Vision Res.

Vol. 12, pp. 2015-2022. Pcrgamon Press 1972. Printed in Great Britain

SACCADES

AND THE QUICK

PHASE OF NYSTAGMUS

SAMUEL RON,’ DAVIDA. ROBINSON~ and ALEXANDERA. SKAVENSKI~

The Clayton Laboratories,Departmentof Medicineand the Departmentof Biomedical Engineering,The Johns Hopkins University, Baltimore, Maryland 21205, U.S.A. (Received

18 November 1971; in revised form 28 April 1972)

ALTHOUGHseldom explicitly stated, it seems to be often assumed that the two forms of rapid eye movements, the saccade and the quick phase of vestibular nystagmus are identical (COHEN and SUZUKI, 1963; MACKENSENand SCHUMACHER, 1960; RJZINHART and ZUBER, 1969; ROBINSON,1968 and WESTHEIMJZR, 1954). This notion probably originates in the desire for the simplicity which results when it is assumed that both these movements are generated by the same rather than separate neural mechanisms. A search of the literature seems to indicate that this idea has not yet been substantiated by quantitative experimental results. Such results were not provided in the studies by COHEN and SUZUKI (1963), GOTO, TOKUMASI and COHEN(1965), PULEC (1968) and UEDA and SUZUKI(1965), to mention only a few. If saccades and the quick phase of nystagmus were distinctly different, that finding would considerably complicate our ideas of the organization of the oculomotor system. For quick phases to be the same as saccades, the shape of their time courses and the relationship between their amplitude and duration should be the same. Such a quantitative comparison cannot be found in the literature. In fact, many of the quick phases in published records of nystagmus, elicited by various methods such as caloric or electrical stimulation, show little resemblance to saccades in their amplitude-duration relationship or the shape of their trajectories (COHENand SUZUKI, 1963; PULEC, 1968 and YAMANAKAand BACH-YRITA, 1970). Consequently, the present study was undertaken to compare the shape and amplitude-duration relationship of saccades and quick phases of vestibular nystagmus in the alert monkey. It has also been generally assumed that the time course of a quick phase of nystagmus was independent of the method by which it was induced and also independent of the nature of the visual surroundings. It turns out that these things make marked changes in the quick phase and saccadic time courses and this investigation also describes the nature and extent of these changes. METHODS Experiments were carried out on five monkeys. Under pentobarbital sodium anesthesia, each had chronically implanted in it a coil for recording eye movements, a crown to immobilize the head and a chamber through which stimulating electrodes could be lowered to the cerebellum and brain stem. The coil of fine wire was implanted on the monkey’s globe beneath Tenon’s capsule and the four recti insertions. When used in conjunction with alternating magnetic fields, the coil provides an accurate method for recording vertical and horizontal eye movements with a sensitivity of 15’ arc and a bandwidth of 2 kHz. These techniques have been described in detail elsewhere (FUCHSand ROBINKIN, 1966; ROBINSON,1963).The t Present address: Department of Ophthalmology, The Chaim Sheba Medical Center, Tel Hashomer, Israel. 2 Present address: Department of Ophthalmology, Woods Research Building of the Wilmer Institute, The Johns Hopkins University, Baltimore, Maryland 21205, U.S.A. 2015

