Nervous System

 

I Evolutionary and Behavioral Aspects of The Brain

The study of the evolution of the brain and behavior is a paradox. We cannot study behavior in fossils or the internal structure of the brain of organisms long dead. In our lifetimes, we experience but a very brief interlude in the long evolutionary process. What leads us to believe that it is possible to investigate the evolution of the brain and its relation to behavior?

An answer lies in the encouragement investigators obtain from examining the brains of those contemporary organisms which have close links with their fossil ancestors. Patterns in the brain are found which are common to these and different species suggesting parallels with common ancestry. Comparative behavior provides similar encouraging parallels. In practice, it is reasonable to suggest that comparative neurology and comparative behavior have information to offer which is relevant for evolutionary theory and, conversely, that evolutionary perspective provides a framework for making sense out of comparative material.

Historical highlights and approaches

C. L. Herrick was moved by such considerations when he founded the Journal of Comparative Neurology in 1891. His original vision was that the study of comparative neuroanatomy could be integrated with the study of comparative behavior.

Students of C. L. Herrick, such as C. J. Herrick and G. E. Coghill, continued in the tradition of the elder Herrick after his death. C. J. Herrick (1948) concentrated on the connections found in amphibian nervous systems, relating his information to the findings of workers who used other forms and interpreting the possible significance of the described connections. Coghill (1929) spent much of his life in the study of the development of the nervous system of the salamander, in conjunction with a study of the development of behavior. Both Coghill and C. J. Herrick worked in an era of considerable interest in comparative neuroanatomy. Much of the work of the era was summarized by Ariëns Kappers, Huber, and Crosby (1936).

Concurrently, others studied comparative behavior effectively without a simultaneous study of the structure of the nervous system, but they did not overlook the possibility of relating behavior to the nervous system. Among these were Yerkes (1916), Hunter (1913), Loeb (1918), Noble (1931), and Parker (1919). Characteristically, their studies were concerned with discriminative capacities, learning abilities, tropisms, stereotypical behavior, and sensitivities in various species. [See the biographies ofHunterandYerkes.]

A third group, frequently employing methods of intervention in the nervous system as a means of studying the relation of the nervous system to behavior or various units of behavior, was led by such outstanding men as Lashley, Sherrington, and Pavlov. Lashley (1929) was noted for his studies of the role of the mass of cerebral tissue in learning and intelligence. Sherrington (1906) was concerned primarily with the problems of the organization of the nervous system in the regulation of reflex actions. Pavlov (1927) had a similar broad concern but an entirely different approach, stressing the general concept that the study of the elicitation or suppression of reflexes by systematically paired, concurrent stimulation was the key to understanding the role of the cerebral cortex. [See the biographies ofLashley; Sherrington;Pavlov.]

Current status

Studies of comparative neurology and comparative behavior continue today both as separate ventures and as combined enterprises. Although the work of many of the men whose names are mentioned here and of numerous contemporary workers has never been specifically addressed to the problems of the evolution of the brain and behavior, much of the data is relevant.

The information accumulated and available is overwhelming but riddled with hidden error, gaps, and misconceptions. Since the work on behavior and the nervous system is being done on various species, we are obliged to make explicit our views of the relationships of these organisms, including man, if we hope to extrapolate results from one species to another. Specifically, we must clarify the evolutionary perspective with which comparative work is to be viewed.

Attributes of the central nervous system

The common origins of the various vertebrate species are reflected in the organization of the brain and spinal cord. Man, as would be expected, possesses the neuroanatomical characteristics of the vertebrates in general, the mammals in particular, and the primates especially. Those portions of man’s nervous system which are common to the vertebrates have been described as primitive, and those which man shares with mammals—pre-eminently with the primates—are referred to as recent portions of the nervous system. For example, man possesses “paleocortex” (primitive outer cell layers of the forebrain) and “neocortex” (more recently evolved outer cell layers of the forebrain). Among the primates, progressively greater proportions of neocortex are found in successively evolved representatives (Clark 1960).

Fourth-level sensorimotor continuity

Interneurons of the spinal cord, serving spinal afferents, and neurons of the brain stem, serving cranial nerve afferents, distribute axons to the cerebellar cortex. The cerebellar cortex, in turn, connects with deep cerebellar nuclei, which send axons to the medial reticular formation. These connections form the fourth level of sensorimotor continuity. The arrangement is one which allows sensory feedback to have a regulative impact on the actions of the medial reticular formation. Broadly speaking, sensory elements responding to muscle tension and elements responding to moderate cutaneous stimulation are sources of negative feedback. Elements responding to intense stimulation such as pain are sources of positive feedback, which serves to sensitize all reflexes except the first-level reflex to the local stimulus.

The vestibular nerve connects directly with the fourth level. Typically, the result of vestibular activity is dependent on asymmetry of stimulation of the vestibular receptors on the two sides. Thus, if the right side of the head is lower than the left, the extensors of the right side are stimulated, while the flexors of the left side are stimulated. The action is magnified by the vestibular contribution to the fourth level, the fourth level acting on the third level. The balance between the two sides of the body which is achieved is characteristic of the role of the cerebellum in equalizing the muscle tone (sensitivity of postural reflexes) on the two sides of the whole body during stationary postures.

Although the cerebellar level is represented in all vertebrates, it becomes progressively more elaborate as the vertebrates themselves exhibit greater differentiation of segmental structures.

