xml version="1.0" encoding="UTF-8"?> Psychosurgery: First Chapter

Chapter I

Normal Frontal Lobes

Ideas concerning the functions of the frontal lobes have been in process of modification over many centuries. Certainly the most imaginative were those of the ancient Greek poets who conceived of Athene, the goddess of wisdom, as springing full-panoplied from the brow of Zeus. To the psychologists, the frontal lobes are concerned with adaptive responses. The less frontal lobe the more stereotyped is the response, the greater the development of the frontal lobe the more flexible and varied is the response to incoming stimuli. Only the human individual can respond to such exalted concepts as patriotism, the hereafter and God.

When the frontal lobes are completely removed from an animal, it shows marked loss of adaptability in competition with its unoperated control. It is true that the animal can perform under various test situations, but these situations, such as maze running, are of much greater interest to the experimenter than they are to the animal. A rat without frontal lobes may learn to jump for a circle instead of a square when the shock comes, but it is at a distinct disadvantage in the fight for survival among fellow rats on the town dump.

The frontal lobes increase progressively in size throughout the vertebrate phylum and particularly in the primates. Tilney has also called attention to the enlargement of these areas in the ascent of man. Modern man has far outstripped his remote ancestors and his anthropoid cousins. Tilney conceives of the frontal region as the "accumulator of experience, the director of behavior and the instigator of progress."

In order to appreciate the significance of the frontal lobes, it is necessary to start with some description of the gross and microscopic anatomy and to take into consideration the findings of comparative anatomy, of embryology and of experiments on animals. The frontal lobes are bounded on the lateral aspect by the central fissure and the sylvian fissure. On the medial aspect, they are separated from the limbic lobe by the calloso-marginal fissure and at the base are bounded posteriorly by the olfactory trigone. Perhaps only 30% of the frontal cortex appears on the surface, the rest of it being buried in the various sulci, with a special development of the frontal operculum adjoining the insula. The mass of each frontal lobe, when dissected free from other areas, is approximately one-third of the total mass of the hemisphere excluding the basal ganglia.

The human frontal lobe represents a huge development when compared with the frontal lobes of lower animals. In subprimate species, the frontal lobes consist almost entirely of motor and premotor cortex. In primates, the motor centers are still predominant, although there is a considerable area at the tip of the frontal pole that is not electrically excitable. At the same time, the importance of these motor areas in the frontal cortex rises progressively. In the rat, for instance, the whole cerebral cortex can be removed without causing any more than transient motor defects; in the chimpanzee, on the other hand, restricted removals of the motor zone cause persistent hemiplegia, which is increased to total incapacity if both sides are operated upon. But we are only indirectly interested in the motor functions of the frontal lobes. They are impaired only temporarily and in subtle ways by prefrontal lobotomy, unless some accident happens that destroys areas beyond the planned incisions. Those areas in advance of the motor regions are the ones with which we are primarily concerned.

The cellular architecture of the frontal lobes varies from place to place and has been the subject of investigation for eighty years since Meynert's first efforts to correlate structure with function. Even earlier than this, Betz had described the cells that bear his name in the paracentral lobule. Campbell and Elliott Smith undertook systematic studies, which were later followed up by Brodmann, the Vogts and especially by Von Economo and Koskinas. Brodmann, whose chart is extensively used, gave numbers in arbitrary fashion to areas that showed differences in cortical lamination and these numbers have been employed for nearly forty years to designate the various areas. With further study, the number of subareas has grown, so that the Vogts described some sixty different areas in the frontal lobes alone. Von Economo and Koskinas, recognizing the inevitable confusion that attends such minute subdivision, attempted to group the areas into larger regions, on the whole with encouraging results.

In order to understand the somewhat unfamiliar yet logical terms employed by Von Economo and Koskinas, it may be stated that they start off with division of the cerebral cortex into seven lobes (see Frontispiece). Therefore, F stands for frontal, P for parietal, T for temporal and so on. Roman script (A, B, C, D, E, F, etc.) is then used to differentiate the main subdivisions of each lobe. In the case of the frontal lobe, these letters run from A to N. Most of these subdivisions have smaller variations from one part to another, hence additional designations are given, (op) for operculum, (γ) for gigantopyramidal, (I) and (L) for transition areas adjacent to the insular and limbic lobes, (m) for magnocellular, (p) for parvocellular and finally (Δ) and (γ) and some others for atypical developments. With these signposts to guide the student, it is not so difficult to visualize the location of any given area when named. These authors designate 35 different areas in the frontal lobes, including the transition zones where one type of cortex merges into a neighboring one (Figures 1, 2 and 3).

