| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
The Journal of Neurophysiology Vol. 80 No. 4 October 1998, pp. 2222-2228
Copyright ©1998 by the American Physiological Society
RAPID COMMUNICATION
Department of Biomedical Engineering, McGill University, Montreal, Quebec H3A 2B4, Canada
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Results
Discussion
References
|
|---|
Green, Andrea M. and Henrietta L. Galiana. Hypothesis for shared central processing of canal and otolith signals. J. Neurophysiol. 80: 2222-2228, 1998. A common goal of the translational vestibuloocular reflex (TVOR) and the rotational vestibuloocular reflex (RVOR) is to stabilize visual targets on the retinae during head movement. However, these reflexes differ significantly in their dynamic characteristics at both sensory and motor levels, implying a requirement for different central processing of canal and otolith signals. Semicircular canal afferents carry a signal proportional to angular head velocity, whereas primary otolith afferents modulate approximately in phase with linear head acceleration. Behaviorally, the RVOR exhibits a robust response down to ~0.01 Hz, yet the TVOR is only significant above ~0.5 Hz. Several hypotheses were proposed to address central processing in the TVOR pathways. All rely on a central filtering process that precedes a "neural integrator" shared with the RVOR. We propose an alternative hypothesis for the convergence of canal and otolith signals that does not impose the requirement for additional low-pass filters for the TVOR. The approach is demonstrated using an anatomically based, simple model structure that reproduces the general dynamic characteristics of the RVOR and TVOR at both ocular and central levels. Differential dynamic processing of otolith and canal signals is achieved by virtue of the location at which sensory information enters a shared but distributed neural integrator. As a result, only the RVOR is provided with compensation for the eye plant. Hence canal and otolith signals share a common central integrator, as in previous hypotheses. However, we propose that the required additional filtering of otolith signals is provided by the eye plant.
Numerous types of ocular responses may be elicited by linear acceleration stimuli, but it is perhaps the translational vestibuloocular reflex (TVOR) whose role is best understood. Like the rotational vestibuloocular reflex (RVOR) driven by the semicircular canals, the otolith-driven TVOR functions to stabilize visual targets during head movement. Nevertheless, the TVOR differs markedly from the RVOR in certain key respects. First, at the sensory level, the canal acts as a low-pass filter to angular head acceleration such that signals on primary canal afferents encode head velocity over the frequency range relevant for locomotion (<10 Hz). By contrast, primary otolith afferents carry a signal that remains approximately in phase with, or even leads, linear head acceleration in the same frequency range (Fernandez and Goldberg 1976 Model description
The proposed hypothesis will be investigated with the simple unilateral model structure shown in Fig. 1. Although the same dynamic processing strategy is applicable for compensatory reflexive responses in both the horizontal and vertical planes, the strategy will be described here in the context of rotations and interaural translations in the horizontal plane. For illustration purposes, only conjugate eye deviations (E) about an average ocular set point, which may correspond to a converged state, will be considered. Sensory drive for the RVOR and TVOR is provided by angular velocity,
Model analysis
To gain insight into how a single distributed integrator circuit can reproduce appropriate RVOR and TVOR behaviors, the ocular and central transfer functions during vestibular stimulation in the dark will be briefly outlined. Responses in the light are similar but reflect the addition of visual loops that were implemented to simulate a pursuit system with a bandwidth of 1.3 Hz. Visual feedback serves to enhance compensatory reflex responses at low frequencies (Barnes 1993 Frequency response
The dynamic behaviors of the RVOR and TVOR in the model were calculated analytically with the parameters in the Fig. 1 legend. Figure 2 provides gain and phase plots for the reflexes during VOR in the dark and in the light while viewing either an earth- or head-fixed target. The ideal TVOR gain varies as a function of fixation distance and therefore is shown normalized with respect to vergence, expressed here in meter-angles (MA, the reciprocal of fixation distance). The ideal TVOR gain in these units has an approximately constant value of 0.57°/cm/MA (Telford et al. 1997
Simulated performance
The model was implemented in SIMULINK (Mathworks, MA) at a sampling rate of 100 Hz. Model simulations during either pure angular rotation while viewing a far target or interaural linear translation while viewing a target 20 cm away are illustrated in Fig. 3. Results at 0.2 and 4 Hz are provided to contrast performance with frequency. As expected from Fig. 2, the RVOR (Fig. 3A) exhibits a gain of 0.87 in the dark at both frequencies, which increases to near unity gain at 0.2 Hz in the presence of visual feedback. The PVN cell modulates in phase with ipsilaterally directed head velocity at 4 Hz but lags considerably at 0.2 Hz, reflecting the cell's lumped premotor nature. During pure head rotation, PH cells at the output of F(s) always code for contralaterally directed eye position in the model.