Fifth-level sensorimotor continuity

The midbrain tectum and associated central gray also receive directly from rostrally conducting interneurons of the spinal cord and brain stem. The tectum sends fibers to the reticular formation and upper spinal cord. The fifth level of sensorimotor continuity is thus formed. Optic, auditory, vestibular, and somatic sensory systems converge on the midbrain tectum and/or central gray. These sensory systems provide guidance for primitive, whole-body reactions. The reactions, like the isolated reflexes of the first and second levels, fall into two mutually exclusive categories. They consist of whole-body reactions toward or away from a stimulus. Towardturning and away-turning form one dichotomy, while forward lunging and backward progression form another. The importance of this level for food-seizing, whole-body approach (including sexual approach), and whole-body withdrawal is evident. Coincident with approach and withdrawal, the midbrain (central gray particularly) is implicated in the linkages which define “distress” or “well-being” in terms of concurrent intrinsic reactions such as “distress calls,” “cooing” vocalization, autonomic changes, and endocrinological reactions.

Fifth-level connections are common to all vertebrates, but they are especially well developed in birds. The fifth level is probably most important to mammals when they are as yet immature. When the seventh level (recently evolved forebrain level) reaches maturity in mammals, it provides a more elaborate and flexible guidance system for wholebody activities. In the long history of the vertebrates, the functioning fifth level probably made possible the evolution of the seventh level, which, in a sense, supersedes the fifth level. In the ontogeny of the mammal, the immediately functioning fifth level allows for the more gradual development of the seventh.

Sixth-level sensorimotor continuity

The same order of sensory interneurons which entered into connection with levels 3 and 5 connects with the primitive (mid-line and intralaminar) thalamus. The primitive thalamus, besides having numerous central interconnections, connects with the striatum and hypothalamus. Both the striatum and hypothalamus connect with the midbrain level. It is suggested that the striatum paces somatic activity and the hypothalamus paces the concurrent visceral activity, both the striatum and hypothalamus being dependent on the connections through the primitive thalamus. The primitive thalamus is, thus, critical in adjusting the rate of reaction of the organism. Perhaps because of its systematic central connections, the presumptive outcome of the activity conducted through the primitive thalamus would be to increase or decrease the rate of sustained withdrawal or approach at the same time concurrent visceral activity is intensified or retarded. To be of survival value, the outcome must be in accord with the demands placed on the organism, as these demands are represented by the convergence of sensory input to the thalamus.

Level 6, together with its associated primitive forebrain connections, is represented in all vertebrates. The contribution of the olfactory nerve is represented at the sixth level as it reaches the striatum, hyypothalamus and paleocortex (or its counterpart in primitive forms) almost directly. On theoretical grounds, olfactory stimulation is thus in a position to determine the rate of visceral and somatic activity. The contribution to the cortex is likely to play a vital role in the memory of specific odors. Various odors are, or become, important for determining whether the organism hastens its approach or retreat. Other substances indicate whether the organism is in territory where it usually comes to rest.

The nerve cell

The primary morphological element for information transmission in the nervous system is the neuron, a special type of cell with projections extending from the cell body. In physiology, those processes which conduct electrochemical impulses toward the cell body are called dendrites and are essentially protoplasmic extensions of the cell body. They usually extend a short distance and tend to branch profusely. In this fashion a great increase in the surface area of the nerve cell is provided, so that many other neurons may be linked with it. There are, however, large numbers of neurons in the nervous system which have no dendritic processes. The single projection which carries information away from the cell body is the nerve-action potential and the return to the resting potential require from 0.5 millisecond to 2.0 milliseconds. When an action potential is generated in a segment of a fiber, the resulting entry of ions into the cell from outside produces depolarization of the adjacent membrane as well. This marginal depolarization acts as the stimulus to a further breakdown in permeability along the membrane, and the process continues as a chain reaction along the length of the fiber. In this manner the nerve impulse is propagated over the length of the fiber in a fashion not unlike that in which ignition travels along the fuse of a firecracker. The range of velocity of the nerve impulse is very wide, varying from approximately 1 to 300 meters per second, depending upon the diameter of the fiber and other factors. The size of the nerve-action potential is not a function of the strength of the stimulus applied to the nerve. It is true that the referred to as the axon or nerve fiber. The axon is frequently of considerable length, and although it may possess collateral branches along its extent, the greatest branching is at its termination. Axons either terminate in muscle or form junctions with other neurons. Nerve fibers are generally covered by a fatty sheath called myelin, and in the case of axons lying outside the central nervous system, the myelin is in turn surrounded by a thin membrane known as the neurilemma. The neurilemma participates in regeneration of peripheral axons following injury or transection of the fiber, and since there is an absence of neurilemma in the central nervous system, regeneration does not normally occur there to any significant degree. While there are many structural types of nerve cells and great diversity in the arrangement of their projections, neurons may be functionally divided into three classes. These are sensory (afferent) neurons, motor (efferent) neurons, and associational or internuncial neurons. (Figure 1 provides a diagram of a motor neuron.) Exceptions to this schematization as discussed below are special efferent fibers which are found in sensory systems but do not serve a motor function.

The nerve impulse

The ability of neurons to conduct nerve impulses along their length is largely a function of the cell membrane. When the neuron is in the resting state the cell membrane is said to be polarized—that is, a steady electrochemical potential difference is maintained over the membrane by virtue of its selective ability to prevent sodium ions from passing from extracellular to intracellular space and to permit potassium ions to move inward through the membrane. As a result of this differential, the concentration of ions inside the cell becomes electrically negative with respect to the outside, and the magnitude of this difference in potential is usually between -50 and -90 millivolts.

