Neural Bases Of Behavior
Nervous tissue consists essentially of two types of cells: neurons, concerned with the conduction of nerve impulses; and neuroglia, or glial cells. The neuroglia are specialized cells which support the neurons. They probably have other functions as well, such as the transport of nutrients to the neurons and the transport of the products of metabolism away from them.
A neuron consists of a cell body, where most of its metabolic activities are carried out, and processes or fibers. The latter are of two types: dendrites, which are usually short and carry impulses toward the cell body; and axons, which are usually longer and carry impulses away from the cell body. The anatomic and functional region at which a nerve impulse passes from the axon of one neuron to the dendrite or cell body of another is termed a synapse. Bundles of nerve fibers travelling together within the central nervous system (brain and spinal cord) are termed tracts. Those travelling in the peripheral nervous system, i.e., outside the brain and spinal cord, are organized into nerves which may be regarded as anatomic tables composed of nerve fibers, blood vessels and lymphatics, and connective tissue elements. Most of the fibers of the nervous system are enveloped in a sheath of fatty, white material called myelin. Myelinated nerve fibers give much of the brain and spinal cord its characteristic white appearance on cross section (white matter). The outer 2 to 6 mm. of the cerebral hemispheres are gray in color owing to the high density of nerve cell bodies and unmyelinated nerve fibers. This is the cerebral cortex. Islands of gray matter are found also throughout the brain. These represent aggregates of nerve cell bodies and are called nuclei. They are among the important anatomic structures to be discussed in this chapter. The central portion of the spinal cord is also gray, consisting largely of cell bodies and synaptic foci.
Traditionally, neurons are classified on a functional basis into afferent, efferent, and internuncial. Afferent neurons are regarded as sensory, conducting impulses toward the central nervous system. Efferent neurons are largely motor and conduct impulses away from the central nervous system. All other neurons are termed internuncial, conducting impulses between other neurons within the central nervous system.
The cranial and spinal nerves and their associated ganglia, bulbous areas where nerve cell bodies are aggregated, constitute the peripheral nervous system. The afferent or sensory fibers of this system respond to stimulation of receptor end-organs sub-serving the various sense modalities and conduct impulses to the central nervous system. The efferent or motor nerve fibers are of two types: regular motor and autonomic. The regular motor fibers terminate in skeletal muscle which is potentially under conscious control but which may react reflexly as in the knee jerk elicited by tapping the tendons below the patella. The autonomic fibers terminate in the cardiac muscle, smooth muscles, and glands of the body. These are not under direct conscious control.
The Central Nervous System
For embryologic and functional reasons, the mammalian brain is divided into three portions: the rhombencephalon (hind-brain), mesencephalon (midbrain), and prosencephalon (forebrain). These, with the spinal cord, constitute the neuraxis. Going from the forebrain down is termed the caudal direction, and from the cord up, the rostral direction.
The hind-brain includes the medulla oblongata, pons, and cerebellum (Fig. 7). The medulla or bulb contains all the tracts which course between the spinal cord and higher centers. In addition, the nuclei of several cranial nerves are situated here, as are important centers for the regulation of breathing, circulation, and other visceral functions. The pons contains tracts which course between the cerebrum above, medulla below, and the cerebellum behind. In addition, the nuclei of some cranial nerves are located here. The cerebellum is concerned chiefly with the coordination of muscle groups making possible smooth and skillful movements of body parts.
The midbrain is a short segment between the pons and the forebrain. It contains nuclei of many nerves and many vital areas of the integration of nervous activity. The hind-brain and midbrain contain an important central core of fibers and scattered cells known as the reticular formation. The functional significance of this area is described later.
The caudal portion of the forebrain is termed the diencephalon. In addition to other structures, it contains the thalamus and hypothalamus, areas containing many aggregates of gray matter (nuclei) concerned with the integration of nervous and endocrine activity and "relay stations" for nervous pathways communicating between lower centers and the cerebral hemispheres.
The rostral portion of the forebrain, or telencephalon, the most recent part of the brain to be studied phylogenetically and anatomically, dominates the rest of the brain in primates, and especially man. It is organized into two great hemispheres which communicate extensively with each other through several bundles or commissural fibers. The largest of these is called the corpus callosum (Fig. 8).
