The "gold standard" for recording the three-dimensional rotation of the eye involves placing two coils of wire, embedded in a soft plastic ring, on the sclera of the eye, then placing the subject inside a set of orthogonal oscillating magnetic fields, and using the currents induced in the eye coils to deduce the position of the coil, and hence of the eye, in space. Eye movements are actually eye rotations, which can be described mathematically by a special class of matrices, rotation matrices, or, alternatively, by a rotation vector related to the axis of the rotation. This article deals with the mathematical tools needed to implement the signal processing from such a multifield, dual-coil system and compute the precise rotational movement of the eye. One reason for making such careful measurements is to study an interesting constraint on eye movements, called Listing's law, which expresses ocular torsion, or rotation of the eye about its line of sight, in terms of the direction of gaze. Techniques for experimentally quantitating these constraints are also presented. Following a treatment of the "ideal" case, with coils and eye in perfect alignment, the additional techniques for dealing with various departures from ideality that are almost always encountered experimentally are examined. A final section deals with developing a validation protocol for eye movement analysis techniques using mechanical and computer simulations of eye movements.

The rostral fastigial nucleus (RFN) of the cerebellum is thought to play an important role in postural control, and recent studies in conscious nonhuman primates suggest that this region also participates in the sensory processing required to compute body motion in space. The goal of the present study was to examine the dynamic and spatial responses to sinusoidal rotations in vertical planes of RFN neurons in conscious cats, and determine if they are similar to responses reported for monkeys. Approximately half of the RFN neurons examined were classified as graviceptive, since their firing was synchronized with stimulus position and the gain of their responses was relatively unaffected by the frequency of the tilts. The large majority (80%) of graviceptive RFN neurons were activated by pitch rotations. Most of the remaining RFN units exhibited responses to vertical oscillations that encoded stimulus velocity, and approximately 50% of these velocity units had a response vector orientation aligned near the plane of a single vertical semicircular canal. Unlike in primates, few feline RFN neurons had responses to vertical rotations that suggested integration of graviceptive (otolith) and velocity (vertical semicircular canal) signals. These data indicate that the physiological role of the RFN may differ between primates and lower mammals. The RFN in rats and cats in known to be involved in adjusting blood pressure and breathing during postural alterations in the transverse (pitch) plane. The relatively simple responses of many RFN neurons in cats are appropriate for triggering such compensatory autonomic responses.

Although many previous experiments have considered the responses of vestibular nucleus neurons to rotations and translations of the head, little data are available regarding cells in the caudalmost portions of the vestibular nuclei (CVN), which mediate vestibulo-autonomic responses among other functions. This study examined the responses of CVN neurons of conscious cats to rotations in vertical planes, both before and after a bilateral vestibular neurectomy. None of the units included in the data sample had eye movement-related activity. In labyrinth-intact animals, some CVN neurons (22%) exhibited graviceptive responses consistent with inputs from otolith organs, but most (55%) had dynamic responses with phases synchronized with stimulus velocity. Furthermore, the large majority of CVN neurons had response vector orientations that were aligned either near the roll or vertical canal planes, and only 18% of cells were preferentially activated by pitch rotations. Sustained head-up rotations of the body provide challenges to the cardiovascular system and breathing, and thus the response dynamics of the large majority of CVN neurons were dissimilar to those of posturally-related autonomic reflexes. These data suggest that vestibular influences on autonomic control mediated by the CVN are more complex than previously envisioned, and likely involve considerable processing and integration of signals by brainstem regions involved in cardiovascular and respiratory regulation. Following a bilateral vestibular neurectomy, CVN neurons regained spontaneous activity within 24 h, and a very few neurons (<10%) responded to vertical tilts <15° in amplitude. These findings indicate that nonlabyrinthine inputs are likely important in sustaining the activity of CVN neurons; thus, these inputs may play a role in functional recovery following peripheral vestibular lesions.

Extracellular spikes were recorded from secondary vestibular neurons in the cat medial vestibular nucleus (MVN) and were identified as type I or II neurons by horizontal rotation. Type I neurons were further classified as excitatory or inhibitory premotor neurons on the basis of their axonal termination in the contralateral or ipsilateral abducens nucleus, demonstrated by spike-triggered averaging of abducens nerve discharges, or by antidromic activation using systematic microstimulation within the abducens nucleus.