Figure 1

Figure 1. Cortical architecture according to Von Economo and Koskinas. Lateral surface. The figure is somewhat distorted by retraction of the opercula to demonstrate the insula.

Figure 2

Figure 2. Cortical architecture according to Von Economo and Koskinas. Medial surface.

Figure 3

Figure 3. Cortical architecture according to Von Economo and Koskinas. Basal surface.

Von Economo and Koskinas have gone farther than mere analysis. Following their detailed plan of studying each separate area, they have synthesized their observations, bringing together various individual types of cortex no matter where they are located into general types, of which they designate five. Three of these represent the typical six-layered cortex, while two types, termed Type 1 and Type 5, represent specialized developments. The main anatomic criterion for such division is the development of the granule layer. In the cortex of Type 1 the granule layer is absent, while in the cortex of Type 5 the granule layer is outstanding. Type 1 cortex represents the motor and Type 5 the primary sensory areas of the cerebral cortex (Figure 4).

Figure 4a

Figure 4a. Types of cortical architecture. Type 1 is agranular and essentially motor; type 5 is granular and essentially sensory; types 2, 3 and 4 run from the slightly granular to the strongly granular. Type 4 cortex is found at the base of the frontal lobe. Lateral surface.

Figure 4b

Figure 4b. Medial surface.

The intervening three types of cortex, which cover most of the cerebral mantle (excluding the olfactory cortex), show a trend from the slightly granular Type 2, to the moderately granular Type 3 and to the strongly granular Type 4. Von Economo and Koskinas term these respectively the frontal, parietal and polar types of cortex, because of predominance in corresponding areas (Figure 5). The last designation is particularly interesting in connection with the frontal lobes. To begin with, Type 4 cortex is found well developed at the occipital pole adjacent to the primary visual striate cortex. This cortex is rather narrow and has a strongly developed granule layer IV, partaking thus of sensory characteristics. From the functional viewpoint, it may be considered as an area responsible for the elaboration of primary sensory impressions into concepts, or possibly for visual awareness or alertness or vigilance. Type 4 cortex is also found in the depth of fissures in the superior parietal area.

There is no primary sensory granular cortex of Type 5 in the frontal lobe, but at the base of the lobe there is a rather strongly developed granular modification similar to that seen in the occipital pole (Figure 5d). The cortex of the gyrus rectus is the thinnest of all, measuring only 1.5 mm in the walls of the sulci. There is practically no layer V, the granule layer resting directly on the layer of polymorphous cells. Von Economo and Koskinas also found a large area of intermediate Type 3 cortex near the frontal pole. They believe that this type of cortex, represented so widely in the parietal and temporal areas, has the function of relating incoming impressions to those that have gone before; hence they speak of Type 3 cortex as carrying out gnostic functions. They do not speculate upon the significance of the presence of cortex of Type 3 and Type 4 in the frontal lobe. In their map of cortical localization, they merely designate the base of the frontal lobe as the seat of "character changes." They specifically state that their work deals with anatomic findings and that those who follow are welcome to make their own interpretations and draw their own conclusions from these unexplained findings. We are taking them at their word. We would suggest that Type 4 cortex at the base of the frontal lobe mediates a general awareness or consciousness of the self, in the most primitive form as visceral self-consciousness, but capable of further development into personal self-consciousness and so on up to spiritual self-consciousness.

Figures 5a, 5b, 5c and 5d. Differing types of cortical architecture encountered in the frontal lobe, after Von Economo and Koskinas.
Figure 5a   Figure 5b
Figure 5a. Type 1 cortex. From Area FB, Area frontalis agranularis.   Figure 5b. Type 2 cortex. From Area FC, Area frontalis intermedia.
Figure 5c   Figure 5d
Figure 5c. Type 3 cortex. From Area FD, Area frontalis granularis.   Figure 5d. Type 4 cortex. From Area FG, Area recta.

Connections of the frontal cortex with other areas in the brain have been worked out in detail on animals but less so in man. In 1902, Anton and Zingerle wrote a monograph on the subject, pointing out that the layering of the longitudinal fibers is well developed but not as much so as in the occipital lobe.