2016

SAMUEL RON. DAVID A. ROBINSON AND ALFXANDER A. SKAVENSKI

crown was a light aluminum ring that encircled the calvarium and was fastened to it by eight stainless steel screws whose sharpened tips penetrated the upper table of the skull. The chamber was implanted over a trephine hole in the skull. The dura was left intact and electrodes or cannuli were passed into the brain within a sterile guard tube. The identification of stimulated sites was later verified by histological reconstruction (RON, 1971). The animals were supported post-operatively by chloramphenicol and penicillin. All the experiments were done, in subsequent daily recording sessions, on the unanesthetized, alert animal. Normal saccades were studied by utilizing the spontaneous movements the alert animal made in looking about the laboratory (referred to as a structured visual field). Spontaneous saccades were also studied in total darkness and with an illuminated ganzfeld by placing a blank opaque surface just in front of the monkey’s eyes. Saccades under these conditions were studied when the animal was kept alert by food and novel auditory stimuli (periods of ganzfeld and darkness were kept short to prevent loss of alertness) and also when alertness was allowed to decrease by periods of quiet when the animal appeared to be bored or drowsy. Vestibular nystagmus induced by rotation was also studied under these three visual conditions. To obtain rotational nystagmus, the primate chair and eye movement measuring system were mounted so that the animal could be rotated in a horizontal plane. A potentiometer measured the angle of the head in space and the magnetic field eye monitor measured the angle of the eye in the head. Vestibular nystagmus was arti5cially induced by electrical stimulation of the vestibular nuclei, injecting KCI into the flocculus, and caloric irrigation. Monopolar electrical stimulation was done with a cathodal train of constant current, @5 msec pulses at a frequency of XlO/sec and a typical stimulus train length of 2 sec. About 40 sites were stimulated in the vestibular nuclei and vestibular nerve. The method of cortical spreading depression (by injecting potassium chloride) has been used extensively to achieve temporary, functional elimination of certain parts of the central nervous system (h&GIRIAN and Buaas, 1970). Floccular spreading depression was caused by injecting 1-6 ~1. of 25% KCI into the flocculus through a 25 gauge hypodermic tube. Nystagmus results presumably by depolarization of Rurkinje fibers, projecting from the flocculus, causing disinhibition of the vestibular nucleus on that side. Subsequent histological reconstruction veri5ed the location. These experiments were repeated three times (separated by intervals of at least one hour) at five different flocullar sites in one monkey. To evoke caloric nystagmus, water at various temperatures (0”. 10” and 20°C) was injected in the right ear. Ten trials wem made when both ears were irrigated: the right ear with cold water and the left ear with hot water (max. WC). A total of 26 caloric stimulation experiments were done on four monkeys. All the data were recorded on a tape recorder, retrieved from the tape and printed by an ultraviolet mirror galvanometer direct writing recorder with a frequency response (-3 dB) of 1.9 kHz.

RESULTS

Alertness and visualjield structure Figure l(a) shows some typical spontaneous saccades made by an alert monkey. The relationship between the amplitude and duration (A and D, Fig. l(a)) of a saccade is well known for man (BECKERand FUCHS, 1969) and monkey (Fuchs, 1967). Our own data, which agrees with those of others, are shown in Fig. 2, curve SF., for the typical juvenile rhesus monkey. We have adopted the attitude that any rapid eye movement which is not reasonably described by this curve is not “normal”. The velocity of saccades is evidently a sensitive index of alertness. As an animal becomes used to the experimental situation it soon becomes listless and if no attempt is made to interest it, it often falls asleep. Figure 1(b) shows examples of “saccades” made by a drowsy monkey whose eyes, however, are still open. They have been chosen as extreme examples to illustrate how slow the movements can become. A few more examples are shown by the data points in Fig. 2 marked x. They resemble many of the abnormal eye movements seen during REM sleep episodes in this animal by FUCHS and RON (1968). In intermediate states between alertness and drowsiness, less dramatic but statistically significant slowing of saccades occurs. We have not attempted to quantify this because we have no independent measure for alertness. In fact, saccadic eye velocity is probably a more sensitive measure of alertness than most other parameters commonly used. Whenever eye movements are analyzed from recording periods when little control over alertness was exercised, saccadic

Saccades and the Quick Phase of Nystagmus

0

200

400 Time,

660

800

msec

FIG. 1. Examples of saccades and nystagmus quick phases in normal visual surroundings. (a), Normal voluntary saccades in the alert monkey. Method of determining amplitude, A, and duration, D, shown. (b), Slow saccades when the animal is drowsy. (c), Rotatory nystagmus. (d). (e) and (f), Examples of the various types of nystagmus that can be elicited by artificial means such as electrical stimulation, caloric stimulation or KCI injection. All records have the same time scale. t50-

I IO

20 Amplitude,

30

40

50

deg

FIG. 2. Amplitude-duration relationships of saccades (solid lines) and quick phases (dashed lines) of rotational nystagmus (NY.) during various visual conditions; S.F., structured visual field (the normally illuminated laboratory); Gnz., a ganzfeld; Drk., total darkness. Lines are linear regression lines, each fitted to at least 50 data points obtained from one monkey. Fiftyfour data points are shown for the Ny. SF. regression line to show typical scatter. Similar data points for other curves are omitted for clarity. Points marked x are individual saccades, as in Fig. l(b), that occurred when the animal was drowsy.