When an appropriate stimulus is applied to a nerve fiber, the membrane at the point of stimulation suddenly becomes permeable to sodium ions, and these move across the membrane and inside the fiber. This process of membrane depolarization is represented by a rapid reversal of the resting potential from, for example, —70 millivolts to +40 millivolts. This change of about 110 millivolts is referred to as the nerve-impulse, or nerve-action, potential. Immediately following depolarization, the permeability of the membrane to potassium increases, and potassium ions shift from inside to outside the cell causing a reversal of the nerveaction potential and a restoration of the normal resting level (repolarization). The occurrence of the nerve-action potential and the return to the resting potential require from 0.5 millisecond to 2.0 milliseconds. When an action potential is generated in a segment of a fiber, the resulting entry of ions into the cell from outside produces depolarization of the adjacent membrane as well. This marginal depolarization acts as the stimulus to a further breakdown in permeability along the membrane, and the process continues as a chain reaction along the length of the fiber. In this manner the nerve impulse is propagated over the length of the fiber in a fashion not unlike that in which ignition travels along the fuse of a firecracker. The range of velocity of the nerve impulse is very wide, varying from approximately 1 to 300 meters per second, depending upon the diameter of the fiber and other factors. The size of the nerve-action potential is not a function of the strength of the stimulus applied to the nerve. It is true that the

magnitude of the nerve-action potential may vary from fiber to fiber, but within a neuron and under constant conditions it is usually invariant, occurring in full amplitude or not at all, although there are certain exceptions to this rule. Also of importance is the fact that the nerve impulse does not decrease in size as it sweeps along the fiber. This all-or-none principle applies only to the nerveaction, or spike, potential of the axon. There is another kind of potential change which immediately precedes the generation of the nerve impulse whenever a stimulus—physiological or artificial—is applied to a neuron. This is the local excitatory potential, which is graded in amplitude and, thus, does not function in an all-or-none fashion. The size of the excitatory potential is proportional to the magnitude of the stimulus applied and represents a partial depolarization of the cell membrane. It is propagated only in a weak and decremental fashion, if at all. If the stimulus is not very strong, the local potential will decay rapidly and no further neural events will occur. If, however, the stimulus is sufficiently strong, the local potential will increase in amplitude and reach the threshold for triggering the all-or-none nerve-action potential.

The spike potential is followed immediately by a brief period during which the axon is absolutely refractory and will not respond to any stimulus, regardless of magnitude. However, the fiber very soon enters a period of relative refractoriness before returning to the resting level. During the relative refractory period the fiber will respond to stimulation, but only if it is of greater strength than is required by the resting nerve. Although the refractory period lasts only a matter of milliseconds, it is nonetheless clear that the rate of discharge possible in a given fiber is limited by the presence of periods of absolute refractoriness. The exact duration of the refractory period is a function of nerve diameter and structure, and the recovery rate in the most rapidly conducting fibers permits them to discharge nerve impulses at a frequency of approximately one thousand per second.

Development of the central nervous system

On the back, or dorsal, surface of the embryo there lies a sheet of cells called the ectoderm. In early development this layer becomes thickened, and because cells lying toward the edge of this plate have a more rapid growth rate than cells in the middle, a neural groove is formed. As this groove becomes deeper, it eventually closes off on top, forming a neural tube which extends from head to tail of the embryo. The tube then detaches itself from the remainder of the ectodermal plate. It is from this neural tube and a group of detached ectodermal cells lying along its side—called the neural crest—that the whole adult nervous system is formed.

In the developing embryo the brain first becomes apparent as an enlargement at the anterior end of the tube; the remainder of the tube becomes the spinal cord. With further growth the embryonic brain differentiates into three vesicular subdivisions: the forebrain (prosencephalon), midbrain (mesencephalon) and hindbrain (rhombencephalon).

Six weeks after fertilization there is further subdivision of these primary vesicles. The prosencephalon divides into an anterior portion, the telencephalon, or endbrain—which will eventually form the cerebral hemispheres (cerebral cortex, rhinencephalon, and basal ganglia)—and a posterior part, the diencephalon, or “twixtbrain” (thalamus and hypothalamus). The mesencephalon remains relatively small and undifferentiated, but from it will be derived its constituent components: the tegmentum, colliculi, and cerebral peduncles. The rhombencephalon divides into anterior and posterior portions, the metencephalon, or afterbrain (cerebellum and pons), and myelencephalon, or marrowbrain (medulla oblongata), respectively. The long, tail-like remainder of the neural tube will become the adult spinal cord. A schematic representation of these developmental stages of the brain is given in Figure 2.

The cavity which extends throughout the length of the embryonic neural tube forms, in the fully developed fetus, the ventricles of the brain and central canal of the spinal cord, which contain cerebrospinal fluid.

In mammals, one of the most remarkable events during development is the expansion and elaboration of the cerebral cortex. The degree of development of the cerebral mantle is greatly out of proportion to that of the rest of the brain. Its relationship to brain stem is not unlike that of the cap of a mushroom which has grown over, around, and partially down the sides of the stem. In addition, in higher mammals the surface of the cortex is further increased by numerous foldings or convolutions. A groove created by these foldings is referred to as a fissure or sulcus, and the ridge formed between two fissures is called a gyrus.

