Symposium on the Role of the Vestibular Organs in Space Exploration (1970)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Evoked Potential and Microelectrical Analysis of Sensory Activity Within the Cerebellum' RAY S. SNIDER AND KARL LOWY University of Rochester SUMMARY The extensive to-and-fro connections between the cerebellum and the vestibular system have been reviewed. Purkinje-cell inhibition to the vestibular system is recognized in the direct pathways, and the role of the crossed fastigiovestibular pathway is discussed. There is no overlap of auditory and vestibular areas in the cerebellar cortex, but there may be in the nuclei fastigii. In the cerebello- cerebral projections, there is overlap at the cerebral levels. Microelectrode studies on the auditory area indicate that it has electrophysiological properties similar to those reported for other cerebellar areas. The function of the cerebellum in the habituation of nystagmus is discussed and some electro- physiological interpretations given. The material to be presented is divided into three parts. Part I is a review of some of the interrelationships between the cerebellum and the vestibular system, including connections with the cerebrum. Part II presents evoked- response and microelectrode data collected from the cerebellar cortex, with special emphasis on the auditory and vestibular systems. Part III contains data pertinent to a discussion on cere- bellar function involved with habituation of nystagmus. By way of emphasizing developmental rela- tionships, figure IA is shown. The basic vestib- ular area (in black) arises from the alar plate and develops into a "typical" cerebellar cortex. In lower vertebrate forms, this is the dominant part of the cerebellum which develops. As shown by the concentric dotted lines, however, the newer portions of the cerebellum in higher animals mushroom upward from this area. The base continues to be related to the vestibular system, whereas the newer portions of the cere- bellum receive fibers from both the tactile and proprioceptive systems. The tail, leg, arm, and 1 This work was supported in part by grants NB-04592 and NB-06827 from the National Institutes of Health. face receiving areas are shown. There is one area in the anterior lobe and one in the posterior lobe. Sensory impulses arising from the tail region of the animal are located adjacent to the vestibular system. As the sensory areas repre- senting the surface of the animal spread out on the dorsal surface, the two face areas overlap. There are only minor differences between the projections of the two systems, and one usually speaks of the tactile and the proprioceptive areas as being coextensive. The face areas are repre- sented largely by projections from the trigeminal system, and there is overlap with projections from two additional sensory modalities; i.e., auditory and visual (ref. 1). It is curious that the auditory area is located so far from the vestibular area, since both systems originate in end organs so intimately related. Instead, the auditory area is almost coextensive with the visual area, and they both overlap the face areas as is shown in figure IB. This is a dorsal view of the cerebellum and the midline structures labeled, according to Larsell's terminology, lobules I through X. To the right of the drawing are shown the figurines as viewed from the dorsal surface. The vestibular areas are located at either end, and the tactile and proprioceptive 24 S

246 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION A A, FIGURE 1. —A: Basic drawing of lateral view of cerebellum with vestibular areas (black) at base. Smaller area to left represents portion in anterior lobe. Concentric dotted lines represent growth of cerebellum, especially neocerebellum. Tactile and proprioceptive receiving areas for the tail (T), leg(L), arm (A), and face (F) are indicated. The dorsal stippled area represents visual and auditory receiving areas. B: Basic drawing of dorsal view of cerebellum showing figurines spread on surface to represent tactile receiving areas. Roman numerals on left correspond to Larsell's subdivisions. The auditory receiving area (stippled) is in lobule VI and part of VII. The vestibular areas (stippled) are in lobules I, IX, X. C: Basic drawing of dorsal surface of cerebellum showing (stippled) vestibular areas, fastigial nuclei, and vestibular end organ (nuc), vestibular nuclei. Arrows represent connections. representations of the different parts of the body represented in between. In the region of lobule VI are the auditory and visual areas. Figure 1C summarizes the major vestibulo- cerebeUar connections. As a result of the studies of Cajal (ref. 2), Ingvar (ref. 3), Larsell (ref. 4), and Dow (ref. 5), the afferent projections were well established. As shown, there is both a direct projection from the end organ and an indirect one containing an additional synapse in one of the vestibular nuclei, and, for unknown reasons, both go to the same basic areas of the cerebellum (ref. 6). Also shown in figure 1C is a summary of the cerebellovestibular connections. Dow (refs. 5 and 7) showed direct fibers from flocculonodular lobe to the lateral and medial vestibular nuclei. The recent studies of Walberg and Jansen (ref. 8) furnish the evidence of a direct pathway from the anterior and posterior lobes to the vestibular nuclei, while the studies of Brodal, Pompeiano, and Walberg (ref. 9) furnish the details of the projection of the cerebeUar nuclei to the vestibular nuclei. A direct pro- jection from the flocculonodular lobe to vestibular hair cells has been described by Llinas et al. (ref. 10). In summary, it is pointed out that there are two cerebellovestibular pathways, and, with few exceptions, vestibular structures which send

