PDFlib PLOP: PDF Linearization, Optimization, Protection Page inserted by evaluation version www.pdflib.com –
[email protected]
Ocular Selectivity of Units in Oculomotor Pathways" WU ZHOUb,'pd AND W. M. KING' bNeuroscienceProgram & Centerfor Visual Science University of Rochester Medical Center Rochester, New York 14642 University of Mississippi Medical Center Deparfments of Neurology and Anatomy Jackson, Mississippi 39216-4505 During linear motion in space, the translational-vestibulo-ocular reflex (TVOR) produces disjunctive eye movements whose amplitude and direction depend on the spatial location of the subject's fixation point, as well as the vestibular sensory signal.' The processing of vestibular and eye movement signals in the pathways used to produce the TVOR have not been systematically studied in primates. To begin such a study, we have returned to the basic question of how eye position and speed are encoded by neurons in vestibulo-oculomotor pathways. The salient new issue is how left and right eye position and speed are encoded during the TVOR when these parameters may be different. Recent behavioral studies suggest that some premotor pathways (e.g., the TVORI and smooth pursuitz) may generate disjunctive eye movement commands independently of the classical vergence system. Thus units in these pathways might encode monocular eye position and speed rather than binocular conjugate and vergence position and speed. METHODS
Single-unit discharge patterns were recorded in monkeys trained to track visual targets that moved in three-dimensional space. Movements of each eye were recorded with scleral search coils. Target trajectories were selected to generate conjugate and disjunctive eye movement patterns that dissociated the positions and speeds of the two eyes (FIG.l), so that any changes in a unit's discharge could be related selectively to the movements of either eye. Eye movement data and unit discharges were averaged over 50-msec intervals during fixation and smooth pursuit and were analyzed using multiple regression techniques to evaluate the relationship between firing rate and eye movement parameters. Ocular selectivity was defined as the extent to which a unit's firing rate was correlated with the movements of one or the other eye. Ocular selectivity indices were computed for eye position, OSK [equal to (K, - K,)/(K, + &),where K, and Kl are unit discharge sensitivities to right or left eye position], and eye velocity, OSR [equal to (R, - Rl)/(R, + RI),where R, and RI are unit discharge sensitivities to right or left eye velocity]. According to these aThis work was supported by National Institutes of Health grant ROlEY04045 and Office of Naval Research grant NO0014 to Dr. King. dAddress for correspondence: Wu Zhou, Department of Anatomy, University of Mississippi Medical Center, 2500 North State Street, Jackson, MS 39216-4505. E-mail: wuz@cvs. rochester.edu 724
ZHOU & KING OCULAR SELECTIVITY
725
Abducens Unitl 2-1
A.
:'[
n
Prepositus H. Unitl 0-7
D.
1OL
'*I 50
C.
F.
FIGURE 1. Responses of a binocular left burst-tonic abducens unit (A-C) and a monocular left tonic prepositus unit (D-F) during smooth pursuit of a visual target that moved in threedimensional space. Panels A and D show unit responses during conjugate pursuit. Panels B, C, E, and F show unit responses during asymmetric pursuit (one eye is stationary). The abducens unit's firing rate was modulated regardless of which eye moved. However, the prepositus unit's firing rate was modulated only when the left eye moved. In each panel, the top trace is left eye position (deg), the middle trace is right eye position (deg), and the lower trace is unit firing rate (spisec).
726
ANNALS NEW YORK ACADEMY OF SCIENCES
definitions, a unit will be related monocularly to the right eye if ocular selectivity is 1 and monocularly to the left eye if ocular selectivity is - 1. If the ocular selectivity is 0, then the unit’s discharge is equally related to movements of either eye (binocular unit). Ocular selectivity was studied in burst-tonic or tonic neurons located in the abducens nuclei, nucleus prepositus hypoglossi, and the cerebellum.
RESULTS
FIGURE 1 shows examples of a binocular unit recorded in the abducens nucleus (panels A-C) and a monocular unit in the prepositus hypoglossi nucleus (panels D-F). The discharge of the abducens unit was modulated during conjugate movements (A), and during asymmetric tracking with the right (B) or the left (C) eye. The ocular selectivity indices were 0.04 for position and 0.23 for velocity, indicating that this was a binocular unit with a firing rate related to movements of both eyes. In contrast, the discharge of the prepositus unit was modulated only in relation to 1E and F. The ocular movements of the left eye as can be seen by comparing FIGURES 2 selectivity indices of this unit were -1.0 for both eye position and velocity. FIGURE shows the distribution of ocular selectivity indices for 74 abducens units (A), 18 prepositus units (B), and 27 cerebellar units (C). Abducens neurons could be divided into three nonoverlapping groups (F(2,69) = 3 9 7 . 4 , ~< 0.0001) according to ocular selectivity for eye position: monocular left (OSK = -0.87,47% of units), monocular right (OSK = 1.0, 9.7%), and binocular units (OSK = -0.15, 43.1%). In contrast, the majority of preopositus and cerebellar units were monocular (prepositus, 77.8% of units; cerebellum, 66.7% of units). Similar groups could be defined using the ocular selectivity indices for velocity.
DISCUSSION These data suggest that unit discharges in oculomotor pathways may selectively influence the movements of one or the other eye. In an earlier study, we showed that position-vestibular-pause (PVP) cells in the vestibular nuclei also exhibit monocular eye position sensitivity,3 suggesting that monocular ocular selectivity is a common feature of premotor eye movement pathways. These data imply that an analysis of ocular selectivity could provide a physiological means to identify premotor pathways associated with the production of motor commands for each eye. Identification of such pathways (reminiscent of the classically described three-neuron vestibuloocular reflex pathways) would be a crucial step in understanding how the TVOR produces disjunctive eye movements dependent on target location and eye position. The ocular sensitivity analysis also suggests that monocular pathways converge on abducens neurons, since many of these cells are binocular. It is likely that the physiological division of abducens units into subpopulations of monocular and binocular units does not correspond to the anatomical division of abducens units into subpopulations of motoneurons and internuclear neurons. Thus, motoneurons may be monocular or binocular. We speculate that significant convergence of left and right eye movement pathways on abducens cells might be present at birth. During the early development of conjugate eye movements, these convergent connections could be reinforced selectively on some neurons by synchronous activation of synapses in left/right monocular pathways during conjugate movements, but weakened on other neurons during disjunctive eye movements. Thus, the distribution of ocular selectiv-
ZHOU & KING OCULAR SELECTMTY
721
h
u
v
Em
-+4
.-
s I
z
2 0
hl P
5
m
$2 a z t
728
ANNALS NEW YORK ACADEMY OF SCIENCES
ity in t h e abducens nucleus may result from t h e s a m e mechanism that produces ocular dominance columns i n visual cortex. We further speculate that ocular selectivity will be altered by surgical, optical, or pharmacological treatments that disrupt binocular alignment during later life. REFERENCES 1991. Eye movement responses to linear head motion in the 1. PAIGE,G. D. & D. L. TOMKO. squirrel monkey. 11. Visual-vestibular interactions and kinematic considerations. J. Neurophysiol. 65: 1183-1196. 2. KING,W. M. & W. ZHOU.1995. Initiation of disjunctive smooth pursuit in monkeys: Evidence that Hering’s law of equal innervation is not obeyed by the smooth pursuit system. Vision Res. 3 5 3389-3400. K. M. V., R. D. TOMLINSON, W. M. KING,G. D. PAIGE& E. Q. NA. 1994. Eye 3. MCCONVILLE, position signals in the vestibular nuclei: Consequences for models of integrator function. J. Vest. Res. 4 391400.