Three-dimensional eye positions, when expressed as rotation vectors, are constrained to lie in a head-fixed Listing's plane. The offset and orientation of Listing's plane changes when the head is tilted. To assess the influence of age on this phenomenon, young (less than 30 years old) and older (>65 years old) human subjects were seated upright, pitched nose up and nose down, and rolled right ear down and left ear down. Listing's plane was computed from eye movements recorded using a dual scleral search coil while subjects scanned a complex visual scene. During pitch, Listing's plane counterpitched with respect to the head, while during roll, it translated in a manner consistent with "ocular counterrolling". There was no significant difference in this reorientation of Listing's plane between the young and older subjects. The only obvious difference between the two age groups was that the "thickness" of Listing's plane was greater in the older subjects. This suggests that aging has a small, but definite, influence on Listing's law.

The response of the vestibular system after acoustic neuroma surgery was investigated in nine patients. The otolith system was studied by means of ocular counterrolling, assessed by video oculography. Horizontal vestibulo-ocular reflex (VOR) function was tested by the sinusoidal harmonic acceleration test using electronystagmography. The results were compared with those obtained from a normal control population. The response to slow rotation tests was symmetric, but the gain was significantly reduced when compared to the normal population. Phase lag was significantly increased. No difference in ocular torsion was observed with lateroflexion of the head to the ipsilateral side in comparison with lateroflexion to the contralateral side. Moreover, the overall ocular counterrolling was well within normal limits. We conclude that the semicircular canal response differs from the otolith response. The component of the torsional VOR mediated by otolith stimulation appears to be more robust than the horizontal VOR mediated mainly by the horizontal semicircular canal system. Ocular counterrolling induced by lateroflexion does not reveal abnormalities in patients with surgically produced unilateral peripheral loss.

The nystagmus following yaw earth-vertical axis rotation often reverses direction, a phenomenon known as the "secondary phase". The purpose of this study was to examine the existence and the spatial and temporal properties of the secondary phase of post-rotatory nystagmus following off-vertical axis rotation (OVAR). Eleven normal human subjects were rotated at 120 or 180 degrees/s about an off-vertical axis and stopped in the left ear down or right ear down lateral position. Horizontal and vertical eye positions were recorded with a scleral search coil, and horizontal and vertical slow component eye velocities were computed. Our results indicate that (a) there is a robust secondary phase nystagmus following OVAR, and (b) the direction of the secondary phase nystagmus tends to align with earth-horizontal. These results can be explained by a minor modification of an existing VOR model that has been shown to produce secondary phase responses.

The purpose of this review is to assess the role of short-latency pathways in the vestibulocollic reflex (VCR). First the current knowledge about the disynaptic and trisynaptic pathways linking semicircular canal and otolith afferents with cat neck motoneurons is summarized. We then discuss whether these pathways are sufficient or necessary to produce the responses observed in neck muscles by natural vestibular stimulation and conclude that they are neither. Finally, alternate pathways are considered, most likely involving reticulospinal fibers, which are an important part of the neural substrate of the VCR.

The purpose of this review is to assess the role of short-latency pathways in the vestibulocollic reflex (VCR). First the current knowledge about the disynaptic and trisynaptic pathways linking semicircular canal and otolith afferents with cat neck motoneurons is summarized. We then discuss whether these pathways are sufficient or necessary to produce the responses observed in neck muscles by natural vestibular stimulation and conclude that they are neither. Finally, alternate pathways are considered, most likely involving reticulospinal fibers, which are an important part of the neural substrate of the VCR.

Our goal was to study potential substrates for cortical modulation of vestibular reflexes in the cat. In initial experiments, injections of wheat-germ-agglutinate-horseradish-peroxidase into Deiters’ nucleus and the rostral descending nucleus revealed bilateral colonies of retrogradely filled neurons in cortical areas 6, 2, and 3a (about 60 cells per colony). In cats anesthetized with chloralose-urethane, we stimulated areas 2 and 3a with trains of pulses while recording from ipsilateral vestibular-nucleus neurons, which were characterized by their responses to sinusoidal tilts and tested for the presence of antidromic responses to stimulation of the upper cervical cord. A majority of the neurons was affected by cortical stimulation, showing either facilitation, inhibition, or a mixture of the two. Stimulation in area 2 was more effective than stimulation in area 3a. Despite the anatomic presence of direct cortico-vestibular projections, properties of facilitation and inhibition suggest that both were evoked by polysynaptic pathways. Cortical effects were broadly distributed to vestibular neurons without regard to responses of these neurons to sinusoidal tilts. There was no significant difference between effects on lateral and medial vestibulospinal tract neurons, but, as a group, vestibulospinal neurons were much more likely to be affected by cortical stimulation than neurons not antidromically activated from the C2 segment. We conclude that, by their influence on vestibulospinal neurons, neurons in cortical areas 2 and 3a should be able to modulate, in behaving animals, vestibular reflexes acting on the neck and limbs.