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 247 impulses to the cerebellum also receive impulses from it —often, from the same area. The func- tional implications of these to-and-fro projections will be discussed below. Figure 2 shows the presence of various evoked responses within the (tuber vermis) auditory area of the cerebellum when the cerebral cortex was stimulated: A, auditory area; B, visual area; C, somatic area. Although there was extensive overlapping of these three cerebral systems with- in this region (see ref. 11 for details), this is the only part of the cerebellum from which visual and auditory responses could be obtained. Auditory responses could not be recorded from the ves- tibular areas, and vestibular responses could not be recorded from auditory areas. However, as shown in figure 3, there was definite interaction of visual and auditory impulses in this area inasmuch as a conditioning "flash" input altered the re- sponsiveness of the test (click) input signal, which followed at varying intervals up to 120 msec when Flaxedil was used as an anesthetic, and for 360 msec when Chloralosane was used. Since it is difficult to find overlapping pathways in these two systems, it is not unreasonable to assume that there is overlapping within the cerebellar cortex. Furthermore, as shown in figure 4, there is interaction of impulses arising in the auditory area of the cerebrum and click-induced impulses within the auditory area of the cerebellum, with a recovery time between conditioning and test stimuli of as much as 400 msec when Chloralosane anesthetic was used. These experiments cannot be performed satisfactorily upon animals under barbiturate anesthesia (ref. 11). On the basis of the data presented thus far, one might conclude that there is no place within the nervous system where the vestibular and audi- tory systems overlap. However, this is not true because there is overlap in the least likely of all places, the cerebellocerebral projections. As shown in figure 5, electrical stimulation in the tuber vermis elicits evoked responses within the auditory area of the cerebrum in addition to responses in the visual area (ref. 12), while electrical stimulation within the vestibular areas of the cerebellum (fig. 6), i.e., lingula and uvula and also the flocculonodular lobe, as shown by 250 Uv 10ms B FIGURE 2. —Evoked responses recorded from tuber vermis (dorsal, first turn). Cat under Chloralosane medication when responses in left column were taken; under Flaxedil medication when responses in right column were taken. A: Middle ectosylvian (auditory) cortex stimulated, 8.5 V, O.I msec, single biphasic pulse. B: As A, except visual cortex (posterolateralis) stimulated. C: As A, except 5.0 V to posterior cruciate gyrus (somatic receiving area). Five traces superimposed. Calibration signals are 250 fi.V (vertical) and 10 msec (horizontal). Ruwaldt and Snider (ref. 13), evokes potentials in a cerebral area overlapping the cephalic part of the auditory area. In figure 7 is shown a summarizing diagram of these two studies (refs. 12 and 13). The stippled regions represent the auditory areas, while the cross-hatched region represents maximal evoked responses when the vestibular areas of the cerebellum were stim- ulated (diagonal lines represent regions of lower responses). There is obvious overlap of the two in the anterior parts of the auditory area. The vestibular area in the cerebrum overlaps part of the tactile face area, but this is not the case in the cerebellum. It should also be pointed out that Infantellina, Sanseverino, and Sperti (ref. 14) have found physiological evidences of a prominent vestibular projection to the pafaflocculus, and this part of the cerebellum projects to the ves- tibular area of the cerebrum as shown in this diagram. METHODS The microelectrode studies are given in the second part of this paper. The cats were pre- pared for acute surgery for the recording sessions after administering 12.5 mg/kg of the short-acting