Fully two-thirds of the cortex in humans lies buried in fissures. The importance of the cerebral mantle in human behavior is to some degree expressed by the fact that it weighs as much as all other central-nervous-system structures combined.

The reflex

A simple reflex arc consists of at least two or three neurons and provides a mechanism whereby a relatively fixed and rapid behavior pattern may occur in response to an appropriate sensory stimulus.

A two-neuron reflex arc comprises a receptor and an afferent neuron bearing information about the stimulus to the central nervous system and a motor neuron with which the sensory cell synapses and through which the response is produced. This simple reflex pathway is referred to as a monosy nap tic reflex arc, since only one synaptic junction is involved. An example is the well-known knee jerk. This reflex may be produced by a tap on the tendon just below the kneecap. This elicits a slight stretching of the muscle fibers and results in stimulation of muscle-spindle receptors. Sensory impulses are conducted into the spinal cord, where motor neurons are excited, their discharge producing quick, phasic contractions of the muscles and a consequent “jerk” of the leg. Monosynaptic reflexes also include arcs in which there are more sustained or tonic contraction of muscle, and these play an important role in the control and maintenance of posture.

In a three-neuron or multineuronal arc there may be one or many neurons interposed between the sensory and motor cells. In general, the greater the number of these interposed links (interneurons) between the sensory input and final motor outflow, the more complex and less stereotyped the reflex.

The sensory impulses which produce a reflex may also be propagated over fibers that reach the cerebral cortex and, thus, result in awareness of the stimulus. However, this sensation is incidental to, and not critical for, the elicitation of the reflex. Furthermore, although a reflex pathway may function without the participation of other spinal or brain areas, this potential for independence is unusual in the normal activity of the organism. Reflexes which involve only one segment of the spinal cord are referred to as segmental reflexes; but as already pointed out, most reflexes include the participation of more than one spinal segment (intersegmental reflexes) and frequently the brain as well (suprasegmental reflexes). Suprasegmental reflexes may be extremely complex, in the sense that organization and coordination of the relatively simple spinal and intersegmental reflexes are carried out by the higher centers. For example, the postural supporting and shifting reactions, which are required for walking and which involve alternating contraction and relaxation of the limbs, are exquisitely integrated by delicately balanced mechanisms in the brain stem. A reflex may be somatic or autonomic, and it may involve either spinal or cranial nerves or a combination of both.

Examples of reflexes, in addition to the postural adjustments mentioned briefly above, are withdrawal of a limb from a painful stimulus before the pain is appreciated as a sensation; the constrictor response of the iris in response to bright light; coughing; sneezing; gagging; vomiting; adjustments in heart rate, blood pressure, and gastrointestinal activity. It is evident from this representative, but incomplete, listing that reflexes are appropriate moment-to-moment adaptations of the organism upon which its very survival depends.

The importance of reflexes for behavior in general may be elaborated and summarized by the following statements: reflexes are innate in the sense that they are the result of genetically transmitted neural mechanisms; they are hierarchically organized so that very simple reflexes can be combined, by the activity of brain centers, into more complex response patterns; while the most simple reflexes within an organism are fixed and stereotyped, complex reflexes tend to be variable and can be modified by experience; finally, phylogenetically, nervous systems have evolved from diffuse and crude reflex mechanisms to more specialized linkages capable of clearly differentiated response patterns. In higher species the borderline between elaborate adaptive reflexes and complex flexible behavior which can be learned is not clear. In man, evolution culminates in the capacity for a high degree of symbolization and the appearance of language and speech. This capacity includes the ability to abstract general principles from specific items of information and memory and the development of an extraordinarily complex social organization accompanied by the ability to transmit it.

From the foregoing, it is evident why the reflexes are commonly considered the basic units of the nervous system and the functional elements from which all other behaviors are derived. We have observed an apparent structural, physiological, and behavioral continuity extending from the simple segmental reflex to complex human behavior. However, whether this continuity between reflex and symbolic capacity is actual or illusory remains today very much an open question, for although there obviously exist common underlying structures and processes, it is not possible to describe or account for a large part of motivated behavior and learning—much less ideation, reasoning, or abstraction—in terms of basic reflex mechanisms.

Spinal cord

A basic function of the spinal cord is the conduction of nerve impulses to and from

Table 1—The cranial nerves

NUMBER

NAME

ORIGIN (OR TERMINATION)

FUNCTIONS

Sensory

motor

I

Olfactory

Telencephalon (ventral rhinencephalon)

Smell

II

Optic

Thalamus

Vision

III

Oculomotor

Midbrain

Ocular muscle proprioception

Eye movement Eyelid elevation Contraction of iris Accommodation of lens

IV

Trochlear

Midbrain

Ocular muscle proprioception

Eye movement

V

Trigeminal

Midbrain and pons

Touch, pain, and temperature sensation from face, jaw, head, ear, cornea, sinuses, teeth, gums, tongue, mucous membranes of mouth, and meninges Proprioception of masticatory muscles

Muscle movement of jaw, middle ear, and palate

VI

Abducens

Pons

Ocular muscle proprioception

Eye movement

VII

Facial

Pons

Taste (anterior two-thirds of tongue)

Muscle movement of face, scalp,

VIII

Vestibulocochlear (a) Vestibular division (b) Cochlear division

Medulla

Equilibrium Hearing

IX

Glossopharyngeal

Medulla

Taste (posterior one-third of tongue)