Responses to linear accelerations in the earth-horizontal plane (typically provoked by tilts of the head or body) are characterized by a stimulus direction that produces the maximal excitation. Although changes in cardiovascular, sympathetic, and respiratory outflow are maximized during pitch, no collection of central vestibular neurons had been identified where pitch responses predominate. In the present study, response properties of neurons in the medial vestibular nucleus were examined in decerebrate cats placed on a turntable. Activation of otolith afferents was provided by constant velocity rotation with the turntable axis tilted 5 degrees from the vertical. Responsive neurons exhibited a sinusoidal modulation in their firing rate; the optimal excitatory stimulus direction was derived from responses to clockwise and counterclockwise rotations. Many of these neurons were also tested for input from horizontal semicircular canals using 0.5 Hz sinusoidal rotation about an earth-vertical axis. Of 22 tilt-sensitive neurons in the medial vestibular nucleus whose optimal stimulus direction was determined, 9 were best stimulated by pitch, 10 by stimuli in one of the two vertical semicircular canal planes, and 3 by roll. Of the 33 neurons in this nucleus tested for possible convergent inputs from the otolith organs and the horizontal semicircular canals, 8 responded to both the constant velocity (otolith) stimulus and to the sinusoidal rotation, 7 appeared to receive otolith, but not horizontal canal, input, while 18 had a canal, but no otolith, response. Thus, besides serving as a relay for horizontal canal signals, the medial vestibular nucleus may also be an important relay for information about orientation within the sagittal (pitch) plane.

1. In decerebrate cats with intact cerebellums, we studied the responses of neurons in the caudal areas of the vestibular nuclei to natural vestibular stimulation in vertical planes and to neck rotation. The activity of most neurons was recorded in the caudal half of the descending nucleus. 2. One goal of our experiments was to compare the dynamic and spatial properties of responses to sinusoidal vestibular stimulation with those seen in previous experiments in which the caudal cerebellar vermis, including the nodulus and uvula, was removed. This part of the cerebellum receives vestibular input and projects to the caudal areas of the vestibular nuclei, suggesting that it could influence responses to stimulation of the labyrinth. 3. As in our previous experiments, most neurons could be classified as receiving predominant input either from the otoliths or from one vertical semicircular canal. When mean gain and phase and response vector orientations were compared, there were no obvious differences between the behavior of neurons in the partially decerebellate preparation and the one with the cerebellum intact, demonstrating that in the decerebrate cat the nodulus and uvula have little or no influence on the processing of vertical vestibular input in this region of the vestibular nuclei. 4. Only 23 of 74 (31%) of neurons tested responded to neck rotation. This contrasts with the much larger fractions that respond to this stimulus in Deiters' nucleus and in the rostral descending nucleus. We also recorded from neurons near the vestibular nuclei, mainly in the external cuneate nucleus. All of them (9 of 9) responded to neck rotation. 5. Responses to neck rotation also differed in their dynamics from those found more rostrally in the vestibular nuclei. Dynamics of more rostral neurons resemble those of neck muscle spindles, as do those of external cuneate neurons. The dynamics of caudal vestibular neurons, on the other hand, have a steeper gain slope and more advanced phases than do those of neurons in the more rostral vestibular nuclei. This suggests the possibility of involvement of additional receptors in the production of these responses. 6. In the more rostral vestibular nuclei, responses to vestibular and neck rotation are most often antagonistic, so that head rotation results in little or no response. This is not the case in the caudal areas of the vestibular nuclei, where less than half the neurons tested displayed antagonistic behavior. Further experiments are required to put the neck projection to the caudal vestibular nuclei in a functional context.