248 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION FIGURE 3. —A: Cat under Chloralosane medication (70 mglkg). Recording electrode on first turn of tuber vermis (see insert). Conditioning stimulus: 3-W neon lamp 2 in. from atropined eye —left. Test stimulus: 50-dB click from 8-in. speaker 2 m away. Numerals indicate time in milliseconds between two stimuli. Note 10-msec time signal for traces through 140-msec interval and then slower sweep for the remaining 250 ^V. Unipolar recording. B: As A. except Flaxedil medication was used (15 mglkg). Records taken from tuber vermis (second turn). Insert shows amplitude of test response when compared with control. Note overlapping areas of visual and auditory responses and the shorter recovery time when Flaxedil medica- tion was used. (From ref. II.) anesthetic agent, Surital. The animals were then maintained under Flaxedil medication while attached to a stereotaxic instrument. In some cases Chloralosane, 30 mg/kg, was adminis- tered. A micromanipulator was attached to the stereotaxic instrument in order to allow fine manipulation of the microelectrode. Care was taken throughout the experiments to maintain body temperature above 34° C and, in the case of the Flaxedil-treated animals, to maintain ade- quate respiration. Glass capillary tubing with a tip diameter of 1 to 3 microns (resistance of 5 to 20 megohms), filled with either 3 M potassium chloride or 2 M potassium citrate, was used for recording extra- cellular unit activity. The microelectrodes were attached to probes containing one of the input valves of the cathode follower input. The other side was grounded. The signal was amplified by a Grass amplifier and the re- sponses were photographed from the face of a 5-inch cathode-ray tube. Two types of stimu- lating electrodes were used. One, the so-called concentric electrode, was made by placing in- sulated stainless-steel wire in metal tubing. These electrodes were especially useful when the central white matter of the cerebellum was stimulated. Surface stimuli were applied di- rectly to the pia by means of bipolar. 26-gage silver wire with chlorided tips. Usually the dis- tance between the tips was 1 to H mm. At the conclusion of the experiment, a small direct

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 249 FIGURE 4.— A: Cat under Chloralosane medication (70 mglkg). Recording (bipolar) electrode on lobulus simplex (adjacent to tuber vermis, see insert), and conditioning stimulus was electric shock (6.5 V, 0.2 msec} applied to anterior ectosylvian gyrus. Test stimulus was 50-dB click from 8-in. speaker 2 m in front of animal. Numerals indicate millisecond interval between conditioning and test stimulus. Note faster sweep for intervals through 120 msec. 200-p.V calibration. B: As A, except Flaxedil medication (15 mgjkg) was used and the recording electrode ivas moved to first turn of tuber vermis. See insert for differences in recovery time of conditioning and test responses at various intervals. Amplitude of test response expressed in percentage of control (100 percent) response. current was imposed upon the microelectrode, and a lesion which was usually less than 250 microns in diameter was placed in order to iden- tify the area of the recording tip when histological sections were prepared. These sections were stained by the Nissl method. RESULTS In figure 8 is shown a series of traces recorded from the auditory area of the cerebellum when a single shock was applied directly to the auditory branch of the eighth nerve. The recording elec- trode was adjacent to the Purkinje-cell layer, and the high-amplitude units appear to represent dis- charges of individual Purkinje cells. No stimuli were given in column A (spontaneous activity), but electrical stimuli were applied in order to obtain the tracings shown in column B (see shock artifact at the beginning of the sweep). The inhibition of units lasted for 10 to 20 msec, depending upon the strength of the stimulus. This was an unexpected finding, since excitation rather than inhibition was the anticipated result. The microelectrode data shown in figure 9 illustrate the interaction of cells in the cerebellar cortex. The records were taken from the Pur- kinje-cell layer. As shown in 9/1, stimulation in the white matter of a folium within the first turn of the tuber vermis induced a prolonged inhibi-

250 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION FIGURE 5. —CoJ under Chloralosane (70 mglkg) medication. ^ Electrical stimulation of folia of first turn of tuber vermis (see insert). 12 V, 0.25 msec. Monopolar recording elec- trode. Note localization of responses in cerebral auditory and visual areas. To obtain this localization, it was neces- sary to use stimuli just above threshold. Calibration, IS msec (horizontal) and 500 p.V (vertical). FIGURE 6. — Cerebral responses resulting from electrical stimulation of lingula (B); pyramis (C); uvula (D); and uvula (E). Cat prepared under ether anesthesia and maintained under dihydro-ft-erythroidine. A: Photomicrograph showing electrode tract in lingula showing position of stimulation point (L2)/or the responses shown in B. B: 20-V, 0.5-msec bipolar stimulation. Note localization of responses. C: Evoked responses in cerebrum resulting from surface stimulation of most posterior folium of pyramis (20 V, 1.0 msec). D: As C, except the most anterior folium of uvula was stimulated (see insert). E: As C, except middle folium of uvula was stimulated (see insert). (From ref. 13.)