WSlON12112-F

2017

2018

SAMUEL

RON.DAMDA. ROBINSON AND ALEXANDER A. SKAVENSKI

durations showed a great deal of scatter in plots such as Fig. 2 with the scatter always in the direction of longer durations. In total darkness, spontaneous saccades have a longer duration as shown in Fig. 2, curve Drk., even when the animal is kept alert by loud, unusual noises. This phenomenon has already been observed in human saccades by BECKERand FUCHS(1969). When there is illumination but the animal is deprived of visual form by a ganzfeld (Gnz. in Fig. 2), saccadic durations lie between the normal (S.F.) and dark (Drk.) curves. This observation is in contrast to the case for humans (BECKERand FUCHS, 1969) where saccades performed in a ganzfeld or in darkness showed the same amplitude-duration relationship. Our results indicate that alertness and both structure and illumination in the visual field all markedly affect the speed of saccades. Rotatory nystagmus The majority of the quick phases of nystagmus elicited during rotation differed slightly from voluntary saccades in shape; they had more rounded ends as shown in Fig. l(c). However, the amplitude-duration relationships of quick phases were close to those of voluntary saccades performed under the same three visual conditions. Some actual data points and the linear regressions obtained for these three relationships are shown in Fig. 2 (i.e. for a structured visual field-NY. S.F., a ganzfeld-Ny. Gnz., and in the dark-NY. Drk.). The linear regression lines of both types of rapid movements under the same experimental conditions shifted together from fast saccades and quick phases when performed in a structured visual environment to slower ones in the dark. Thus, the speed of rapid eye movements was altered by the visual state of the animal regardless of whether the movement was a spontaneous saccade or a vestibular quick phase. However, the quick phases were always slightly faster than spontaneous saccades of the same amplitude for a given visual state (Fig. 2). An interesting corollary of this observation is that saccades are thus not performed by the oculomotor system in the shortest possible time. It should also be remarked that the amplitude-duration relationship of quick phases did not depend on the stimulus intensity, that is, on the speed of rotation. Abnormal nystagmus There are three differences observed for artificially induced (non-rotational) nystagmus: the quick phases were almost always slower than normal, most of them had unusual shapes and their velocities (or durations), for a given amplitude, could be altered by changing the stimulus intensity. Examples of quick phases during artificially induced nystagmus are shown in Fig. 1 (d-f). These records were chosen from periods of intense stimulation to better illustrate large quick phases at high trace speeds. As a result, the slow phase velocities (close to 100°/sec) are not a great deal less than the quick phase velocities (typically 3OO”/sec).Lower stimulus intensities result in a more normal sawtooth nystagmic pattern. Jn all cases the most striking feature is that the durations of the movements are abnormally long. This is made clear by the amplitude-duration relationships in Fig. 3 for the quick phases of nystagmus induced by caloric and electrical stimulation and spreading depression. For all amplitudes the duration was much longer than for rotatory quick phases; by as much as 240 per cent for small movements. The regression line for caloric stimulation is given for only one stimulus intensity, and for the KC1 injection the regreaaion line pertains to one short period of the evoked nystagmus because the quick phase amp&udeduration relationship may change over long periods as the intensity of the depression changes. The

Saccades and the Quick Phase of Nystagmus

2019

were conducted several times on different days on each animal and while the results differed slightly from test to test, all the curves lay near those shown and were, in any event, above the normal curve. An example of the dependency of quick phase duration on stimulus intensity is given in Fig. 3. For stimulation of the vestibular nuclei with three progressively increasing currents one obtains three different linear lines of amplitude vs. duration (El. Ny. 1, 2, 3). As illustrated in Fig. 3, when stimulus intensity (current) was increased from El. Ny. 1 to El. Ny. 3, the duration of the quick phases decreased. In both electrical and caloric stimulation, where artificial stimulus intensity could be easily varied, the speed of the quick phase increased with stimulus intensity, the best physiological measure of which is, of course, slow phase velocity. experiments

150.