The duration of post-rotatory nystagmus is known to be shortened by head tilt, a phenomenon that has been attributed to rapid discharge of the velocity storage mechanism. The relative importance of the various sensory signals associated with post-rotatory head tilt is unknown. Using both earth-vertical axis and off-vertical axis rotation, we investigated this issue in humans by combining sudden termination of constant velocity rotation with several post-rotatory maneuvers that stimulated combinations of otolith organ, somatosensory, and vertical semicircular canal afferents. Results indicated that horizontal post-rotatory nystagmus was shortened by maneuvers that move the head from upright to off-vertical and by cessation of dynamic otolith stimuli. Somatosensory, neck afferent, and transient vestibular stimulation had no consistent effect. We conclude that tilt suppression of postrotatory nystagmus and the short time constant following off-vertical axis rotation is primarily a function of otolith influence on the velocity storage system.

This study examines the response of neurons in the medullary reticular formation of the decerebrate cat to sinusoidal yaw rotations in the plane of the horizontal semicircular canals. Responsive neurons that could be antidromically activated from the spinal cord appeared to be less sensitive to the rotary stimulus than the rest of the population or responsive neurons. Most neurons had response dynamics similar to those of semicircular canal afferents.

Annals of the New York Academy of SciencesVolume 656, Issue 1 p. 500-506 Spatial Transformation in the Vertical Vestibulocollic Reflexa V. J. WILSON, V. J. WILSON The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorP. S. BOLTON, P. S. BOLTON The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorT. GOTO, T. GOTO The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorR. H. SCHOR, R. H. SCHOR The Eye and Ear Institute University of Pittsburgh Pittsburgh, Pennsylvania 15213Search for more papers by this authorY. YAMAGATA, Y. YAMAGATA Present affiliation: Department of Ophthalmology, Hyogo College of Medicine, NishinomiyaCity, Hyogo 663, Japan.Search for more papers by this authorB. J. YATES, B. J. YATES The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this author V. J. WILSON, V. J. WILSON The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorP. S. BOLTON, P. S. BOLTON The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorT. GOTO, T. GOTO The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this authorR. H. SCHOR, R. H. SCHOR The Eye and Ear Institute University of Pittsburgh Pittsburgh, Pennsylvania 15213Search for more papers by this authorY. YAMAGATA, Y. YAMAGATA Present affiliation: Department of Ophthalmology, Hyogo College of Medicine, NishinomiyaCity, Hyogo 663, Japan.Search for more papers by this authorB. J. YATES, B. J. YATES The Rockefeller University 1230 York Avenue New York, New York 10021Search for more papers by this author First published: May 1992 https://doi.org/10.1111/j.1749-6632.1992.tb25231.xCitations: 3 a Supported by National Institutes of Health grants NS02619, NS24930, and DC00693. AboutPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Citing Literature Volume656, Issue1Sensing and Controlling Motion: Vestibular and Sensorimotor FunctionMay 1992Pages 500-506 RelatedInformation

1. To investigate the neural substrate of vestibulospinal reflexes in decerebrate cats, we studied the responses of pontomedullary reticulospinal neurons to natural stimulation of the labyrinth in vertical planes. Our principal aim was to determine whether reticulospinal neurons that terminate in, or are likely to give off collaterals to, the upper cervical segments had properties similar to those of the vestibulocollic reflex (VCR). 2. Antidromic stimulation was used to determine whether the neurons projected to the neck, lower cervical, thoracic, or lumbar levels. Dynamics of the responses of spontaneously firing neurons were studied with sinusoidal stimuli delivered at 0.05-1 Hz and aligned to the plane of body rotation, that produced maximal modulation of the neuron (response vector orientation). Each neuron was assigned a vestibular input classification of otolith, vertical canal, otolith + canal, or spatial-temporal convergence (STC). 3. We found, in agreement with previous studies, that the largest fraction of pontomedullary reticulospinal neurons projected to the lumbar cord, and that only a small number ended in the neck segments. Neurons projecting to all levels of the spinal cord had similar responses to labyrinth stimulation. 4. Reticulospinal neurons that received only vertical canal inputs were rare (1 of 67 units). Most reticulospinal neurons (48%) received predominant otolith inputs, 18% received otolith + canal input, and only 9% had STC behavior. These data are in sharp contrast to the results of our previous studies of vestibulospinal neurons. A considerable portion of vestibulospinal neurons receives vertical canal input (38%), fewer receive predominantly otolith input (22%), whereas the proportion that have otolith + canal input or STC behavior is similar to our present reticulospinal data. 5. The response vector orientations of our reticulospinal neurons, particularly those with canal inputs (canal, otolith + canal, STC) were predominantly in the roll quadrants. There was no evidence of convergence of inputs from like canals across the midline (e.g., right anterior + left anterior). 6. Two characteristics of the VCR, STC behavior and bilateral input from symmetric vertical canals (in some muscles), cannot be accounted for by the reticulospinal neurons that we studied. Because these characteristics are also not seen in vestibulocollic neurons, they are likely to be the result of the appropriate convergence of vestibular signals in the spinal cord. 7. Pontomedullary reticulospinal neurons seem particularly well suited to play a role in gravity-dependent postural reflexes of neck and limbs.