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 251 FIGURE 1.—Outline drawing of lateral surface of cerebral hemisphere and dorsal view of cerebellum showing (dotted) cerebellar areas which project to dotted areas of cerebrum. The vestibular areas of cerebellum, i.e., lingula (LING), uvula (L'V), nodule (NOD), flocculus (FLOC), and paraflocculus (PFL), project to cross-hatched and diagonal-lined area of cerebrum. Note that the two projection systems overlap in the anterior ectosylvian gyrus. ANT is anterior lobe and P is pyramis. (From ref. 13.) tion of Purkinje-cell activity. As clearly shown by Eccles and associates (ref. 15), such prolonged inhibition lasting approximately 400 msec can be explained by the inhibitory action of Golgi II cells on incoming signals via granule cells, or by direct basket-cell inhibition of Purkinje-cell ac- tivity. In 9B are shown the results of stimulating the pial surface of an adjacent folium and the induction of inhibition of Purkinje-cell activity in the tuber vermis for 200 msec. There are several explanations for this; however, the one we favor is the inhibition induced by the Purkinje-cell axon collateral which passes from one folium to an adjacent folium and induces relatively short inhibition directly on other Purkinje cells. Figure 10 shows additional extracellular micro- electrode records in which the electrode was placed at the junction of the molecular and Purkinje-cell layers of the tuber vermis. In figure 10/1 a single shock was applied to the white matter of the same folium, according to the pro- cedure of Eccles et al. (ref. 15). There was an initial deflection which represented ascending fiber activity and then rhythmic 20/sec discharges for less than 100 msec. Since reduction of the strength of the shock by only 10 percent failed to excite activity, this was arbitrarily assigned as threshold: figure 10B shows similar data col- lected when the shock was increased to l£ times threshold. In this case the same two Purkinje cells fired for 350 msec, while the higher voltage one continued for an additional 150 msec before stopping. These data show not only the rhyth- micity of Purkinje cells but also the sharp locali- zation of active units which can be obtained when small stimulating electrodes are used at near

252 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION FIGURE S. — Decerebrate cat. Single-unit recording via 3 M KCl-filled micropipet with 15-Mfl resistance. The tip was located adjacent to Purkinje-cell layer of a folium in first turn of tuber vermis. Record reads from below upward. A: Spontaneous activitv. The high-voltage spikes were interpreted to represent Purkinje-cell discharges. The origin of the lower voltage spikes was unknown. B: At the beginning of each sweep, n single pulse I'l- i times threshold (9 V) was applied to end (in osseous spinal lamina) of the auditory nen'e. Note the 25- to 30-msec inhibition of activity. Calibration, 10 msec (horizontal) and I mf' (vertical). threshold values. Figure IOC shows the effect of applying a single electrical pulse to the pial surface and recording from the Purkinje-cell layer of the same folium. There is an initial burst of 300 to 500/sec activity which lasts for approximately 100 msec and then gradually slows. This is followed by a period about 600 msec long in which there is inhibition of the higher voltage single units observed in the pre- stimulatory period which, because of the ampli- tude and the electrode tip location, were assumed to be Purkinje cells. As shown by Eccles et al. (ref. 15), basket-cell discharges are fast (up to FIGURE 9. —A, B: Decerebrate cat. Micropipet (3 M KCI) 15-Mfl resistance inserted into folium of tuber vermis. Continuous record: read below upward. Single-shock electrical stimulus induced inhibition of unit activity. Calibration, 100 msec and I mV. 500/sec) and cause enduring inhibition of Purkinje cells. This, plus the fact that these cells are located in the Purkinje-cell layer, in- dicates that these data may be indicative of basket-cell discharges. In figure 11 are shown the effects of local ap- plication of 2 percent strychnine on the pial surface of a folium when the central myelinated fibers of the folium are stimulated electrically. A is the control record showing the arrival of the afferent volleys in the Purkinje-cell layer and a 200-msec inhibitory period which follows. In B are shown the alterations which occur 2 minutes after application of the strychnine, while the traces shown in C include activities which were observed 5 minutes after local application of the

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 253 FIGURE 10.-A. B: Decerebrate cat with 13-Mtl resistance microelectrodes (3 M KCl pipets) inserted into the tuber vermis. Single electrical stimuli were given. Records read from below upward. Calibration, 50 msec (hori- zontal) and 1 mV (vertical). C: Cat under Chloralosane medication (70 mglkg). Record reads from below upward. Single electrical shock introduced first a fast burst of activity, then an enduring inhibition. Calibration, 50 msec and I mV. FIGURE 11. —A. B, C: Shows data collected when a 10-mfl, 3 M KCl-filled micropipet was inserted into the Purkinje- cell layer of the tuber vermis and used as a recording elec- trode. Single electrical shocks were given at /'/2 times threshold. Note prestimulatory and poststimulatory effects. Record reads from below upward. Same re- cording and stimulating sites for A, B, and C. Calibration, 100 msec (horizontal) and 1 mV (vertical). Two percent strychnine applied locally 2 minutes before B was taken and 5 minutes before C was taken. drug. Note reduction in size of the afferent volleys and the shortened inhibitory period. From these data collected from single-unit activity in the auditory-visual area of the cere- bellum, one can conclude that the electrical activity does not differ from that described by Eccles and associates (ref. 15) for the somato- sensory areas. Unfortunately, our studies on the vestibular areas of the cerebellum have not been completed. However, it is likely that there will be few if any unique features of this region other than the type of sensory input used to acti- vate the areas. Thus, unlike the cerebral cortex, the cerebellar cortex possesses intrinsic activity which varies little from area to area. At this point a fundamental question may be asked. If there is this uniformity within the cerebellar cortex, what function does it serve? For example, why have both a direct vestibulo- cerebellovestibular projection in addition to to-