Swallowing Salivary secretion

X

Vaaus

Medulla

Cutaneous sensation of external ear Sensation from pharynx, larynx, trachea, and esophagus Visceral afferents from thoracic and abdominal cavity linings and organs

Control of activity of heart, gastrointestinal tract, blood vessels, and other visceral organs of thorax and abdomen Control of pharyngeal, laryngeal, and esophageal muscles

XI

Spinal accessory

Medulla and spinal cord

Muscles of pharynx, larynx, and palate Muscles for flexing neck, rotating head, and raising shoulders

XII

Hypoglossal

Medulla

Position sense of tongue

Tongue movements

higher levels of the central nervous system. Except for nerve activity carried over the cranial nerves, all remaining sensory information and motor outflow travels through the spinal cord and the nerves which extend outward from it to the periphery. Spinal functions have often been determined by studying animals in which the cord has been experimentally severed or in humans in whom spinalcord transection has been the result of accidental injury. In man there are 31 pairs of nerves arranged along the length of the spinal cord, which is encased in the bony vertebral column. Each pair of nerves lends to the cord the appearance of external segmentation. These 31 segments have been designated as follows, proceeding from the top (rostral) portion of the cord to the bottom (caudal) end: 8 cervical, 12 thoracic, 5 lumbar, 5 sacral, and 1 coccygeal. If the spinal cord is completely transected above the fourth cervical segment, death occurs immediately, since the nerve fibers originating in the respiratory centers of the lower brain stem are severed from their connection with the muscles of respiration. Humans may survive interruption of the spinal cord at lower levels. If the transection is complete, voluntary movements are permanently lost below the level of the transection. The ability of the isolated cord to maintain postural reflexes and motor activity depends upon the species of organism and the type of reflex under consideration. For example, in lower forms such as amphibians and reptiles, reflex withdrawal of limbs in response to noxious stimuli and reflex swimming movements are retained to a remarkable degree. In some mammals, such as cats and dogs, occasional and usually brief periods of reflex standing may occur if adequate assistance and support are provided. Other basic postural and motor reflexes, such as running movements, may also be present, but these, as well as the standing reflexes, are highly variable.

In man, although the motor neurons of the cord are still connected with skeletal muscle, all reflexes are lost during the period referred to as spinal shock, which often lasts for several weeks after transection. Gradually, some reflexes reappear—first flexion of the limbs, then the reflexes of extension. These soon become exaggerated, since, following transection, reflexes originating in the cord below the cut function without the normal inhibitory influences from higher levels of the central nervous system. Eventually, either flexion or extension spasms will predominate clinically. All sensation, which is normally conducted from the periphery to levels of the cord below the transection, is permanently absent as the result of severance of sensory fibers in the cord ascending to the brain. Autonomic functions—for example, those of the urinary bladder and rectum—are seriously disturbed immediately following interruption of the cord but often show some recovery, indicating the ability of the cord to integrate, at least partially, certain aspects of these reflexes. In summary, it is clear that although certain basic postural, motor, and autonomic reflexes remain or recover after section of the cord, these have only limited autonomy and require the higher centers for their full and adaptive coordination.

Midbrain. The midbrain, like the pons and medulla, contains nuclei of cranial nerves and many fibers in passage, both of which provide communication between higher and lower levels of the central nervous system. In addition, the roof, or dorsal, region of the midbrain is referred to as the tectum and contains two pairs of protuberances, the anterior, or superior, colliculi and the posterior, or inferior, colliculi. These bilateral pairs of elevations are collectively known as the quadrigeminate bodies and are essentially aggregations of gray matter. The remainder of the midbrain is composed of the centrally located tegmentum—which includes the midbrain portion of reticular formation—and large fiber bundles, which for the most part are laterally placed. The quadrigeminate bodies are primarily sensory reflex centers. The superior colliculi receive fibers directly from the retinas of the eyes, and although in higher mammals other pathways and structures mediate complex processes such as pattern perception, the colliculi nonetheless participate with adjacent regions in certain reflexes such as blinking in response to novel visual stimuli, pupillary constriction to light striking the eyes, and certain reflex conjugate eye movements.

The inferior colliculi receive fibers from the auditory pathways which enter the medulla and ascend through the pons to the midbrain. These collicular midbrain stations are concerned with startle reflexes and eye, head, and body-orienting reactions to unexpected noises. An additional level of skeletal muscle control is also contributed by other nuclei within the midbrain.

Thalamus

Sensory information from every modality, with the exception of olfaction, is relayed through thalamic nuclei before being transmitted upward to the telencephalon. Figure 3 shows the location of the thalamus. From these thalamic sensory nuclei, impulses are projected to appropriate and specific receiving areas in the cerebral cortex. In addition, neural information exchanged between the cerebral cortex, on the one hand, and the brainstem reticular formation and cerebellum, on the other, reaches thalamic stations en route. It is important to emphasize that the thalamic relay nuclei, like other synaptic stations in the central nervous system, do not simply send on raw information in an unchanged form from one part of the brain to another; but, in a manner not well understood, modify the neural discharge patterns and their temporal relationships in such a way as to provide simultaneously for the apparently contradictory processes of both increased selectivity, on the one hand, and integration of neural information from several sensory modalities, on the other. The crucial role of the thalamus in functions of the reticular system will be discussed in a later section.

