

We investigated the interaction of smooth ocular pursuit (SP) and the vestibulo-ocular reflex (VOR) during horizontal, combined eye-head tracking (CEHT) in patients with abnormalities of either the VOR or SP movements. Our strategy was to apply transient stimuli that capitalized on the different latencies to onset of SP and the VOR. During CEHT of a target moving at 15 deg/sec, normal subjects and patients with VOR deficits all tracked the target with a gain close to 1.0. When the heads of normal subjects were suddenly and unexpectedly braked to a halt during CEHT, the eye promptly began to move in the orbit to track the target, but eye-in-orbit velocity transiently fell to about 60-70% of target velocity. In patients with deficient labyrinthine function, following the onset of the head brake, eye movements to track the target were absent, and SP movements were not generated until about 100 msec later. In patients with deficient SP, CEHT was superior to SP tracking with the head stationary; after the onset of the head brake, tracking eye movements were initiated promptly, but eye velocity was less than 50% of target velocity and increased only slightly thereafter. These results indicate that at least two mechanisms operate to overcome the VOR and allow gaze to track the target during CEHT: (1) the SP system provides a signal to cancel a normally-operating VOR (this cancellation signal is not needed by labyrinthine-deficient patients who have no VOR to cancel), and (2) a reduction of the gain of the VOR is achieved, an ability that is preserved even in patients with cerebral lesions that impair SP.

Dynamic modulation of vestibuloocular reflex gain during passive head rotation

To measure the performance of the visually enhanced vestibuloocular reflex (VEVOR), we exposed four normal subjects to the onset and subsequent offset of t 15 deg/second velocity steps of passive horizontal head rotation while they viewed a stationary light spot. Gaze and head movement data were collected using the search coil technique, and, after saccade removal and digital filtering, the measured position waveforms were differentiated to obtain gaze and head velocity. We used a modeling approach to analyze the data. Based on currently accepted schemes,*X2 a simple model of the VOR was created that incorporated elements characterizing semicircular canal dynamics and VOR latency, as well as an element providing a constant gain value. We coupled this with a model describing how visual inputs may augment VOR signals to maintain target fixation during rotation (based on Robinson et al.).3 Optimal parameter estimation techniques were employed to determine values for model parameters that caused model simulations to accurately reflect measured data.4 If the VOR acts perfectly with a gain of near 1.0, one would expect that gaze would not be significantly perturbed, despite perturbations of the head, because the generated eye movements would be almost completely compensatory. However, as depicted in FIGURE 1, the VE-VOR data showed substantial gaze perturbations when the head begins to move from rest (at 0.0 seconds), but much lower level perturbations when the head was subsequently stopped (at 2.0 seconds). This asymmetric degree of gaze perturbation was typical for all of our subjects and could not be predicted using the aforementioned model. The improved performance when the head was stopped could not have arisen due to augmentation of the VOR with signals derived from visual inputs; the latency of visual processing (requiring at least 50 mseconds for visual pro~essing)~ would delay such visual contributions until long after the observed compensatory eye motion had already occurred. Also, if vision did play a role when the head stopped, one might expect it to play a similar role at the onset of head motion, which it clearly does not. We reasoned that if the internal VOR gain were initially at some value less than 1.0 when a head rotation is initiated, the magnitude of the induced eye rotations will