254 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION and-fro connections between vestibular nuclei and the cerebellum? A partial answer may come from an analysis of the complex and subtle func- tional interrelationships which exist between these two structures. (1) Dow (ref. 5) showed that ablation of the nodulus plus part of the uvula in the primate produced a syndrome of disequilibrium which lasted for 1 to 2 months. A transient nystagmus was observed. Bard et al. (ref. 16) showed that flocculonodular lobe lesions in dog prevented motion sickness and that this effect was long enduring. (2) Direct cerebello-oculomotor connections have been known since KlinofTs work (ref. 17). Whiteside and Snider (ref. 18) were the first to give electrophysiological evidence for these and pointed out that the latency of response was so short that direct Purkinje-cell fibers might be considered. Manni, Azzena, and Dow (ref. 19) showed that single units in the oculomotor nucleus were affected by both cerebellar and vestibular nuclei stimulation. With these prominent connections to the oculomotor nucleus, it is easy to accept a cerebellar control of eye movements. Such effects have been known since Mussen's work (ref. 20). Koella (ref. 21) was the first to point out that, so far as eye move- ments are concerned, the cerebellum "appears to be organized with reference to three-dimensional space." The recent work of Cohen et al. (ref. 22) not only supports this concept but emphasizes that the tuber vermis region, i.e., so-called audi- tory-visual area, when stimulated in the un- restrained animal, produces conjugated horizontal eye movements similar to those seen when the horizontal canal (vestibule) is stimulated. Yet there is no evidence to show that the horizontal canal has connections to this part of the cere- bellum. (3) Evidence in the third category is concerned with the little-known role which the cerebellum plays on the habituation of vestibular nystagmus. Halstead first observed in 1935 (ref. 23) that pigeons with cerebellar lesions in the region of the tuber vermis failed to show habituation of vestibular nystagmus for rotatory stimulation. Di Giorgio and Pestellini (ref. 24) made similar studies on the guinea pig. However, the most extensive study has been done by Wolfe (ref. 25), and the next two figures show some of these data taken from unrestrained cat during horizontal canal stimulation. In figure 12, note that the burst of fast activity in nucleus fastigii precedes slightly the slow wave in vermis which occurs during point of reversal; i.e., point of zero acceleration. Additional data show that changes in the activity of the fastigial nuclei are time locked to changes in the activity of the medial vestibular nucleus related to the fast phase of horizontal nystagmus. Figure 13 shows data which indicate that there was a normal nystagmic response at the beginning of record, but 24 seconds after stimulus onset the animal habit- uated to left rotation, while maintaining a normal response to right rotation. A unilateral lesion was placed in right tuber vermis, and 32 hours after surgery this record was taken. Note the slow wave in the nucleus fastigii which appears at zero acceleration during first part of record and which becomes less prominent when animal habituated to left rotation. Seventy-two hours after surgery, the animal did not habituate in either direction. Daily tests on this animal failed to elicit habituation until the 18th postoperative day when the animal reverted to unilateral habituation and on the 21st day showed habitua- tion to both right and left stimulation. Wolfe (ref. 25) concluded that tuber vermis and nucleus fastigii are part of a control mechanism of the fast phases of nystagmus and that the habituation mechanism is located in the tuber vermis. It is difficult to understand the changes which occurred in the infraslow cerebellar potentials (IFSPC) which have a frequency of only five to seven per minute and are too slow to be related to respiration. The IFSPC appeared in all animals before stimulus onset and increased in frequency and amplitude after the stimuli began. However, as the animal habituated, the IFSPC decreased in amplitude and occasionally dis- appeared completely during habituation. Thus Wolfe's work (ref. 25) indicates that the tuber vermis contains neural mechanisms capable of modifying the fast phase of vestibular nystagmus and that there is diminished amplitude of potentials during habituation.

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 255 too tVHMM tret ^ I Nf I CM I SEC FIGURE 12. — Records taken from free-moving cat during horizontal oscillation of platform. The upper trace shows platform movement at 0.33 cps. The second trace shows electro-oculogram (EOG). Note periodicities related to platform oscillation, The third trace shows EEG recording from right side of tuber vermis. The fourth trace shows EEG recording from left thalamo- cortical radiations (LTCR), and the fifth trace shows EEG record taken from left nucleus fastigii (LNF). Calibration, 25 mm/sec (horizontal); EOG = 500 pVIcm, vermis = 100 jtK/cm; LNF andLTCR = 50 nVlcm (vertical). (From ref. 25.) fOC fOC ICMI FIGURE 13. — Records taken from free-moving cat on horizontal oscillating platform with cycle of 0.33 cps (trace I). Trace 2 shows electro-oculogram (EOG). Note the changes in latter half which were correlated with "loss of fast phases to the left and a marked decrement in slow phase output to the left." Trace 3 open. Trace 4 is EEG record taken from right nucleus fastigii (RNF). The bottom three tracings are continued from the first three. Note that the loss of the fast phase of nystagmus {to the left) continues throughout the bottom tracings. Surgical removal of the second turn of tuber vermis was accomplished 32 hours previously. Calibration, 25 mm/sec (horizontal) and 1 mV\cm (vertical). (From ref. 25.)