Hypothalamus

The hypothalamus is perhaps best thought of as a major center for the integration of a variety of motivated behaviors, such as feeding, drinking, sexual activities, and emotional responses. It also is of major importance in the regulation of autonomic responses which are not only critical for the moment-to-moment physiological equilibrium of the organism but are also components of the motivated behaviors themselves. Figure 3 shows the location of the hypothalamus.

The functional assembly of these visceral responses and motivated behaviors depends upon the integrity of the hypothalamus. For example, in a series of classical experiments which were designed to determine the central-nervous-system level essential for coordinated emotional behavior, all structures above the hypothalamus were severed or removed surgically in experimental cats. Following recovery from the operation, an intact and wellcoordinated pattern of defensive-aggressive display was still present. When, however, a brain-stem transection was made behind the hypothalamus, the integrated rage response—which in the cat consists of baring of the teeth, hissing, spitting, extrusion of the claws, striking out with the claws, arching of the back, elevation of body hairs, and sometimes urination and defecation—was no longer present in an integrated pattern. Disjointed and isolated elements of emotional behavior in response to tactile stimuli were still occasionally expressed through receptive and motor mechanisms in the brain stem below the hypothalamus, but the coordination of these responses, which makes the behavior protective and adaptive, was permanently lost. It is worth noting that the defensive-aggressive response pattern consists of an assortment of skeletal motor and autonomic reflexes which are exquisitely combined and organized by the hypothalamus into a precisely timed sequence of actions. It should be pointed out, however, that although the hypothalamus is critical for the integration of this and other behaviors, it too responds to higher influences from the basal ganglia and cortex of the cerebral hemispheres.

Metabolic and endocrine response. In addition to sympathetic and parasympathetic regulating areas in the hypothalamus, there are reciprocal control areas for other behaviors. For example, the ventromedial nucleus of the hypothalamus is an inhibitory center for eating behavior, while a more lateral region is concerned with excitation of this behavior. Evidence of the existence of critical areas such as these often comes from experimental studies in which small regions of the brain are either destroyed or electrically stimulated in experimental animals. For example, in the case of the ventromedial nucleus, stimulation produces immediate cessation of eating for the duration of the stimulus, and destruction of this nucleus results in overeating and, eventually, obesity. The overeating following this lesion presumably occurs because the excitatory center in the lateral hypothalamus is released from the inhibitory control of the ventromedial nucleus. Conversely, stimulation of the lateral hypothalamus elicits the onset of eating, and a lesion in this area leads to starvation.

As part of its controlling role in the expression or inhibition of eating, drinking, sexual behavior, and additional metabolic functions, the hypothalamus exerts an important influence over the major endocrine gland, the pituitary, which lies at the base of the brain and is connected with the hypothalamus by a stalk containing neural fibers which extend from the hypothalamus down into the posterior lobe of the gland (see Figure 3). The anterior lobe of the pituitary is influenced by the hypothalamus through a vascular system, so that chemical substances released by secretory cells in the hypothalamus may act upon the anterior pituitary. The pituitary, in turn, plays a central role in the control of the several other endocrine glands with which it has reciprocal relations. When hormones of these glands enter the general circulation, they act on not only the pituitary itself but the hypothalamus as well, to modify its activity. Thus, an equilibrium of bodily functions concerned with metabolism and consummatory behavior is maintained by a circular system of neural and hormonal checks and balances, with the hypothalamus and pituitary performing the central integrating steps in these processes.

The cerebral cortex

The cerebral cortex is a convoluted, or folded, mantle of gray matter arranged as layers of cells over the surface of the cerebral hemispheres. It virtually covers and surrounds the upper levels of the brain stem. The cortex has an average thickness of about 2.5 millimeters, but this varies considerably from one region to another. Fibers leading to and from subcortical areas and connecting cortical areas with each other make up the white matter within the hemispheres. The corpus callosum, a massive bundle of fibers, crosses the midline of the hemispheres and provides a direct connection between the right and left cerebral cortices.

The cortex represents the most recent and highest development of the nervous system. In man it is proportionally larger than in any other species, and to the cortex are attributed those activities which are most distinctly human: complex learning and reasoning; symbolization, abstraction, and generalization; and the development of language and speech. While a substantial portion of the cortex is given over to sensory receiving areas and regions concerned with the initiation and control of motor responses, the higher integrated activities just mentioned depend upon other cortical regions —the so-called cortical association areas—as well;these regions are critical in the mediation of complex perceptions and cognitions.

At the gross anatomical level, the cortex is divided into lobes. In addition, areas which are presumably different in their cellular architecture have been traditionally assigned numbers (Brodmann’s areas). In some instances Brodmann’s areas correspond to functional zones, but in other cases they do not. The relationship of the lobes and functional areas is shown in figures 4 and 5.