0

IOO-

z

E 5 ._ z 5 D

50-

L

IO

1

20 Amplitude,

30

40

50

deg

FIG. 3. The amplitude-duration relationships of nystagmus quick phases in a structured visual field during artificially induced vestibular nystagmus; El. Ny., electrical stimulation at the three intensities: 1,0.2 mA; 2,0.4 mA and 3,0.6 mA; Cal. Ny., caloric nystagmus; KCI. Ny., spreading floccular depression by KC1 injection and S.F. normal voluntary saccades (as in Fig. 2) for comparison. Data points were omitted for clarity. Each linear regression line was calculated from at least 60 data points obtained from several experiments in the same monkey.

DISCUSSION

The major question to which this research was addressed is whether saccades and quick phases are generated by the same neural mechanisms. If the two curves F.T. and Ny. F.T. in Fig. 2 had been indistinguishable the answer would be clearly yes. Up to 25”, the two types of movements are very similar but for larger movements the difference is quite apparent and it is clear that quick phases are slightly faster than saccades; I8 per cent faster for a 40” movement, for example. The difference is not due to idosyncratic differences between animals (which do exist) for it persists when saccades and quick phases are compared in the same animal. However, the difference is slight enough that it seems unnecessary to hypothesize separate neural circuits for the two types of rapid eye movements. The small difference may only indicate that rapid eye movements are initiated in the same neural circuit but in slightly different ways by the visual and vestibular systems.

2020

SAMUEL

RON. DAMDA ROBINSON AND ALEXANDER A. SKAVENSKI

On the other hand, the fact that visual field content (structure, ganzfeld, dark) affects quick phases and saccades in the same way is not only a novel finding but is strong evidence that the same neuronal circuit does, in fact, produce saccades and quick phases. One often tends to think of the vestibular-ocular reflex as an elementary brain stem system which stabilizes the visual axes in space without much regard for what the eyes see. This is evidently not so, for whether the retina is illuminated or filled with contrast contours can affect the quick phases of nystagmus, and in the same way that these things affect saccades. In summary, the data presented here on saccades and naturally induced vestibular quick phases tend to quantitatively substantiate the commonly held belief, as shown qualitatively by some investigators (COLLARD, CONTAUX, THIEBAUT and THIEBAUT, 1967 ; JUNG and KORNHUBER, 1964), that the two movements are the same and are the product of the same neuronal mechanism. This would further suggest the hypothesis that a saccade reflects the ability of the brains of frontal eyed, binocular, foveate animals to control and utilize brain stem neuronal circuitry for rapid eye movements inherited from ancestors without such a varied oculomotor repertoire but who had, nevertheless, a well developed vestibular-ocular reflex. Whether nystagmus is artificially induced, probably by any means, the result is quite different. The quick phases, while still relatively quick, bear little resemblance to normal saccades and quick phases. It is interesting to note that had one compared saccades and the quick phases of artificially induced nystagmus, they would appear to be the products of quite different mechanisms. We suggest that the slowing of artificially induced quick phases may be due to some sort of brain stem depression that is similar to the slowing of saccades and naturally induced quick phases during loss of alertness or loss of visual field content. We have noticed, in working with this monkey preparation for about 5 years, that the speed of saccadic eye movements is a good index of the state of alertness. For those interested in measuring or controlling alertness in man or animals saccadic velocity could be a very useful parameter to observe. The drop in velocity with loss of alertness might also be of interest to physiologists who work with animals who are paralyzed by cervical transection in which alertness is difficult to obtain or assess. If loss of alertness can cause a marked drop in saccadic velocity the same mechanism may also be at work during artificially induced vestibular nystagmus. Anyone who has experienced caloric nystagmus will know how nauseating and disorienting it is. There is a direct conflict between vestibular, visual and somatosensory afferents concerning rotation in space. Most important, one cannot maintain fixation on any object in the environment and this situation almost never occurs naturally under any combination of rotations of the subject or his environment with respect to an inertial reference frame. Thus the visual sensation is not merely in conflict with both vestibular and somatosensory information, it is self-contradictory since it lies outside the realm of natural experience. The fact that several monkeys vomited during electrically evoked nystagmus suggests that the experience is probably as distressing for them as it is for us. If rather subtle changes in the state of alertness can cause noticeable changes in the speed of saccades, it seems reasonable that the altered levels of excitability that occur during the distressing and conflicting sensations of abnormally induced nystagmus might easily show up in altered patterns of quick phase movements. Whatever the cause, artificial nystagmus, although a useful tool in neurophysiology, should be viewed with caution if one wishes to study quantitutiueZy the time course, mechanics and neural events of vestibular quick phases or saccades.