256 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION DISCUSSION While the sensory receiving areas of the cere- bellum are well known, the present paper stresses the point that, despite the anatomical proximity of the auditory and vestibular end organs, the receiving areas are rather widely separated. However, the two systems may converge at the level of the nucleus fastigii and they definitely overlap in the cerebral receiving areas. (See fig. 7.) The biological significance of these cere- bellocircuits has not been worked out (ref. 26). Hopefully, microelectrical analysis of these cerebellar areas when paired with measurable behavioral performance will provide some answers. With the studies of Brookhart, Moruzzi, and Snider (ref. 27) and Granit and Phillips (ref. 28), and especially the recent studies of Eccles et al. (refs 15 and 29), a challenging start has been made. The inhibitory function of the basket, Golgi II, Purkinje, and stellate cells has been established (ref. 29). Of special interest to the present discussion is the paper of Ito, Obata, and Ochi (ref. 30) which showed the direct inhibitory action of Purkinje cells on cells in the lateral vestibular nucleus, as well as the indirect effect of Purkinje cells on nucleus fastigii, thence to cells in lateral vestibular nucleus. A sig- nificant unknown in most studies on efferent cerebellar systems is the role of the nuclear cells. Except for cerebellovestibular connec- tions, the Purkinje cell should be considered a very important interneuron (not an effector) synapsing on the efferent neuron in the cerebellar nuclei. The cerebellar control of eye movements is considered along with the work of Fernandez and Fredrickson (ref. 31), which showed that lesions of the nodule produced disequilibrium, positional nystagmus, and prolonged vestibular reactions to rotatory and caloric tests. Of special interest to the present discussion was the prompt and consistent inhibition of nystag- mus produced by electrical stimulation of the nodule. Usually the eyes showed conjugate deviation toward the side of the slow component as if the stimulation inhibited the mechanism from whence the fast component originated. Evidence is also given for a cerebellar role in habituation of the fast phase of nystagmus. However, it is difficult to obtain data on how the cerebellar cortex functions during habituation because the time course is so much longer than the electrophysiological measurements associ- ated with the stretch reflex, for example. Wolfe's data (ref. 25 ) would indicate that the tuber vermis is an improtant part of this circuit. However, since there are no known vestibular connections to the tuber vermis, then the main afferent volleys remain unknown unless they come from the proprioceptive activity of eye muscles. The studies of Higgins, Partridge, and Glaser (ref. 32) on limb musculature indicated that the "cerebellar effect results in muscle tension- leading stretch response in phase under condi- tions which should stabilize the stretch reflex, thus reducing an inherent tendency to oscillate." If nystagmus were considered a special case of eye-muscle oscillation, then the cerebellum could reduce an "inherent tendency to oscillate" and participate in physiological mechanisms under- lying habituation. Under these conditions, lesions of the cerebellum would temporarily eliminate habituation until other areas could assume part of the function. While experimental evidence seems com- patible with some of these assumptions, the basic difference between the precise short- term regulatory function controlling skeletal and eye muscles and the long-term effects of habituation must not be overlooked. Two obvi- ous gaps must be bridged before such a functional role of the cerebellum can be established. (1) Additional data must be obtained concern- ing the dependence on proprioception of cere- bellar control of eye muscles, and (2) the com- plexities of habituation must be analyzed with data relevant to the function pf the cerebellum in long-enduring motor performance. Such ex- perimental data are overdue since, as early as 1943, Rosenblueth, Wiener, and Bigelow (ref. 33) suggested that the "main function of the cerebellum is the control of the feedback nervous mechanisms involved in purposeful motor activity."