Sensory cortex

The anterior portion of the parietal lobe immediately behind the central fissure (Figure 5) is called the postcentral gyrus and contains the primary somatosensory receiving region (areas 3, 1, and 2). These areas receive sensory information directly from the posterioventral nucleus of the thalamus and, thus, are important in the initial perception, discrimination, and localization of stimuli exciting nerve endings and receptors on the surface of the body and from within its somatic musculature. However, a primitive consciousness of certain somatic sensations probably exists in man at the thalamic level, even in the absence of cortex. These thalamic sensations are at best poorly differentiated and not well localized. For example, a patient with a lesion in the primary somatosensory receiving area of the cortex may indicate that he appreciates the fact that stimulation has occurred but may have great difficulty in distinguishing degrees of intensity of stimulation or in localizing the stimulated point on the body

Fibers carrying the somatic sensations of touch, pressure, pain, proprioception, and warmth and cold are not well separated in either the thalamus or cortex, but the cortical receiving area is topographically organized in a manner similar to that of the motor cortex: lower extremity representation at the uppermost portion of the sensory area, with upper extremities and head represented at the basal portion of the region. Our knowledge of topographical precision within these sensory areas is primarily the result of electrophysiological mapping studies in which punctate physical stimuli are applied to the surface of the body and the resultant electrical potential changes, called evoked sensory potentials, are recorded with small electrodes on the surface of the cortex. A distinct and localized evoked potential occurs at the specific cortical area which represents the part of the body being stimulated.

Although the reader may have gained the impression from the foregoing that the central fissure forms an absolute boundary between motor and sensory cortex, this is not actually the case. Motor responses can be elicited from the sensory areas, and sensations have a degree of representation anterior to the central fissure. However, the responses become fewer and weaker as we progress from motor into sensory cortex or vice versa. It also is appropriate to point out at this juncture that taste, in addition to the somatosensory modalities mentioned above, is also represented in the parietal lobe at the base of the postcentral gyrus. [SeeTaste AND Smell.]

Considered from the phylogenetic point of view, many sensory and motor functions have undergone a process known as corticalization. That is, functions which in lower forms appear to be mediated subcortically depend in higher species on the integrity of the cortex. The instance cited, showing visual losses to be greater in man than in lower animals following destruction of the striate cortex, is an example of corticalization. In the motor sphere this evolutionary process is equally evident. For example, in the rat the motor cortex is not well organized either cytoarchitectonically or physiologically. If it is completely removed, it is difficult for even the sophisticated observer to discern any motor impairments. In the cat or dog, where there is greater organization and finer differentiation of motor areas, motor weakness and a substantial loss of movement follow injury to this region. The recovery, however, is usually prompt although not necessarily complete. In primates—especially man—the reliance upon cortex for normal activity is even more crucial, and motor-cortex injury produces greater and more enduring paralysis than in any of the species described above.

Frontal lobe. The prefrontal region, or association cortex of the frontal lobe, lies rostral to those areas from which motor effects can be elicited. Neurons of the prefrontal region make no fiber contribution to the pyramidal motor pathways, nor do they receive impulses directly from thalamic sensory-relay nuclei. They do, however, have connections with other subcortical structures, including the thalamus, hypothalamus, corpus striatum, midbrain, lower brain stem, and cerebellum, as well as with many cortical areas.

In monkeys, experimental injury to portions of the prefrontal cortex may result in a number of behavioral alterations, including hyper activity and stereotyped pacing, increased distractability and inattention, response perseveration, and possibly disruption of short-term memory. A combination of these disturbances may account for the difficulty that monkeys with lesions of the prefrontal cortex have in solving problems which require sequential responding or in holding and retrieving information over a delay period between stimulus presentation and response, during which delay the stimulus object is not visible to the animal. As is generally the case with injury to the association cortex in monkeys and lower species, symmetrical bilateral lesions produce much greater defects than do unilateral ones. In fact, in some instances unilateral lesions may result in no observable loss of function. Another consequence of prefrontal damage is in the sphere of emotional behavior. Normal monkeys and chimpanzees working on very difficult discrimination problems often display emotional disturbances, including violent temper tantrums, as a manifestation of the frustration resulting from the many errors made. After prefrontal lobotomy, a surgical procedure which severs the connections between the prefrontal lobe and subcortical areas, the animal no longer becomes upset with its errors and works calmly and quickly on the problem, even though the lobotomy may have actually increased the frequency of errors.

These observations on emotional behavior led to the development of the prefrontal lobotomy in man for the relief of neurotic and psychotic symptoms. This operation has been almost entirely replaced in recent years by the verbal and experiential therapies, electroshock treatment, and the psychotropic drugs. However, variants of the lobotomy procedure are still occasionally employed in the alleviation of intractable pain in terminal diseases. The behavioral effects of lobotomy in man are highly variable, and reports have been frequently contradictory. There is no clear evidence that prefrontal lobotomy significantly alters the intelligence of man, and it is possible that the defects which have been shown by some studies may actually reflect motivational and attitudinal, rather than intellectual, changes.

Hemispheric dominance

The fact that the left and right sides of the brain appear grossly identical does not mean that there is necessarily equivalence of function for the two hemispheres. For example, damage to the right parieto-occipital cortex of man often produces impairment of performance of nonverbal perceptual motor tasks, while similar injuries to the left hemisphere may be without consequence with respect to this class of behavior. For many complex human activities there is a dominant hemisphere, but for a particular individual the dominant hemisphere is not necessarily the same for all such activities. Because the pyramidal motor pathways cross as they descend, individuals who are unequivocally right-handed are cerebrally dominant in the left hemisphere for this function. In fact, most people, whether righthanded or left-handed, show left cerebral dominance for speech, but not all do; for example, individuals who are dominant for speech in the right hemisphere are more likely to be left-handed. Thus, for a particular individual the hemisphere which is dominant for speech may or may not be the one that is dominant for handedness.