We demonstrated that the required differential dynamic processing of canal and otolith-related sensory signals may be achieved in a single shared central processor simply by virtue of the location at which each sensory signal enters the circuit. In agreement with previous hypotheses, a common neural integrator for otolith and canal signals is envisioned. However, our proposal differs in that a second central low-pass filter is not required for the otolith signal. Canal signals enter a central feedback loop so as to achieve both a single integration and compensation for the eye plant. By contrast, otolith signals enter the loop such that a single integration but no eye plant compensation is performed. Hence, the eye plant provides the second filtering of otolith signals at frequencies above ~0.5 Hz. This scheme is consistent with behavioral observations of a robust, compensatory TVOR only at frequencies above the bandwidth of the eye plant (Paige and Tomko 1991a
INTRODUCTION
Abstract
Introduction
Results
Discussion
References
). Second, although the high-pass properties of the RVOR exhibit a robust response down to typically 0.01 Hz, those of the TVOR are only significant at frequencies above ~0.5 Hz (Paige and Tomko 1991a; Paige et al. 1996; Tokita et al. 1981). Third, the appropriate ocular compensation for a pure head translation depends inversely on target distance so that ideally no response is required in an unconverged state (Busettini et al. 1994; Paige and Tomko 1991b; Schwarz and Miles 1991; Telford et al. 1997). Hence TVOR performance must be interpreted in conjunction with binocular viewing context. In comparison, although the required RVOR gain also depends on target location (Crane et al. 1997; Hine and Thorn 1987; Snyder and King 1992; Viirre et al. 1986), its performance is easily gauged relative to an ideal gain of 1 for far target viewing.
; Telford et al. 1997). Yet, little is known about how and where sensory otolith signals are processed centrally. Relatively few studies considered central responses during pure linear translation, and in general either observed responses were not correlated with ocular behavior (Angelaki et al. 1993; Bush et al. 1993; Lannou et al. 1980), or the cells described could not be specifically related to well-known vestibuloocular cell types in the vestibular nuclei (VN) (Perachio 1981; Xerri et al. 1987). Recently, neuronal responses to translation during eccentric vertical-axis rotation were isolated from angular response components by comparing on-axis and off-axis responses. The well-characterized position-vestibular-pause (PVP) and eye-head-velocity (EHV) VN cell types (Cullen et al. 1993; Cullen and McCrea 1993; Lisberger et al. 1994; McCrea et al. 1996; Scudder and Fuchs 1992) were shown to carry convergent canal, otolith, and visually derived signals (Chen-Huang and McCrea 1997; McConville et al. 1996; McCrea et al. 1996). Furthermore, the otolith-related components of their response profiles modulated in phase with linear velocity rather than acceleration, suggesting that some degree of signal processing had already taken place.
; Raphan et al. 1996; Telford et al. 1997). However, because the TVOR has high-pass characteristics with a lower corner frequency of ~0.5 Hz, two simple integrations (i.e., with large time constant filters) are clearly not sufficient to describe behavioral observations. Either one filtering process can better be described as a low-pass filter with a bandwidth in the vicinity of 0.5 Hz (leaky integrator), or one near-ideal integration must be cascaded with a high-pass filter (Paige and Tomko 1991b). Additional attenuation of low-frequency signals may be provided by combining low- and high-pass filtering stages (Telford et al. 1997). Alternatively, it was proposed that a head jerk signal (derivative of head acceleration) could be used in place of a low-pass filter in the first processing stage (Angelaki et al. 1993; Angelaki and Hess 1996; Hain 1986) because a differentiation would contribute the same phase shift as an integration with sign reversal, but the gain would have high-pass characteristics. This strategy, in isolation, gives the correct response only for pure sinusoidal inputs, although a system with jerk-like response properties cascaded with an appropriate low-pass filter could be used to introduce additional high-pass filtering. Regardless of the approach, all current hypotheses for otolith processing agree that a preliminary central filtering stage precedes an integration performed by the so-called neural integrator (Robinson 1981), which is shared with the rotational system.