MICROELECTRICAL ANALYSIS OF SENSORY ACTIVITY 257 REFERENCES 1. SNIDER, R. S.: Recent Contributions to the Anatomy and Physiology of (he Cerebellum. Arch. Neurol. Psychiat., vol. 64,1950, pp. 196-281. 2. CAJAL, S. R.: Histologie du Systeme Nerveux de 1'Homme et des Vertebres. Maloine, Paris. 1909. 3. INGVAR, S.: Zur Phylo- und Ontogenese des Kleinhirns nebst einem Versuche zu einheitlicher Erklarung der zerebellaren Funktinn und Lokalisation. Folia Neuro- biol., Leipzig, vol. 11, 1918, pp. 205-495. 4. LARSELL, O.: The Cerebellum: A Review and Interpreta- tion. Arch. Neurol. Psychiat., vol. 38, 1937, pp. 580- 607. 5. Dow, R. S.: Effect of Lesions in (he Vestibular Part of the Cerebellum in Primates. Arch. Neurol. Psychiat., vol. 40,1938, pp. 500-520. 6. BRODAL, A.: Anatomical Aspects on Functional Organi- zation of the Vestibular Nuclei. Second Symposium on the Role of the Vestibular Organs in Space Explora- tion, NASA SP-115,1966, pp. 119-139. 7. Dow, R. S.: The Fiber Connections of the Posterior Parts of the Cerebellum in the Rat and Cat. J. Comp. Neurol., vol. 63, 1936, pp. 527-548. 8. WALBERG, F.; AND JANSEN, J.: Cerebellar Cortico- vestibular Fibers in the Cat. Exptl. Neurol., vol. 3, 1961, pp. 32-52. 9. BRODAL, A.: POMPEIANO, O.; AND WALBERG, F.: The Vestibular Nuclei and Their Connections. Anatomy and Functional Correlations (Ramsay Henderson Trust Lectures). Oliver & Boyd, Edinburgh, London, 1962. 10. LLINAS, R.; PRECHT. W.; BRACHE. F.; AND HUERTAS, J.: The Inhibitory Cerebello-Vestibular System in the Frog. Proc. Intern. Union Physiol. Sci., vol. 7,1968, p. 269. 11. DEURA, S.; AND SNIDER. R. S.: The Interaction of Vari- ous Impulses in the Cerebellum. J. Neurol. Sci., vol. 1,1964, pp. 178-196. 12. HENNEMAN, E.; COOKE, P. M.; AND SNIDER. R. S.: Cere- bellar Projections to the Cerebral Cortex. Proc. Assoc. Res. Nerv. Ment. Dis., vol. 30, 1950, pp. 317-333. 13. RUWALDT, M. M.; AND SMDER, R. S.: Projections of Vestibular Areas of Cerebellum to the Cerebrum. J. Comp. Neurol., vol. 104, 1956, pp. 387-402. 14. INFANTELLINA, F.; SANSEVERINO, E. R.; AND SPERTI, L.: Cerebellocerebral Relationships: Projections of Cere- bellar Paraflocculus on the Cerebral Cortex in the Cat. Arch. Sci. Biol., vol. 49, 1965, pp. 97-122. 15. ECCLES, J. C.: LLINAS, R.: AND SASAKI. K.: The Inhibi- tory Interneurones Within the Cerebellar Cortex. Exptl. Brain Res., vol. 1, 1966, pp. 1-16. 16. BARD, P.; WOOLSEY, C. N.; SNIDER, R. S.; MOUNT- CASTLE, V. B.; AND BROMILEY, R. B.: Delimitation of Central Nervous Mechanisms Involved in Motion Sick- ness. Fed. Proc., vol. 6, 1947, p. 72. 17. KLINOFF, J.: Ueber die Leitungshahnen des Kleinhirns. Arch. Anat. Physiol. Anat. Abt., 1899, pp. 11-27. 18. WHITESIDE, J. A.; AND SNIDER, R. S.: Relation of Cere- bellum to Upper Brain Stem. J. Neurophysiol., vol. 16, 1953, pp. 397-413. 19. MANNI, E.; AZZENA, G. B.; AND Dow, R. S.: Cerebellar Influence on the Unitary Discharge of Oculomotor Nuclei and Adjacent Structures. Exptl. Neurol., vol. 13, 1965, pp. 252-263. 20. MUSSEN, A. T.: Cerebellum and Red Nucleus. Arch. Neurol. Psychiat., vol. 31, 1934, pp. 110-126. 21. KOELLA, W. P.: Motor Effects From Electrical Stimu- lation of Basal Cerebellum in the Unrestrained Cat. J. Neurophysiol., vol. 18,1955, pp. 559-573. 22. COHEN, B.; GOTO, K.: SHANZER, t;.: AND WEISS, A. H.: Eye Movements Induced by Electric Stimulation of the Cerebellum in the Alert Cat. J. Exptl. Neurol., vol. 13, 1965, pp. 145-162. 23. HALSTEAD, W.: The Effects of Cerebellar Lesions Upon Habituation of Post-Rotational Nystagmus. Comp. Physiol. Monogr., vol. 12, 1935, pp. 1-30. 24. Di GIORGIO, A. M.: AND PESTELLINI. G.: Inibizione Acquisita dei Riflessi Vestibolari: Significato Degli Emisferi Cerebrali e del Cervelletto. Arch. Fisiol., vol. 48, 1948, pp. 86-110. 25. WOLFE, J. W.: Evidence for Control of Nystagmic Habituation by Folium-tuber Vermis and Fastigial Nuclei. Acta Oto-Laryngol., suppl. 231, 1968. 26. SNIDER, R. S.: Functional Alterations of Cerebral Sensory Areas by the Cerebellum. The Cerebellum, Progress in Brain Research, C. A. Fox and R. S. Snider, eds., Elsevier Publishing Co., Amsterdam, London, New York, 1967. 27. BROOKHART, J. M.: MORUZZI, G.; AND SNIDER, R. S.: Spike Discharges of Single Units in the Cerebellar Cortex. J. Neurophysiol., vol. 13. 1950, pp. 465-486. 28. GRANIT, R.; AND PHILLIPS, C. G.: Excitatory and In- hibitory Processes Acting Upon Individual Purkinje Cells of the Cerebellum in Cats. J. Physiol., vol 133, 1956, pp. 520-547. 29. ECCLES, J. C.; ITO. M.: AND SZENTAGOTHAI, J.: The Cerebellum as a Neuronal Machine. Springer- Verlag, New York, 1967. 30. ITO, M.; OBATA, K.: AND OCHI, R.: The Origin of Cere- bellar Induced Inhibition of Deiters Neurones. II. Temporal Correlation between the Trans-synaptic Activation of Purkinje Cells and the Inhibition of Deiters' Neurones. Exptl. Brain Res., vol. 2. 1966. pp. 350-364. 31. FERNANDEZ, C.: AND FREDRICKSON. J. M.: Experimental Cerebellar Lesions and Their Effect on Vestibular Functions. Acta Oto-Laryngol.. vol. 58, suppl. 192. 1964. pp. 52-62. 32. HIGGINS, D. C.; PARTRIDGE, L. D.: AND GLASER. G. H.: A Transient Cerebellar Influence on Stretch Responses. J. Neurophysiol., vol. 25, 1962, pp. 684-692. 33. ROSENBLUETH. A.: WIENER. N.: AND BIGELOW. J.: Behavior, Purpose and Teleology. Philosophy of Science, vol. 10, 1943, pp. 18-24.

