haut de page

mise à jour du
17 juillet 2005
Arch Ital Biol
1972; 110; 3 348-382
Sleep-wakefulness, EEG and behavioral studies of chronic cats without neocortex and striatum: the "diencephalic" cat
Villablanca J, Marcus R
Departments of Psychiatry and Anatomy University of California, Los Angeles, USA


The present research was initiated with the intention of studying the electrothalamogram (EThG) in relation to overt behavior in chronic "diencephalic" cats, i. e., cats without neocortex and striatum. In what appears to be the only previous study in unrestrained, totally neodecorticate cats chronically implanted with subcortical electrodes, the EThG exhibited a constant low voltage (20 µV) fast activity irregardless of the behavioral state This EThG pattern was recorded the day following the operation and remained unchanged for several months. However, there are reports of 8-12 c/sec waxing and waning spindle bursts in the thalamus of acute, immobilized, neodecorticate and "diencephalic" cats; this activity has essentially the same features of cortical spindle activity of intact cats during drowsiness and non-rapid eye movement sleep (NREMs). The pioneer experiments of Morison and Bassett are particularly interesting because the spindle bursts which they observed in the thalamus of neodecorticate cats anesthetized with pentobarbital were present for up to three postoperative days but disappeared thereafter.
Due to the above discrepancies, and since the objection has been raised that the EThG synchrony in the acute experiments may be due to irritation or diaschisis it was felt that a reexamination of the EThG and behavior of unrestrained freely moving decorticate cats was justified. A clarification of this problem could shed light upon the role of the thalamus in the generation of rhythmic activity in the alpha range of the-human EEG as well as upon its behavioral significance.
Early in the course of this EThG study, marked changes were noticed in the sleep-wakefulness patterns, especially in rapid eye movement sleep (REMs). These observations led to a systematic study of the sleep-wakefulness cycle in these " diencephalic" animals, which became an important part of this investigation. Recent research has emphasized the role of caudal brain stem mechanisms controffing REMs. However, the finding of motor hyperactivity in neodecorticate and "diencephalic" cats and dogs could imply that forebrain lesions alter the sleep-wakefulness cycle. Furthermore a virtual absence of NREMs in decorticate cats and a clear, although transitory suppression of REMs after lesions of the basal forebrain area in cats has been recently demonstrated. Because of these last mentioned results and data reported by other authors and ourselves, it was felt that a demonstration of a consistent alteration in sleep-wakefulness in " diencephalic" cats would further document forebrain participation in the control of sleep and wakefulness.
It is well known that barbiturates induce EEG spindle burst activity in intact animals. This effect has also been shown to occur in the diencephalon acutely separated from both rostral d caudal brain stem structures. Since it has been claimed hat barbiturates do not induce spindle activity in decorticate eats, it was decided to reexamine their effects in the "diencephalic" cat. Quite unexpectantly, it was noticed that thiopental induced changes in the sleep-wakefulness patterns. This effect wasstudied in detail since it was interpreted as a pharmacological interference witha hypopthetical mechanism of the rostral brain controlling sleep and wakefulness.
In summary, this paper is concerned with the following aspects of "diencephalic" cats:
1) neurology and overt behavior;
2) sleep wakefulness;
3) EEG-behavioral correlations;
4) effects of thiopental on the EEG and sleep-wakefulness. 352-358
1. - Neurology and overt behavior.
1. Locomotion and reflexes. - The "diencephalic" cats were able to walk on the second postoperative day. During the following one and a half months they were extremely hyperactive, their main usually for hours (a maximum of 17 hours of, uninterrup ed walking was recorded in one cat). Any attempt to stop the cats only enhanced the strength of their progression. In three cats, this tendency was so vigorous that if stopped they would fight, hiss, and growl, and when released, would jump forward. Although the hyperactivity declined progressively, they spent a large part of their time restlessly standing, sitting, or crouching; episodes of obstinate progression persisted in relation to defecation or micturation until they were sacrificed.