PROPOSED STRATEGY FOR SHARED
CANAL-OTOLITH PROCESSING
ang, sensed by the horizontal semicircular canals, C(s), and interaural translational acceleration, 
lin, from the utricles, O(s), respectively. The node labeled PVN is a lumped representation of premotor medial vestibular neurons (PVN), including PVPs, EHVs and burst-tonic (BT) neurons (Scudder and Fuchs 1992). As previously suggested, the prepositus hypoglossi (PH) is presumed to contain a neural filter, F(s), which is a scaled internal model of the eye plant P(s) (Galiana and Outerbridge 1984). PH neurons at the output of F(s) modulate in phase with eye position under most circumstances and may be considered internal estimates or efference copies (E*) of eye position (Delgado-Garcia et al. 1989; McFarland and Fuchs 1992). Reciprocal connections between PVN and PH neurons (Belknap and McCrea 1988; McCrea and Baker 1985) form a positive feedback loop that acts as a distributed neural integrator and provides gaze holding in the absence of visual feedback. Visuomotor (VM) areas (e.g., cerebellum) are represented by a lumped controller, which provides brain stem areas with a combination of retinal slip and position errors during pursuit (Barnes 1993; Lisberger et al. 1987). Although a simplified representation of visual pathways, this formulation is sufficient to explore the effects of visual loops around a brain stem circuit.
View larger version (18K):
[in a new window]
FIG. 1.
Schematic of model used to explore shared central processing of canal and otolith signals. VN (shaded), PH, and VM: vestibular nuclei, prepositus hypoglossi, and visuomotor areas, respectively. The node labeled PVN refers to a lumped premotor VN cell type. Vestibular stimulation is provided by angular head velocity, ang, sensed by semicircular canals, C(s) = Tcs/(Tcs + 1), and linear head acceleration, lin, sensed by otolith organs, O(s) = 1. T describes conjugate target position relative to the head. Neural filter, F(s) = Kf/(Tfs + 1), proposed to exist in the PH, represents an internal model of the eye plant, P(s) = Kp/(Tps + 1), when Tf = Tp. PH output neurons provide an internal estimate, E*, of eye position, E. In this collapsed version of a bilateral network, PVN cells are located on the right side of the brain stem and carry ipsilaterally directed (rightward) head movement signals, and PH cells, located on the left side of the brain stem, code for ipsilaterally directed (leftward) eye movements. Dashed pathways are activated by the presence of either a visual or cognitive goal. Weight, q, of the otolith afferent projection is shown modulated by vergence to simulate changes in TVOR gain with fixation distance. s: Laplace operator. Model parameters used in all simulations are a = 0.244; b = 1.68; p = 1; q = 2.50; Kv = 2.51; Kf = 2.40; Kp = 1; r1 = 0.27; r2 = 1; Tf = 0.28; Tp = 0.28; Tc = 5.
; McCrea et al. 1987), whereas otolith signals are proposed to reach the lumped PVN cell type indirectly via the PH. Utricular afferents project monosynaptically to the VN and the abducens nucleus but not directly to the PH (Carleton and Carpenter 1984; Uchino et al. 1996). Hence it is proposed that some proportion of otolith signals are conveyed to PH via nonpremotor, linear acceleration-sensitive cells located in the lateral or descending VN (e.g., Xerri et al. 1987); both areas are known to project to PH (Belknap and McCrea 1988; McCrea and Baker 1985). Otolith inputs to PH are weighted by vergence to account for the required modulation in TVOR gain with target distance. Parametric modulation is performed at this site for illustrative purposes only; evaluating sites of gain modulation would require a more complex model. The goal here is to demonstrate an approach for appropriate convergence of canal and otolith signals using a shared circuit.