258 THE ROLE OF THE VESTIBULAR ORGANS IN SPACE EXPLORATION DISCUSSION Ito: You are producing evoked potentials in various parts of the cerebral cortex by stimulating the cerebellum. I wonder what pathways are responsible for this rather short- latency evoked response. Snider: I can answer it in relation to the auditory system but not in relation to the vestibular system. I can show you the subsequent data, if you wish additional details. The cerebellar projection passes through the dorsal aspect and synapses in the nucleus fastigii. From the nucleus fastigii it goes forward in the brachium conjunctivum to the anterior medial zone of the medial geniculate body. There is also a second pathway that goes through the reticular formation and is relayed into the cerebral cortex from there. From this anterior and medial zone in the medial geniculate body which is rather narrow, there is a projection to the cerebral cortex. In general, one might consider this the older por- tion of the medial geniculate body, and maybe that is why it has been missed before. Now, of course, the question arises: Is there not a vestibular area nearby, particularly the parieto- occipital for example, which may act as a relay through to the cerebrum? Perhaps, but it is too early to say so. I would appreciate any information you may have on this. Whiteside: Dr. Snider, the question is perhaps an ele- mentary one, but in regard to eye-movement control, I do not understand how to equate the afferent fibers from the extraocular muscles with the apparent absence of a posi- tion sense in the eye, which is functionally claimed and in- deed demonstrable. You can move the eyes and demon- strate quite definitely that there is no position sense as far as the individual is concerned. I am now talking about normal and conscious man. Snider: The trigeminal projection to the cerebellum is rather well outlined, and includes projections not only from cutaneous sensibility but also proprioceptive endings of eye muscles. (S. Cooper, P. M. Daniel, and D. Whit- teridge: Muscle Spindles and Other Sensory Endings in Extrinsic Eye Muscles; Physiology and Anatomy of These Receptors and of Their Connections With Brain-Stem. Brain, vol. 78, 1955, pp. 564-583.) According to Cupedo (R. N. Cupedo: A Trigeminal Midbrain Cerebellar Fiber Connection in the Rat. J. Comp. Neurol., vol. 124, 1965. pp. 61-69), the mesencephalic root of the fifth cranial nerve sends fibers into the base of the cerebellum via the brachium conjunctivum and overlaps with the regions that are called an auditory- and visual-receiving area as well as a proprio- ceptive-face area. One might say that the auditory and visual signals coming into the cerebellum are rather efficiently orga- nized because it is coming into that portion of the cere- bellum which, when electrically stimulated, causes eye move- ment. In other words, when a sudden sound occurs to one side, the eyes are turned, as is the head, toward the signal. The same response holds for a visual input. But the question of where proprioception fits into the movement of the eyes cannot be answered. One would believe that, during nystag- mus, there would be a cerebellar interrelation with the ves- tibule. However, to my knowledge it is not possible to elicit a stretch reflex from extraocular muscles. So there is a gap here, and one has to be very cautious in interpreting cerebellar function. If you think of the cerebellum as being a neurological comparator of exquisite sensitivity, then you can put almost anything in it, including learning mecha- nisms. It is tempting, for example, to call it the head- ganglion-of-habit formation. Also, I would point out that very few lesions of the cere- bellum produce a permanent motor deficit. Compensation readily takes place. One is reminded of some older experi- ments of Spiegel et al. in which they removed one labyrinth and noticed recovery from nystagmus, and then placed a subsequent lesion and the nystagmus returned, but to the other side. This would indicate to me that the cerebellum was one of the compensating mechanisms in the nervous system for the nystagmus. This does not exactly answer your question, but the information is not available. I would appreciate any comments you have.