Saccades and the Quick Phase of Nystagmus

2021

Acknowledgements-This research was supported by Grant EY-00598 from the National Eye Institute of the National Institutes of Health of the U.S. Public Health Service. Computer time was provided by National Institutes of Health Training Grant GM-00576 to the Biomedical Engineering Department of The Johns Hopkins University. REFERENCES BECKER,W. and FUCHS, A. F. (1969). Further properties of the human saccadicsystem: Eye movements and correction saccades with and without visual fixation points. Vision Res. 9, 1247-1258. COLLARD,M., CONTAUX,C., THIEBAUT,M. S. and THIEBALIT,F. (1967). Le nystagmus d’origine cervicale. Rev. Neurol. Paris, 117,677-688. COHEN, B. and SUZUKI, J. (1963). Eye movements induced by ampullary nerve stimulation. Am. J. Physiol. 204, 347-351. DODGE, R. (1903). Five types of eye movement in the horizontal meridian plane of the field of regard. Am. J. Physiol. 8, 307-329. FUCHS, A. F. (1967). Saccadic and smooth pursuit eye movements in the monkey. J. Physiol., Land. 191, 609-631. FUCHS, A. F. and ROBINSON,D. A. (1966). A method for measuring horizontal and vertical eye movements chronically in the monkey. J. uppl. Physiol. 21, 1068-1070. FUCHS, A. F. and RON, S. (1968). An analysis of the rapid eye movements of sleep in the monkey. Electroenceph. clin. Neurophysiol. 23,244-251. Gore, K., TOKUMASI,K. and COHEN,B. (1965). Return eye movements, saccadic movements and the quick phase of nystagrnus. Acta Oto-laryngol. 65,426-440. JUNG, R. and KORNHUBER,H. H. (1964). Results of electronystagmography in man: the value of optokinetic, vestibular and spontaneous nystagmus for neurologic diagnosis and research. In The Oculomotor System (edited by BENDER,M. B.), pp. 428-488. Harper & Row. MACKENSEN,G. and SCHUMACHER,J. (1960). Die Geschwindigkeit der raschen Phase der optokinetischen Nystagmus. Albrecht v. graefes. Archs. Ophthal. 162, 400-415. MEGIRIAN,D. and BURES,J. (1970). Unilateral cortical spreading depression and cortical eye blink responses in the rabbit. Exp. Neural. 27, 34-45. PULEC, J. L. (1968). Clinical electronystagmography. Laryngoscope, 78,2033-2048. REINHART, R. J. and ZUBER, B. L. (1969). Cellular activation patterns in the abducens nucleus during horizontal nystagmus in the cat. Brain Res. l&284-287. ROBINSON,D. A. (1963). A method of measuring eye movements using a scleral coil in a magnetic field. IEEE Trans. on Bio-med. Elect. BME-10, 137-145. ROBINSON,D. A. (1968). The oculomotor control system: a review. Proc. IEEE, 56, 1032-1049. RON, S. (1971). A quantitative study of eye movements evoked by cerebellar stimulation in the alert monkey. The Johns Hopkins University, Ph.D. Dissertation. UEDA, R. and SUZUKI, J. (1965). Studies on the eye-speed during voluntary ocular movements in induced nystagmus in normal test-subjects. In Proc. International Symposium on Vestibular and Oculomotor Problems. Univ. Tokyo, 89-94. WESTHEIMER, G. (1954). Mechanism of saccadic eye movements. A.M.A. Archs Ophthol. 52,71&724. YAMANAKA,Y. and BACH-Y-RITA,P. (1970). Relations between extraocular muscle contraction and extension times in each phase of nystagmus. Exp. Neural. 27,57-65. Abstract-Quick phases of vestibular nystagmus and saccades were compared on the basis of the shape of their time courses and their amplitude-duration relationships in the alert intact monkey. Both rapid movements are almost identical when nystagmus is induced naturally by rotation. Visual field structure, illumination and darkness affect the speed of both types of movements similarly. This indicates that the same pre-motor neural circuit probably creates both saccades and quick phases. When nystagmus is created artificially (e.g. caloric or electrical stimulation) the quick phases are quite abnormal; their speed being reduced by a factor of two to three. R&m&-On compare sur le singe intact et tveillC les phases rapides du nystagmus vestibulaire et les saccades, d’apres la forme de leur &olution temporelle et leurs relations ampiitudeduree. Les mouvements rapides sont presque identiques tous deux si le nystagmus est induit naturellement par rotation. La structure du champ visuel, l’klairage et l’obscuritt affectent d’une faGon semblable la vitesse des deux types de mouvements. Ceci indique que le m&me