The surface of the cerebral hemispheres contains convolutions or gyri separated by grooves termed sulci. Deep sulci are also called fissures (Fig. 7). The latter are used to divide the hemispheres into lobes. The frontal lobe is the anterior one-third of the hemisphere which is rostral to the central sulcus and above the lateral fissure. The occipital lobe lies behind an arbitrary line drawn from the parieto-occipital fissure to the pre-occipital notch (Fig. 7). The remaining portion of the brain is divided into parietal and temporal lobes, lying above and below a line drawn between the end of the lateral fissure and midpoint of the line demarcating the occipital lobe. These sub Figure 8 shows the medial surface of the hemispheres as seen after sectioning the brain in the midline. Some of the surface structures shown and deep structures depicted by dotted lines are important in the limbic system to be discussed later.
The correlation of behavioral deficits with pathologic changes in the brain has probably been the richest source of information about the role of the nervous system in behavior. Traumatic head injuries incurred during the two World Wars have been clinically observed. Other pathologic processes, such as tumors, The correlation of brain damage with behavioral deficit is no simple task. Clinical lesions tend not to be delimited to a single anatomic structure. Hence, deficits observed are related to the interaction of the effects of varying degrees of damage to a variety of structures. Further, the psychologic changes accompanying a particular lesion are dependent on a great many factors in addition to the location and extent of the lesion. The factors include the following: The Nature of the Lesion. For example, a benign tumor may produce symptoms due to simple mechanical irritation of surrounding normal tissue. An invasive, malignant tumor in the same location may rapidly destroy surrounding normal tissue and produce different behavioral deficits and clinical phenomena.
The Rate of Growth of the Lesion. The alterations in brain functioning resulting from a destructive lesion are confounded by the response of other cerebral mechanisms in reaction to the primary changes. The nature and extent of the secondary reactions and the psychologic and physical symptoms which result are a function, among other things, of the rate of development of the primary lesion.
The Age of the Individual. These differences are related to the varying degrees of maturation of the nervous system at different ages as well as differences in psychologic defense mechanisms which characterize different ages. Further, there is a general principle that the plasticity and adaptability of the brain decreases as age advances. This is seen, for example, in aphasia. Recovery of language loss from injury to the left hemisphere is much greater, both in rate and final level of recovery, in a child of seven than a man in his sixties. Apparently the normal, nondamaged brain tissue which surrounds the injured area gradually loses its ability to take over functions of diseased tissue as age advances.
Age differences are seen also in lesions involving deeper brain structures. For example, the late sequelae of the viral encephalitis which was epidemic following the First World War varied greatly with the age of the individual at the time of the initial infection. Many children who contracted the infection later developed primarily psychiatric symptoms, often within a few months of the initial infection. These consisted chiefly of conduct disorders. Many of the children were hyperactive and showed little tolerance for frustration, tending to react impulsively with little provocation in a destructive and socially delinquent manner. In contrast, persons who were already adult at the time of the infection tended to develop later on primarily neurologic symptoms such as tremor and muscular rigidity of Parkinson's disease.
Premorbid Personality. The behavioral consequences of brain damage are the result of alterations in function directly ascribable to the neural tissue damage and the reaction of the rest of the personality to these changes. The latter is greatly dependent upon the characteristic ways in which the person handles stress and the various personality traits and tendencies which existed before the injury. This is seen clearly, for example, in some tumors of the frontal lobes which, in their early stages, characteristically present purely psychologic symptoms. This is true especially in tumors arising from the olfactory groove (28). The psychologic reactions which develop are often an exaggeration of pre-existing personality traits and tendencies. A person who characteristically reacts to social and personal stress with suspiciousness may develop paranoid ideas, one with a despondent disposition may become depressed, and so forth.
Although it is particularly important to be mindful of these various factors which influence the consequences of clinical brain lesions, they are important also in determining the effects of experimental lesions and the various experimental manipulations used to elucidate problems of brain function. A description of these manipulations follows.
Much has been learned about the functional organization of the nervous system by observing the effects of transection of the neuraxis at various levels. Mammals such as cats, dogs, and monkeys have been used extensively for this purpose and their study contributed greatly to classical neurophysiology. Descriptions of several are listed below: Spinal Preparation. The spinal animal is prepared by a transection between the spinal cord and medulla. All complex voluntary and visceral behavior is lost and the animal must be kept alive by artificial respiration. This preparation has been used chiefly for studying those reflexes which are completed within the spinal cord itself.