Some degree of locomotor ataxia, including moderate "goose step "-like gait was observed during the first two postoperative days. Later, locomotion was slower, and clumsy and there was a tendency toward ventroflexion of the head and lowering of the spine. During the early postoperative days, hyperextension of the legs could be induced by suspending the cat by the scruff of the neck or when the Magnus reflex was tested in dorsal decubitus; however, there was no extensor rigidity or hyperextension of the neck when the cat stood or walked. Indeed, stretch reflexes. appeared to be diminished in these animals since the tendon jerks were not exaggerated and patting the pelvic girdle or the spine produced flexion of the limbs, thus suggesting a weakness of the positive supporting reaction. Curiously enough, this latter maneuver always stopped the animal's ongoing movements. On the other hand, the righting reflexes were hyperactive. The cats struggled to recover their upright position after being placed on the floor on their flanks or when suspended from the pelvis with the head hanging. In the cats which lived for up to six months there was a marked flexor rigidity such that when the cats were suspended from a flexed elbow the spine arched forward and the head and hind limbs flexed strongly.
Tactile placing and hopping reactions were absent and only a coarse placing of the forelegs was observed in the long-term cats. Their sitting position was awkward, in that the hind legs were usually thrust forward between the forelegs.
A few apparently purposeful movements were occasionally observed; for example, they could use their forepaws skillfully in an attempt to reject the insertion of the eye pieces of the feeding tube and this was accompanied by avoiding movements of the head.
2. Reaction to external stimuli. - During the first two months, the cats shivered when they were changed to a room slightly cooler than the vivarium, when placed upon cold surfaces or when tubefed a cool meal. This reaction, however, was not reflected by changes in rectal temperature. The threshold to nociceptive stimuli (pinprick, heat) was not changed in the acute state however, in the chronic condition pain sensitivity appeared to be rather enhanced. Although olfactory stimuli induced vigorous sniffing and searching, the cats were unable to locate the stimulus source; the olfactory behavior was absent in those cats in which the olfactory tracts were ablated.
There were no overt signs of vision in" diencephalic" cats. They did not show a visual-palpebral reflex and in walking they bumped into objects irregardless of their size or brightness. In contrast to this apparent blindness, however, the pupils reacted briskly to changes of illumination, and even a very weak light source, & g., a burning match, produced pupillary constriction. A strong light source always produced (after the third mouth) a type of avoidance reaction in that the pupils constricted intensely, the nictitating membranes covered the eye and the head deviated away from the source of illumination.
One of the most striking reactions to external stimuli was the exquisite reactivity to auditory stimulation. Subtle noisse triggered brisk body movements with ocular arousal, perkings of the ears and orientation of the head toward the source of the noise; this reaction usually rapidly habituated. After the third month, a stronger, sudden sound, e. g., a handclap, induced a startle reaction often accompanied by hissing and growling. In two cats, sudden, loud noises produced an escape reaction with sudden, fast, slinky running.
3. Eating and licking. Licking and swallowing occurred when food was placed in front of the animal's nostrils. Tube feeding apparently induced some type of conditioning, i. e., the chronic cats reacted with salivation to the introduction of the feeding tube but later salivation occurred simply by handling the cat at the customary feeding time.
In four of the five chronic cats, spontaneous licking occurred. They often licked their chests and paws with occasional attempts to rub their face with few, weak, paw strokes. This grooming activity, however, was completely inefficient. In three cats after the fourth postoperative month, licking the floor, other cats and the walls often occurred, this activity being triggered by disturbing stimuli.
4. Other behavior. - Two to three months following decortication, the cats often engaged in many iterative motor activities, e. g., soft mewing, licking, or various movements of the legs (scratching, kneading, etc.). These activities were exaggerated by noises or by handling to which some cats reacted with continuous mewing. Stronger handling, like brushing, or pressing their bladders out, often triggered growling, hissing, loud guttural mewing or more rarely, biting. Thus, a striking behavioral feature of all chronic " diencephalic" cats was their increased irritability in response to minor environmental changes.