If the dominant hemisphere is injured, there is a tendency for the homologous cortical area of the opposite hemisphere to assume its function. Thus, if the area of the left hemisphere which controls speech is severely injured, this function will be lost, but it may be recovered if comparable regions of the right hemisphere can take over. Whether recovery of an activity occurs in a particular case is largely a function of the age at which the damage occurs. Plasticity is especially pronounced in early life and declines with maturity and age. Recent evidence indicates that there are persons in whom speech is represented bilaterally in the cortex. In these cases, it is not yet clear whether both hemispheres have come into use as a result of injury to the one which was originally dominant.

Equipotentiality, mass action, localization

Experiments with monkeys have shown that even lesions which destroy both prefrontal lobes will not produce the prefrontal-lobotomy syndrome described earlier if the operation is done very early in life. This illustrates that the central-nervoussystem plasticity involved in the restitution of certain abilities does not necessarily require the presence of an anatomically intact and comparable region in the opposite cortex. Other, as yet undetermined, structures of the hemispheres are sometimes able to assume the role of regions which are morphologically quite different. Many findings such as these have led to the concept of functional equipotentiality of cortical areas, which is almost certainly an oversimplification. But if the limits of plasticity are taken into account, the “equipotentiality” notion has merit, especially in lower mammals, in which structural and functional differentiation is not so prominent and recovery is more frequent and rapid.

For many years, in neurology and psychology, a lively debate has raged over the degree to which behavioral functions are localized in specific areas of the hemispheres. At one extreme, some scientists have taken the view that functions and subfunctions are highly localizable. Proponents of strict localization have created detailed maps of the cortex relating precise cortical regions to a myriad of specific complex psychological activities. At the other end of the spectrum are those who believe that except for the primary sensory and motor areas, the cortex functions as a whole in practically all psychological activities and that localization is a rare finding. Evidence for this “mass action” viewpoint is often taken from rat studies in which large amounts of several cortical areas had to be removed before a deficit in learning or memory was noted, as well as from the fact that in humans certain psychological-test performances are disturbed regardless of the location of the lesion. The concepts of mass action and equipotentiality are obviously closely related. As has often been the case in similar scientific controversies, the evidence of both sides is frequently valid and not so mutually exclusive as the proponents of the extremes would sometimes have us believe.

Recent studies on human cortical injuries and electrical stimulation of the brains of patients indicate that some psychological functions are rather well localized, although the precise region on the cortex may vary somewhat from individual to individual. For other classes of behavior, specific regions have not been found, and it is reasonable to assume that certain complex intellectual activities require roughly equivalent participation of neuronal circuits from many areas of the cortex.

Rhinencephalon, limbic system, and reticular formation

”;Rhinencephalon” literally means “nose-brain”it is basically composed of those structures concerned with olfaction. These are the olfactory bulbs and pathways, the amygdaloid nuclei, the threelayered cortex (paleocortex) on the ventral and mesial surface of the temporal lobe, and the hippocampus (archicortex), which is folded under the neocortex, which comprises the remainder of the covering of the hemispheres and has been discussed in the previous section. The paleocortex and archicortex are relatively primitive structures in terms of point of appearance in evolution and degree of morphological differentiation. [See Taste AND Smell.]

Although the involvement of rhinencephalic structures in olfaction has been assumed and studied for many years, it was not until the past few decades that their role in emotional and visceral activity was also recognized. Out of such investigations of structures concerned with emotion has grown the concept of a limbic system. A rather loose term, “limbic system” is often defined somewhat differently by various investigators. It arose from the even older term “limbic lobe,” which Broca used to designate the cingulate cortex in addition to most of the rhinencephalon. Broca assigned no specific function to this series of interconnected structures, which form a ring, or “limbus,” around the brain stem, but noted its presence as a common denominator of the mammalian and vertebrate brain. The modern concept of the limbic system is more inclusive than either that of rhinencephalon or limbic lobe. The limbic system is now generally taken to include many structures, from brain stem to cortex, that are connected anatomically and are involved in the expression of affect and other motivated behaviors of which emotion is a part. Some of these structures are the orbitofrontal cortex, cingulate cortex, hippocampus, ventral paleocortex, septal nuclei, amygdala, several hypothalamic and thalamic nuclei, and certain midbrain areas. The limbic system also has strong connections with the reticular formation.

It should not be assumed that structures of the limbic system participate only in states of emotional experience or expression. Many components of the limbic system are also active in the control of autonomic activities, which, according to the particular instance, may or may not be involved in affect—for example, digestion, cardiac and vascular regulation, and respiration. In addition, one of the central structures of the limbic system, the hippocampus, plays a crucial role in the process of memory storage. Before the importance of the hippocampus in recent memory was appreciated, a few patients underwent bilateral removal of this structure in an attempt to relieve severe seizures or psychotic behavior. Following these operations, the patients usually had a good memory of events that had occurred up to a few months or weeks before the surgery but virtually no memory of events occurring after that time. Some of the patients have been followed carefully for several years and continue to show a profound inability to remember events which have occurred since the operation.

It is clear from these unhappy clinical observations that the hippocampus is vital in recording perceptual impressions and laying down the memory trace. It is also evident that this structure is not crucial for the retention, retrieval, or expression of stored material that has been well established, since memory of events long past was adequate in these patients. It is presently hypothesized that the human hippocampus acts as the first-stage recorder of experience but that transfer of accumulated information to other systems is soon effected for long-term storage.