Decerebrate Preparation. This animal may be prepared by interrupting the blood supply at the level of the common carotids and basilar artery at the center of the pons, thus depriving the upper midbrain and forebrain of its blood supply. This animal shows much complex voluntary and reflex behavior but many deficits are apparent. In the words of Maclean (35), "It resembles nothing so much as an idling mechanism temporarily devoid of its driver." A decerebrate cat tends to remain in the position it is placed for a long time, shows no grooming or food searching behavior, and no signs of pain or pleasure. It may demonstrate angry behavior if provoked with a noxious stimulus, but the behavior stops abruptly when the provoking stimulus is withdrawn. Such behavior has been termed "sham rage" or "pseudo-affective" because it is undirected, unsustained, and by inference assumed to be unassociated with the subjective feeling of anger. Phenomena of this sort are discussed more fully under the limbic system.
Hypothalamic Preparation. This is similar to the decerebrate animal except that the upper midbrain and hypothalamus are intact. Such animals are grossly defective with impairment of memory function, their behavior little affected by immediate past experience. Hypothalamic animals tend to react with excessive emotion to all stimuli, be they pleasurable or noxious. The rage of the hypothalamic cat is more integrated and convincing than that of his decerebrate brother. As will be described below, such observations have led to the view that the hypothalamus is much concerned with the integration of emotional behavior.
Decorticate Preparation. The most striking difference in brain morphology between man and lower mammals is the great development of the cerebral hemispheres in the former. This makes particularly hazardous the application to man of inferences drawn from the study of lower mammals deprived of their cerebral cortex. In fact, decorticate cats and dogs carry out very complex postural, food seeking, and other behaviors with little difficulty. In contrast, cortical lesions in monkeys and apes produce marked deficits in problem-solving and other complex behavior. In man, gross behavioral deficits accompany extensive cortical damage, as results from interference with blood supply to the hemispheres. The experimental elucidation of the functions of various cortical areas has been carried out extensively with primates using more discrete lesions.
Cerveau Isole and Encephale Isole. In the preparations described above, attention is focused on the portion of the nervous system caudal to the transection. In the present instance, however, the primary interest is in the activity of the isolated neural tissue rostral to the transection. In cerveau isole, the transection is at the level of the midbrain and in encephale isole transection is at the junction of the spinal cord and medulla. Study of these preparations has added, for example, to our understanding of the neural basis of arousal and wakefulness. Encephale isole animals have brain wave patterns characteristic of the waking state, whereas cerveau isole animals show a pattern characteristic of sleep. The relationship of these observations to an important midbrain structure called the reticular formation is discussed later in this chapter.
In contrast to these classical neurophysiologic preparations, the work of recent years has focused on the ablation of discrete brain areas, especially in the cerebral cortex and certain sub-cortical nuclei of monkeys and apes. These refined neurosurgical technics, and the development of reliable technics for a detailed experimental analysis of behavioral deficits, has added greatly to our knowledge of brain organization and function. The use of these technics will be illustrated in the discussion of neocortical function and limbic system.
Spontaneous Electrical Activity Of The Brain
In 1875 an Englishman named Caton reported spontaneous electrical activity from the brains of monkeys and rabbits. He observed further that the amplitude of the potentials recorded over the visual cortex was increased when light fell upon the retina of the eye. Thus an association was established between a level of physiologic activity (intensity of photic stimulation) and the level of electrical activity of a brain region (the cortical areas subserving vision). However, it was not until 1929 that Hans Berger recorded similar electrical potentials in men by electrodes placed on the intact scalp (12). Since these observations, electroencephalograms or EEG's, electrical activity of the brain as recorded at the scalp, have been the object of a great deal of study.
Since alterations in cerebral activity are frequently reflected in changes in the EEG, they have proved valuable for the identification and localization of various kinds of brain pathology such as tumors, infections, and abnormal spikes which precipitate epileptic convulsions. In addition, evidence has been accruing that some emotional disorders usually thought of as psychologic in origin may have characteristic alterations of the EEG in some cases. For example, such abnormal electrical activity has been described in certain behavior disorders in children (56).