The only " social" trait observed in these cats was their tendency to congregate when placed in the same corral. This behavior is probably related to the finding that when placed on a hard, cold floor, but having a small, soft pad, the cats invariably were later found on the pad. After 2-3 months there were other signs of " awareness" to objects or surfaces; thus the cat would stop or turn around after touching a wall with its face (or even whiskers) or upon reaching a border (i. e., the edge of a table) with a paw. A similar behavior occurred when a foot was wet or soiled; in this case the leg was lifted and shaken vigorously as an intact cat would do.
II. - Sleep-wakefulness patterns.
Four stages were identified in the sleep-wakefulness continuum of the "diencephalic" cats. The cat was considered to be awake when it either stood, walked, or engaged in any other activity, e. g. licking or mewing, or when it was quiet but kept its eyelids open, nictitating membranes retracted, eyeballs centered in the orbit and its pupils dilated. The animal was judged to be drowsy when it sat quietly or crouched with eyelids half-closed, nictitating membranes protruding variably and the pupils reduced and fluctuating in size (although no smaller than 5 mm in width) with the eyeballs exhibiting slow, often dissociated movements. The cat was considered to be in NREMs when it lay in a random or curled position with the eyelids closed, nictitating membranes protruding and eyeballs tending to rotate inward and downward, with the pupils smaller than 5 mm in width and exhibiting less fluctations. Finally, the animal occasionally lapsed into typical episodes of REMs with all the behavioral and ocular features described for this stage of sleep in intact and chronic decerebrate cats.
After one month the "diencephalic" cats progressively recovered their presomnic and postsomnic postural patterns which further helped to define the sleep-wakefulness stages. Thus, crouching often occurred at the onset of oculo pupillary sleep patterns and the typical curled posture of intact cats was frequently observed when the eyes exhibited a deeper sleep pattern. Likewise, at the end of a sleep period, the cats often stretched their limbs, arched their backs and even yawned, as intact cats do upon awakening.
There was a large reduction in the sleeping time of all " diencephalic " cats. During the first two sessions, NREMs time decreased to an average of 17% of the total observation period, or less than 50% of the time control cats spend in this stage. REMS time decreased to 1,1% of the observation time or less than onetenth of the control value. Both NREMs and REMs time further decreased and reached a stable level only after the second month. During the last four months, NREMs was under 5% of the total observation time-or about 1/8 of the controls. REMs time was even more markedly affected since after the second month it decreased to an average of 0.65% of the observation time or less than 1/20th of the controls. In fact, REMs was completely absent in ten out of 48 sessions whereas NREMs was never absent. Drowsiness was not significantly affected. The normal cyclic distribution of drowsiness or of the remaining NREMs or REMs could not be demonstrated. Feeding did not affect the distribution of sleep time.
Since the cats in the chronic state were exquisitely reactive to noise and the noise level in the observation room was not uniform day and night, a comparison was made between time spent in sleep from 8:oo AM to 8:oo PM and from 8:oo PM to 8:oo A. M. It was found that the cats slept about the same during the day (6.2% NREMs and 0.8% REMs) as during the night (5.4% NREMs and 1.0% REMs).
Two other features differentiated the sleep of " diencephalic" cats from that of intact cats: 1) the mean duration of REMs period was shortes in " diencephalic " cats (4-1/2 mm) than in intact cats (6-1/3 mm) (35); 2) the amount of phasic muscular events during REMs was higher in "diencephalic" than in control cats.
page 369
2. Sleep-wakefulness. - Although observations concerning sleep wakefulness in decorticate animals have been reported previously, several factors limit their comparability with the present results. First, in most cases the authors were only secondarily interested in sleep. Secondly, REMs was not known as a separate entity at the time many of those experiments were performed. Thirdly, although the extent of the decortication was not always clearly stated, it appears that usually only a neodecortication was done. Finally, quantitative studies of the sleep-wakefulness patterns were not performed in the earlier experiments.