). In the expressions below, the letter s will be used to refer to the Laplace operator. All other parameters are scalars. For ease of exposition, only deviations in activity from resting rates will be considered at the central level.
) and PH filter. Canal dynamics relative to head velocity are approximated by a first-order, high-pass filter (Fernandez and Goldberg 1971). Currently, little is known about the relative importance of different otolith afferent types in driving the TVOR. For simplicity, only regular otolith units with response gains and phases that are relatively flat up to at least 2 Hz in primate (Fernandez and Goldberg 1976) are considered; the otoliths are therefore treated as simple linear accelerometers.
(1)
(2)
when Tf = Tp and G = 1/(1
abKf). The time constant TI = Tf/(1 abKf) may be much larger than the eye plant time constant, Tp, (e.g., TI 
17 s for parameter set in Fig. 1 legend) and provides a central integration of sensory signals. Response components related to canal stimulation were presented previously (Galiana 1991) and will not be described in detail here. Notice, however, that the PVN cell response to rotation has a zero (in numerator) given by the pole (in denominator) of the neural filter F(s). Hence, if F(s) is an internal model of the eye plant (Tf = Tp), then the zero of the PVN response cancels the eye plant pole at the ocular level, as required. By contrast, the PVN cell response to otolith stimulation exhibits no zero. Because its response reflects the long integrator time constant, TI, the otolith signal appears to have been integrated before reaching the PVN cell, and therefore modulation is in phase with translational velocity instead of acceleration. At the ocular level, no compensation is provided for the eye plant, and hence the eye plant itself performs the second integration or filtering of the otolith signal. This proposal is consistent with behavioral observations because the dominant eye plant time constant Tp ( 0.2-0.3 s) corresponds to a pole somewhere between 0.5 and 0.8 Hz. PH cells at the output of F(s) provide an internal estimate of eye position (efference copy), E*. In this model, the accuracy of the efference copy will degrade during pure translation at frequencies above the bandwidth of the eye plant (see DISCUSSION). Notice that both central and ocular responses to otolith stimulation are completely independent of the poles of F(s) (associated here with a time constant of Tf). The only dynamic term in the otolith responses that does depend on Tf, Tl = Tf/(1 abKf) can be fixed to any desired value by suitable choice of abKf.
RESULTS
Abstract
Introduction
Results
Discussion
References
).
View larger version (53K):
[in a new window]
FIG. 2.
Predicted frequency response plots for head rotation (A) and translation (B) in the dark and in the light while viewing either an earth- or head-fixed target. In B, the translational vestibuloocular reflex gain is normalized with respect to vergence. Phases relative to head velocity were shifted by 180° to represent phase relative to the ideal compensatory response.
0.87) in phase with angular head velocity over most of the frequency range (Paige 1983). Because velocity storage was not incorporated in this simple model, RVOR responses in the dark below 0.1 Hz are of lower gain and larger phase lead relative to head velocity than observed experimentally. In the presence of visual feedback, close to unity gain and zero phase are observed across the entire bandwidth, as expected. VOR cancellation in the model relies only on visual feedback, and therefore both the ability to track a moving target and to suppress the reflex diminish with increasing frequency, reflecting the model pursuit bandwidth of 1.3 Hz (Barnes 1993).
) and primates (Telford et al. 1997). During earth- (Fig. 2B, solid line) and head-fixed (dotted line) target viewing, responses are nearly ideal at low frequencies. However, performance in both cases declines with frequency above the pursuit bandwidth so that gains in the dark and in the light are similar at 4 Hz, as observed (Paige et al. 1996).
View larger version (59K):
[in a new window]
FIG. 3.
Model simulations during (A) angular rotation while viewing a far target or (B) interaural linear translation while viewing a target 20 cm away ( 17° vergence for an interocular distance of 6 cm). Head rotations or translations are illustrated at 0.2 and 4 Hz in each case. Translations and rotations to the right are positive. Dotted lines represent ideal compensatory responses. Shaded portions of the simulations indicate dark conditions.