In addition to its usefulness in diagnosis, electroencephalography is a potent research tool for explaining the biologic substrate of psychologic states and processes. For example, characteristic changes occur in the EEG with changes in consciousness, as in sleep (7) and hypnosis (29). Alterations in EEG occur also with simple change of "attention" or "mental set" (18). In this instance, we have an observable alteration in cerebral physiology, as reflected in the EEG, accompanying an event we usually think of as "psychological" or "mental." That is, a neurophysiologic correlate of a psychologic process (3).
The amount of information obtainable from recording the electrical activity of the brain is greatly increased when the electrodes are placed under the scalp and skull directly onto the brain surface (electrocorticogram) or into the brain substance itself (depth electrodes). By these technics the electrical activity of particular brain structures can be selectively monitored. Recent advances in technic have made it possible to implant a great number of small electrodes into the brains of animals where they can remain for long periods of time (chronic preparations). Figure 10 shows an animal with 610 implanted electrodes which invade different parts of the brain and from which recordings can be made singly or in combinations.
Since 1950 Dr. R. G. Heath and his associates, working at Tulane Medical School, have placed electrodes deep into the brains of human patients, largely for diagnostic purposes (15). Characteristic changes have been recorded from various brain structures, e.g., hypothalamus, cingulate gyrus, and hippocampus, in a variety of neurologic and psychiatric conditions including various forms of epilepsy and schizophrenia (16). Correlations have been established also between specific abnormal mental states, such as hallucinations, and the electrical activity of specific brain structures (51). By these and related technics the relationship of specific brain structures to psychologic processes, both normal and abnormal, is being elucidated.
Another point of contact between physiologic events in the brain and phenomena we regard as psychologic is the identification of specific alterations of brain electrical activity accompanying classical conditioning. Much of this work has been done by Russian investigators (1) who have long sought the neurophysiologic basis of learning. Recently several American investigators have described electrophysiologic correlates of conditions at the cortex, some subcortical nuclei, as well as midbrain reticular formation (11, 21, 38).
Evoked Electrical Potentials
The neurophysiologic changes produced in an experimental animal in response to a stimulus are often detectable as changes in electrical potential in different brain areas. Such evoked potentials give clues to the functions and interrelationships of various brain structures. For example, it has been known for a long time that the lateral geniculate body, one of the nuclear masses which constitute the thalamus, forms part of the nervous pathway for vision. That is, it is one of the way stations in the conduction of nerve impulses which arise in the retina in response to photic stimulation and end in the visual cortex of the occipital lobe. Thus, if the retina is stimulated by light, electrical spikes (short-lived changes in electrical potential or relatively high amplitude) may be recorded from microelectrodes implanted in the lateral geniculate body. Some quantitative relationships have been described between the physical parameters of stimuli presented to animals and the characteristics of the evoked electrical changes in the brain. De Valois and his associates (8), for example, recorded spikes in response to photic stimulation from single neurons in the lateral geniculate body of a monkey. They found that the rate of electrical spikes evoked is proportional to the logarithm of the intensity of illumination.
The technic of evoked potentials has been used also to identify neurophysiologic differences among psychiatric disorders. For example, Shagass and Schwartz (53) report that the electrical potentials recorded with scalp electrodes by stimulation of the ulnar nerve at the wrist have different characteristics in patients afflicted with different psychiatric disorders.
Since the experimental manipulation of animals evokes electrical changes in different brain structures, it might be supposed that the electrical stimulation of brain structures would in turn bring about changes of a behavioral or experiential nature. This is indeed the case, and a great number of structures have been stimulated in a variety of species including man. For example, stimulating specific cortical areas, chiefly in the precentral gyrus, elicits specific motoric responses involving one or another muscle groups. In this manner "motor maps" have been made of the cortex. Similarly, "sensory maps" have been constructed by stimulating other cortical areas, or the postcentral gyrus.
The participation of cortical areas in more complex aspects of behavior, such as problem-solving, also has been studied with this technic. For example, using a food reward, a monkey may be taught to solve a delayed-alternation task, a problem which involves a time lag. The animal is presented with two covered food wells. On successive trials he is required to lift the cover that he had not lifted on the preceding trial. Electrical stimulation of portions of the frontal lobes of the monkey will interfere with his ability to perform adequately this task. However, there will be no interference with a similar task involving a visual discrimination without a time lag (57). In contrast, stimulation of portions of the occipital cortex will disrupt performance on the visual-discrimination task but not interfere with delayed-alternation performance. Similar effects may be obtained by ablating the various cortical areas. However, stimulating technics are in general superior to surgical ablations for the analysis of cortical functions since the neighboring tissue is not destroyed and the effects are reversible; that is, functions return to normal when the stimulation is discontinued.