2022

SAMUEL.

RON,DAVIDA. ROBINSON AND ALEXANDER A. SKAVENSKI

circuit nerveux pfemoteur crCe probablement a la fois les saccades et les phases rapides. Quand le nystagmus est cr&? artificiellement, par stimulation calorique ou Blectrique, les phases rapides sont tout-a-fait anormales, leur vitesse &ant mduite dans un facteur de deux ou trois. ZHssmineafsissuagSchnetle Phasen des vestibuhtren Nystagmus und Sakkaden wurden beim wachen und unve&nten At% hinsichthch ihres Zeitvertaufes und ihrer Amplituden-Zeitdauer Verhiiltnisse verglii. Die beiden schnellen Bewegungen sind nahezu identisch, wenn der Nystagmus auf natbrlich Weise durch Rotation hervorgerufen wird. Die Struktur des Gesichtsfelde-s, Helligkeit und Dunkelheit beeinflussen die Geschwindigkeit beider Rewegungen auf g&he Weise. Das ist ein Hinweis darauf, daB miiglicherweise den&be vor-motor&he neuronale Schaltkreis sowohl die Sakkaden als such die schnelen Phasen erzeugt. Wenn der Nystagmus ktinstlich erzeugt wird (etwa durch thermische oder elekttische Stimulation), dann sind die schnellen Phasen abnormal. Ihre Geschwindigkeit ist dann urn den Faktor zwei oder drei reduziert. Pearo~&rcTp&.re @aamaecTsr6ynnpnoro rinc~ar~ H cam 6t.m~ cpaariea~ no @opMe XXnpOTeKari%Uf BOBpCMeB&i kl COOTIiOmeFmaM McTJly ax aMM.uTynoti H mn-9 Y o6e~an. Q6a ~nxa 6r.rc~pirx rcapmtIIlEBnom ~fr(earamt~, 6orrpcTBYrorJuntB korna ElicTarM BbUbmaeTca ecTecTBeHn0 IIyTeM v. CTpyrTypa noJill Spearur, ocBeKIernfea remora ,uetfcTnyroTcxoJnibnu O6pa3OMHa cXOpocTbo6ortX TEnoB JtBnne?nnn. 3~0 yrra3br~ae~aa TO, 9~0, Bepom~o, om x TOTnte npeh4oTopnbtti net&ponarmnT&mnot t$opMapyeTEaEc.ar~~anbt,TaKH6bIcTp~t ~U~~I.&JIEA~CT~I-MB~~CXECK~~~~.IO (wanpm.sep, nprr ~~oparectcoft tuni anercTmptecrtoticrnruynamin) 6r~rp~e +a314 o~mnximo-