Hyperactivity has been a common finding in decorticate dogs. Kleitman and Camille reported a predominance of locomotor activity in decorticate dogs but they did not provide any figures concerning the actual sleeping time of their animals. Hypercativity after decortication has been likewise described for cats and in a more sleep-oriented study, it was found that decorticate cats sleept approximately eight hours whereas intact cats slept for 18-20 hours.
Motor hyperactivity in decorticate animals can be interpreted as the result of a "release" phenomenon. The question then arises as to why this hyperactivity decreased after one to one and a half months. The reasons are probably complex and may involve anatomical (degenerative) alteration of the subcortical structures originally "released" by decortication, compensatory neuronal processes and even prolonged sleep-deprivation.
On the basis of their studies, Mettier et al. questioned the existence of hyperactivity in decorticate animals. An important methodological difference which may explain the discrepancy between their study and the present findings is that their dogs were decorticated in several stages over a period of six months. It has been shown that the effect of a CNS lesion is greatly attenuated or changed when it is performed in several steps.
There appears to be only one extensive EEG and behavioral study concerning both REMS and NREMs in decorticate animals. In that study, NREMs was reported to be almost completely absent whereas REMs was within the normal range (15-20% of the observation time).
The present study appears to be the first demonstrating a permanent, pronounced reduction in both REMs and NREMs following a rostral brain lesion. During the first postoperative month, the insomnia might be attributed to an indirect effect of the animal's motor hyperactivity. However, since the hyperactivity decreased following the second month while the sleeplessness did not, the insomnia is probably a primary phenomenon. An increased sensory input does not appear to be a causative factor either inasmuch as there was no difference in the sleep-wakefulness pattern of the cats when studied under different levels of environmental noise (nighttime versus daytime study).
Unlike the present experiments, Jouvet's study was performed in cats with the limbic system and striatum essentially intact. Furthermore, it has been repOrted that there is a permanent behavioral activity in "diencephalic" cats and rats, whereas neodecorticate cats are inactive and neodecorticate rats are only moderately active. Therefore, it appears that the presence of the striatum and/or rhinencephalon is important for a normal sleep-wakefulness alternation.
The fact that chronic cats with high complete mesencephalic transection do not manifest a marked decrease in either REMs or NREMs suggests that rostral structures do not exert much control over caudal brain stem sleep mechanisms. However, the present finding in "cliencephalic" animals necessitates a revision of this concept. Accordingly, we postulate that in the intact animal, both enhancing and suppressing rostra.l brain influences impinge upon caudal brain stem sleep mechanisms and modulate their function in a balanced manner. Consequently, mesencephalic transection, by completely removing both types of influences, would not induce too large a disturbance. On the other hand, a functional imbalance might occur when only some parts of the rostral brain are removed. More specifically, we postulate that in "diencephalic" cats, a net loss of sleep-enhancing influences descending upon the brain stem from either the neocortex, striatum, rhinencephalon, or from all of these areas results in sleeplessness. The sleep-suppressing effects may hypothetically be ascribed to the hypothalamus and/or subthalamus. There is evidence that these structures posses arousal properties and both structures were undisturbed by surgery in our experiments. Moreover, the hypothalamus has been demonstrated to remain functional even after complete isolation. The hypothalamus may primarily favor wakefulness rather than directly. supress sleep; if this is the case, our basic hypothesis of an imbalance among elements of the descending rostral brain control would still be tenable.
At least two forebrain areas - i. e., the caudate nucleus and the orbitofrontal cortex - ablated in the present experiments have been demonstrated to be involved in sleep, EEG synchronization and behavioral inhibition. The basal forebrain area, which was not directly lesioned in the present cats, also shares these properties and, according to Clemente, is the main component of a descending synchronizing system, which acts reciproally with. the ascending reticular activating system to regulate .... "the level of cortical activation or synchronization.....". The lesioning of the above forebrain areas most probably contributes to the insomnia of the "diencepbalic" cats in the present study However, mainly on the basis of the lack of any striking insomnia in cats with high mesencephalic transection, we propose that the descending modulatory action of the rostral brain consists of both activating and deactivating influences rather than of a descending deactivating system oppossing the action of an ascending activating system.