The purpose of this study was to examine the effect of bilateral labyrinthectomy on quiet stance in the freely-standing cat. Since loss of the vestibular end organs produces marked deficits in motor behaviour, including ataxia and problems with balance, we hypothesized that labyrinthectomized animals would show impairment in quantitative measures of stance. Stance was quantified in terms of the ground reaction forces under each limb and the tonic electromyographic (EMG) activity of selected muscles. Animals were labyrinthectomized by drilling into the vestibule and removing the vestibular epithelium. Following lesion, animals were able to stand unsupported on the force platform within 2 days. To our surprise, the lesioned animals showed little change in stance parameters from the control, pre-lesion state. Thus, our hypothesis of changes in stance parameters was not supported. There was no change in the distribution of vertical forces under the limbs and no increase in sway, as measured by the area of excursion of the centre of pressure over time. The horizontal plane forces, which were diagonally directed prior to lesion, became more laterally directed and larger in amplitude. The change in direction persisted even after the animals had fully compensated for the lesion, but the force amplitudes returned to control values within 10–12 days. The change in horizontal force direction was similar to that observed in normal animals that were required to stand with their paws closer than preferred in the sagittal plane (unpublished observations). EMG activity changed in some muscles but not others, and usually transiently. One limb extensor showed decreases in tonic activity (gluteus medius), but other extensors showed increases (vastus medialis, soleus). It is likely that the changes in EMG levels were due to the biomechanics of the stance configuration related to the changes in direction of the horizontal plane forces. Kinematic recordings are needed to address this issue further.

1. To compare the properties of the vestibulocollic reflex (VCR) with those of vestibular neurons projecting to the neck [vestibulocollic (VC) neurons], we have studied the behavior of the latter in the decerebrate cat. Neurons were identified by their antidromic responses to stimulation in C1-C2, but not C5. Responses to stimulation of vestibular and neck receptors were produced by rotation of the body and head in vertical planes. 2. We determined the plane of whole body (vestibular) or body with head counter-rotated (neck) rotation, which produced the maximal modulation of each neuron (response vector orientation). Neuron dynamics were then studied with sinusoidal (0.02-2 Hz) stimuli aligned with this orientation. 3. On the basis of dynamics and vector orientation, the neuron was assigned a vestibular input classification of otolith, vertical canal, otolith + canal, or spatial-temporal convergence (STC). 4. The properties of this sample of VC neurons are similar to those of a larger population of vestibular neurons whose projection was not identified. For example, the distributions of cells with different types of vestibular inputs were roughly the same; in particular, few cells showed STC responses. In addition, there was no evidence of significant convergence of like canals across the midline (e.g., right anterior + left anterior). 5. Also similar to the larger unidentified population, 80% of VC neurons tested for neck input received such an input. The neck and vestibular responses tended to be antagonistic; the vector orientations were usually opposite, and the response gains and phases similar.(ABSTRACT TRUNCATED AT 250 WORDS)

In decerebrate cats we used response dynamics to classify neurons in Deiters' nucleus and the rostral descending nucleus as receiving input from vertical canals, otolith organs (utricle) or canal + otolith. Many of these neurons, particularly those with convergent vestibular input (canal + otolith, canal + canal) and preferred stimulus directions near roll, also received neck input. Neck and vestibular response vectors tended to point in opposite directions. Neck and vestibular response dynamics were often well matched, so that combined stimuli (head rotation) produced no response over a wide frequency range. In some cases differing dynamics produced an output that appeared to code head position.