The recent development of neurosurgical procedures in human patients has provided an abundance of observations of cortical organization and function. For technical reasons, these procedures are frequently carried out under local anesthesia, making it possible for the wake patient to report the experiences that accompany the stimulation of various cortical areas. Wilder Penfield and his associates in Montreal have pioneered for many decades in neurosurgery of this type (44, 46). Most of these observations have been made on individuals afflicted with "psychomotor seizures," a form of epilepsy in which seizures consist of complex patterns of semi-purposeful behavior, such as pacing around a room, or complex sensory experiences such as auditory, visual, or olfactory hallucinations. Electrical stimulation of appropriate temporal lobe areas, sometimes no more than a millimeter in diameter, may elicit the memory or experience which characterizes the patient's usual seizure. Surgical extirpation of these foci sometimes leads to cure of the illness. In the course of these procedures Penfield and his associates have identified other cortical areas concerned with motor functions, sensation, and speech (44, 45, 46).
An alternative to electrical stimulation of the brain has recently been developed by producing a lesion on the cortex which becomes the focus of abnormal electrical activity. These epileptogenic foci may interfere with the normal functioning of nearby cortex, or, if they spread sufficiently, produce a generalized convulsion. Stamm, Pribram, and their associates (54, 55) have produced such foci of abnormal electrical activity by placing aluminum hydroxide jelly on the cortex. They report, for example, that focal paroxysmal discharges created in the inferotemporal cortex of monkeys interferes with their ability to "learn" new tasks involving visual discriminations. However, there is no loss of ability to perform tasks involving visual discriminations which were learned prior to the induction of the abnormal focus. In other words, the technic makes it possible to differentially interfere with the learning of a new task while leaving intact memory for the solution of an old problem. Similar abnormal discharges created in the frontal lobes interfere with the learning of tasks involving delayed-alternation, as described above in electrical New and very promising technics for investigating brain functions, and especially those relating to "emotions" and "motivations," have developed from observations of Olds and Milner in 1953 (40). They found that a rat will persist in pressing a lever in a Skinner box if, as a consequence of the lever pressing, small electric shocks are delivered by tiny electrodes to certain areas deep in the substance of the brain. Such areas have come to be known as "reward" or "pleasure" centers. They are better termed areas of positive reinforcement. That is, areas, stimulation of which tends to increase the frequency of behavior which has the effect of continuing the stimulation (see Section III). The implications of this observation for psychologic concepts such as motivation and drive is obvious, and a great deal of psycho-biologic research has been stimulated as a result. Many brain areas have been found which have this property, and areas of positive reinforcement have been mapped in the brains of a variety of species ranging from the goldfish (2) to man. The reinforcing properties of some of these areas, particularly those in the brain stem and hypothalamus, are found to be very great. Animals sometimes press a self-stimulating lever 6,000 or 7,000 times per hour without apparent habituation or fatigue. Another index of the rewarding properties of these centers is the willingness of hungry and thirsty animals to continue in this behavior when alternative sources of reinforcement, such as food and water, are also available. Animals also willingly cross over electrically charged metal rods which deliver a painful shock in order to gain access to the lever used to stimulate areas of their own brains.
Other brain areas have been described which have opposite properties. Animals will persist in pressing a lever which avoids or terminates electrical stimulation to these areas of negative reinforcement. The opportunity for making similar observations on human subjects is provided by the use of depth electrodes in certain patients with epilepsy, mental illness, and brain tumors. Areas of positive reinforcement have been described in many subcortical regions, such as the septal region and amygdala (Fig. 8). The subjective reports of subjects stimulated in these various areas differ widely. Descriptions range from intense feelings of well being which cannot be identified with any past sensation or experience to specific pleasurable sensations including sexual arousal (30). Stimulation of other areas in humans, often in close juxtaposition to areas of positive reinforcement, produce disagreeable or painful sensations (30). Still other areas, such as most of the neocortex, are neutral with respect to reinforcement properties.
The sources of information about brain function described above have added greatly to our knowledge of the neural basis of behavior. Some of this knowledge, most quite recent, will be summarized under three interrelated systems: the neocortex, limbic system, and reticular activating system.