There are projection fibers from the agranular frontal cortex to the pons, and from the gigantopyramidal precentral cortex FAγ through the pyramidal tract, all the way down the spinal cord. There are no proven connections with the hypothalamus, but there is electrophysiologic evidence that the frontal cortex sends fibers (possibly only collaterals) to the hypothalamus and to the caudate and putamen. Anton and Zingerle state: "The external sagittal stratum is an anterior projection of the corona radiata, broadening out and surrounding the other fiber systems on its way to the frontal pole. It bends sharply around the frontal horn. Some of the fibers are distributed to the medial surface of the hemisphere, while others fan out as well to engage the superior frontal and the orbital convolutions."

Figure 6 illustrates a defibrillation dissection of the thalamofrontal radiation in a normal brain.

Figure 6

Figure 6. Defibrillation dissection of the thalamofrontal radiation. The head of the caudate nucleus has been removed. The fibers run through the anterior limb of the internal capsule, bend around the anterior horn of the lateral ventricle and spread out to all parts of the frontal lobe, particularly to the medial and orbital regions. (Preparation by Dr. Harold Lehrman from the Laboratory of Neurology, George Washington University Medical School.)

The afferent projection to the frontal lobes comes from the thalamus. There is a point-to-point arrangement of these thalamofrontal radiations. The medial nucleus projects to the frontal pole, its medial portion to the basal frontal region and its lateral portion to the frontal granular area. The lateral group of nuclei project largely to the agranular areas of the frontal lobe. The anterior nucleus of the thalamus projects to the limbic lobe. These are the conclusions arrived at from our study of some lobotomy cases in which incisions in the frontal lobes deviated by accident or design from the usual plane. They will be described in detail in Chapter XV.

Projection pathways in the frontal lobes develop myelin sheaths long before the association pathways. Reference to Flechsig's myelinization chart (Figure 7) shows that in the non-olfactory or isocortex, myelinization begins in the precentral white matter and shortly thereafter beneath the premotor areas. Next in order, the projection pathways from the thalamus to the cortex become myelinated. The best known are those to the primary cortical sensory areas in the parietal, temporal and occipital lobes, but the thalamofrontal radiation is a sizable one and develops its myelin at the same period. It is only later that the association fibers begin to appear in myelin sheath stains. Figure 8 shows the radiation in a seven month old infant. The thalamofrontal projection follows the anterior limb of the internal capsule, skirts the anterior horn of the lateral ventricle and bends sharply medial and downward to lose itself in the gyrus rectus, anterior to the olfactory trigone. The degenerations of this fasciculus have not yet been studied in detail for lack of proper material, but in myelin sheath preparations of frontal poles long after lobotomy there is a recognizable pallor in these regions, indicating degenerative changes.

Figure 7a

Figure 7a. Diagram of lateral surface of cerebral cortex showing order in which myelinization occurs (after Flechsig). The projection areas (cross hatched) become myelinated first, the the intermediate areas (vertical lines) and finally the association areas (white). The numbers refer to the approximate order in which these areas mature.

Figure 7b

Figure 7b. Same. Medial aspect of hemisphere.

In addition to projection pathways, there are three major association bundles. The fasciculus cinguli curves around the rostrum of the corpus callosum, runs beneath the limbic lobe and can be followed into the hippocampal region. This is apparently a trunk line, fibers joining and leaving it all the way along, since the interruption is not followed by any distant degeneration. The fasciculus uncinatus curves over the mouth of the sylvian fissure and links the frontal and temporal lobes, spreading out brush-like in each and interweaving with the fibers of other systems. The corpus callosum is the major commissure between the two sides. Connections exist between the frontal lobes and the parietal and occipital areas by means of superior and inferior association bundles, but they are late in development and are not clearly defined. The last connections to develop are the fibers connecting adjacent convolutions, the U-fibers.

Figure 8a

Figure 8a. Thalamofrontal radiation in a 7 month old infant. The anterior limb of the internal capsule spreads out lateral to the ventricle and bends medial and downward particularly into the gyrus rectus. Myelin sheath stain. Opposite the tip of the frontal horns.

Figure 8b

Figure 8b. Same. Close to the frontal pole.

Our studies have shown that following prefrontal lobotomy, degeneration in the cortex is minimal, whereas it is very pronounced in the thalamus. It would seem likely, therefore, that the effects of prefrontal lobotomy are due to the degeneration of the thalamus. Since the thalamus is the organ par excellence of affective experience, it is conceivable that the selective degeneration of the thalamus is the important factor in the altered emotional state of the mentally sick person following prefrontal lobotomy. Furthermore, it would seem possible to limit the incisions in prefrontal lobotomy to the bundles of fibers that would bring about the preferential degeneration of those areas in the thalamus that are important in the affective life of the psychotic individual. Spiegel and Wycis, with the aid of a stereotaxic instrument, have made a direct surgical attack upon the thalamus, with results comparable to those of lobotomy.