70°) relative to the ideal compensatory response but rises to a gain of 0.31°/cm/MA at 4 Hz with modulation in phase with translational head velocity (Paige et al. 1996; Telford et al. 1997). As expected, performance is close to ideal at 0.2 Hz in the presence of visual feedback. PVN cell activity modulates in phase with translational velocity rather than acceleration at both frequencies in the dark and at 4 Hz in the light, in agreement with experimental observations (Chen-Huang and McCrea 1997; McConville et al. 1996; McCrea et al. 1996). Note that this result stems from the overall dynamics of the circuit (see Eq. 2) and is unrelated to the additional lag on the lumped PVN cell during rotation; using our approach the same observation would be made even in a more detailed model that represents individual premotor cell types. The predicted response of the PH neuron at the output of F(s) is of particular interest. At 0.2 Hz, the cell modulates in phase with translational head velocity in the dark but more closely in phase with head position in the light. Ocular responses at this frequency are compensatory in the presence of visual feedback but lead ideal eye velocity by nearly 90° in the dark; hence, when considered relative to the motor response, the PH neuron provides a good estimate of eye position at low frequencies in keeping with its postulated role in providing an efference copy signal. At 4 Hz, the PH cell again modulates in phase with translational head velocity. However, because ocular responses at this frequency are indeed compensatory, the PH cell now modulates in phase with eye velocity in both the light and dark conditions. Hence the simulations predict a marked change in this cell's response relative to motor output at higher frequencies during pure translation.
DISCUSSION
Abstract
Introduction
Results
Discussion
References
; Paige et al. 1996; Telford et al. 1997; Tokita et al. 1981). Furthermore, we demonstrated that the general characteristics of both the RVOR and TVOR may be reproduced even when only the dominant dynamic properties of the sensors and eye plant are taken into account. More detailed models of sensor and plant dynamics may be incorporated to improve the overall fit of our responses when the characteristics of the TVOR are available over a larger frequency range and the types of otolith afferents important in driving the reflex are identified.
). Using this approach, eye plant compensation during the RVOR relies on the appropriate weighting of projections to the eye plant from these two pathways. The appropriate TVOR response could be achieved if otolith signals were passed selectively only through the integrator pathway before being filtered a second time by the uncompensated eye plant. Central results could be reproduced by assuming that the PVN cell sums signals from both the direct pathway (i.e., canal primary afferent signals) and the output of the neural integrator. Inputs to the neural integrator, in this case, would be vestibular-only cells carrying angular head velocity and/or linear head acceleration signals but no eye position signal. Regardless of the implementation of the central filtering process (i.e., lumped feed-forward vs. a feedback approach), the basic strategy with respect to combined canal/otolith processing is the same; a shared neural integrator for both canal and otolith signals followed by a second integration of otolith signals by the eye plant.
; Angelaki et al. 1993; Raphan et al. 1996). However, these proposals may be difficult to confirm experimentally. An advantage of our hypothesis is that it makes testable predictions at behavioral, anatomic, and physiological levels, which may help to distinguish between a multistage central filtering process for otolith signals versus a single integration followed by filtering by the eye plant. First, because a basic premise of our strategy is that the eye plant itself is used to filter otolith signals, TVOR behavior should reflect the dynamics of the eye plant at high frequencies. Hence, the frequency associated with a second filtering of the otolith signal should correspond well with the dominant pole of the eye plant, and evidence of higher-order eye plant dynamics (not incorporated in the model here) would be expected in the TVOR responses. Recent investigations of high-frequency TVOR responses do indeed appear to reflect these higher-order plant dynamics (Angelaki, personal communication). Second, from an anatomic perspective, although PVN neurons (e.g., PVP and EHV) receive monosynaptic inputs from the semicircular canals, we predict that the majority of otolith afferents make only indirect (di- or trisynaptic) projections onto these cells.