Experiments upon lower animals have been a fruitful source of information upon the functions of the frontal lobes. With the development of newer methods, the old stimulation-extirpation experiments have been replaced and conditions more nearly approaching the normal have been attained. The most promising of these methods is physiological neuronography, in which electrical or strychnine stimulation of an accurately localized area in one part of the brain can be picked up by virtue of synchronous electrical discharges at the point of termination of the stimulated axones. This reveals both the location and the direction of the pathway under examination. Transsynaptic activity can also be recorded electrically, so that the tracing of impulses becomes a matter of systematic search and enduring patience upon the part of the investigator rather than that form of cerebral mythology indulged in by the pathologists of a past generation.

Because results were easily demonstrated, the motor and premotor areas received attention first and there is a large body of information concerning these areas, their connections with other parts of the nervous system and their importance in the control of visceral musculature. It has long been known, for instance, that stimulation of the premotor area will bring about contractions of the stomach and intestines, while Watts and Fulton observed intussusception after bilateral extirpation of Area FB in the frontal lobes. Many investigators have reported elevation of blood pressure upon stimulation of Area FB. Cardiac rate, sweat secretion, skin temperature, skin resistance and vesical contraction can all be influenced by appropriate stimulation of the frontal cortex. The implications are obvious, and are succinctly put by Fulton:

"These newer disclosures concerning the relation of the cerebral cortex to the autonomic system give an adequate physiologic basis for the relationship long recognized between mental states and visceral processes ... The heart and circulation may be worked just as hard, and just as much to the detriment of the body as a whole, from an armchair — or perhaps I should say a swivel chair — as from a rower's seat. Many disturbances of gastrointestinal function undoubtedly have a similar basis and one might add that, instead of referring to them as 'psychogenic disturbances', one may more appropriately designate them as purely physiological aberrations explicable in terms of recognized physiologic mechanisms."

In contrast to the tonic, pressor and activating responses derived from stimulation of the convexity of the frontal lobe, inhibitory manifestations sometimes result from stimulation of the base of the frontal lobe. Bailey and his colleagues found that stimulation of the orbital surface inhibited respiration and gastric tonus. They also found that stimulation of the central end of the severed vagus nerve produced action potentials in the basal frontal cortex. This might be considered additional evidence that the basal frontal cortex was the end-station for visceral afferent impulses. The pathway is probably by way of the thalamus, since there are no known direct connections. Livingston, Chapman and Livingston found that stimulation of the basal frontal cortex in man prior to lobotomy caused inhibition of respiration or elevation of blood pressure, or both. Murphy and Gellhorn have shown by electrical and strychnine stimulation that there is a two-way path between the frontal cortex and the hypothalamus, via the medial nucleus of the thalamus. In connection with neuronographic experiments, they also demonstrated a direct connection between the ventrolateral nucleus of the thalamus and the hypothalamus, and reached the conclusion that continued somatic stimulation may set off uninhibited autonomic discharges. We believe that these connections are also of importance in the question of pain and its emotional and visceral manifestations.

By means of the strychninization technic, Bailey and his coworkers have established projections from the anterior nucleus of the thalamus to the cingulate cortex and Wilbur Smith, upon stimulation of its anterior portion, Area LA, obtained a complex response characterized by autonomic reactions and changes in facial expression with vocalization, which led him to consider this area concerned in emotional expression. Papez and Vonderahe are champions of the limbic cortex as far as the emotional life is concerned. We are more impressed by the evidence in favor of the basal frontal cortex.

There is still a large hiatus in our knowledge of the manner in which visceral afferent impulses are transformed into discomfort and how fear begins and manifests its effects in the viscera. There is some danger in transferring bodily to man the results obtained in animals, yet the more carefully the experiments are conducted, the more likely it seems that the physiologic functions in man differ only in their complexity and in the variety of stimuli that can excite the appropriate responses. There have been developments rather than additions as far as the human economy is concerned, but the fundamentals were laid down in the lowly pattern of the rodent. The higher one goes in the stage of the experimental animal, the more varied, flexible and inconstant the responses. This is in the nature of the nervous system. So we come to the conclusion that the frontal lobes are concerned with adaptation to environment and that the most complex adaptations are found in man. It may be possible at some future time to study the maladaptations in animals that are the equivalent of the neuroses and psychoses in man, but a lot more must first be learned from both animals and man.

Last Modified: October 1, 2004 Comments? Questions? feedback