). An extension of this model incorporating the strategy proposed here to examine viewing distance effects on the TVOR and its interaction with the RVOR should contribute to a better understanding of the central processing underlying the dynamic and spatial characteristics of compensatory vestibuloocular reflexes.
| |
ACKNOWLEDGEMENTS |
|---|
The authors express gratitude for the helpful comments of Drs. R. Baker and D. E. Angelaki in the preparation of the manuscript.
This research was supported by the Medical Research Council of Canada.
| |
FOOTNOTES |
|---|
Address for reprint requests: H. L. Galiana, Dept. of Biomedical Engineering, McGill University, 3775 University St., Montreal, Quebec H3A 2B4, Canada.
Received 12 May 1998; accepted in final form 1 July 1998.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  A. M. Green, H. Meng, and D. E. Angelaki A Reevaluation of the Inverse Dynamic Model for Eye Movements J. Neurosci., February 7, 2007; 27(6): 1346 - 1355. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.-r. Tian, E. Mokuno, and J. L. Demer Vestibulo-Ocular Reflex to Transient Surge Translation: Complex Geometric Response Ablated by Normal Aging J Neurophysiol, April 1, 2006; 95(4): 2042 - 2054. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. E. Andreescu, M. M. De Ruiter, C. I. De Zeeuw, and M. T. G. De Jeu Otolith Deprivation Induces Optokinetic Compensation J Neurophysiol, November 1, 2005; 94(5): 3487 - 3496. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Ramat, D. Straumann, and D. S. Zee Interaural Translational VOR: Suppression, Enhancement, and Cognitive Control J Neurophysiol, October 1, 2005; 94(4): 2391 - 2402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Meng, A. M. Green, J. D. Dickman, and D. E. Angelaki Pursuit--Vestibular Interactions in Brain Stem Neurons During Rotation and Translation J Neurophysiol, June 1, 2005; 93(6): 3418 - 3433. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.W.P. Chan and H. L. Galiana Integrator Function in the Oculomotor System Is Dependent on Sensory Context J Neurophysiol, June 1, 2005; 93(6): 3709 - 3717. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Green and D. E. Angelaki An Integrative Neural Network for Detecting Inertial Motion and Head Orientation J Neurophysiol, August 1, 2004; 92(2): 905 - 925. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Angelaki Eyes on Target: What Neurons Must do for the Vestibuloocular Reflex During Linear Motion J Neurophysiol, July 1, 2004; 92(1): 20 - 35. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Green and D. E. Angelaki Resolution of Sensory Ambiguities for Gaze Stabilization Requires a Second Neural Integrator J. Neurosci., October 15, 2003; 23(28): 9265 - 9275. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Angelaki Three-Dimensional Ocular Kinematics During Eccentric Rotations: Evidence for Functional Rather Than Mechanical Constraints J Neurophysiol, May 1, 2003; 89(5): 2685 - 2696. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Musallam and R. D. Tomlinson Asymmetric Integration Recorded From Vestibular-Only Cells in Response to Position Transients J Neurophysiol, October 1, 2002; 88(4): 2104 - 2113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Merfeld and L. H. Zupan Neural Processing of Gravitoinertial Cues in Humans. III. Modeling Tilt and Translation Responses J Neurophysiol, February 1, 2002; 87(2): 819 - 833. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. P. Medendorp, J.A.M. Van Gisbergen, and C.C.A.M. Gielen Human Gaze Stabilization During Active Head Translations J Neurophysiol, January 1, 2002; 87(1): 295 - 304. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Angelaki, A. M. Green, and J. D. Dickman Differential Sensorimotor Processing of Vestibulo-Ocular Signals during Rotation and Translation J. Neurosci., June 1, 2001; 21(11): 3968 - 3985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Angelaki and J. D. Dickman Spatiotemporal Processing of Linear Acceleration: Primary Afferent and Central Vestibular Neuron Responses J Neurophysiol, October 1, 2000; 84(4): 2113 - 2132. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Angelaki, M. Q. McHenry, J. D. Dickman, and A. A. Perachio Primate Translational Vestibuloocular Reflexes. III. Effects of Bilateral Labyrinthine Electrical Stimulation J Neurophysiol, March 1, 2000; 83(3): 1662 - 1676. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||
