Anatomy of the Prefrontal Cortex

The prefrontal cortex is among the last regions of the brain to develop, in evolution and in individual maturation. In the human, it constitutes nearly one-third of the totality of the neocortex. The most recent prefrontal cortex (pole and lateral convexity of the frontal lobe) is layered and granular, with a distinct layer IV which is heavily populated by granular cells. The more primitive, phylogenetically earlier, ventral and medial prefrontal cortex is dysgranular, poorly layered, and structurally and functionally akin to limbic cortex. Anatomically defined by its connectivity with the nucleus medialis dorsalis of the thalamus, the prefrontal cortex is the best connected of all neocortical regions. It is reciprocally connected, directly or indirectly, with the posterior association cortex, thalamus, hypothalamus, amygdala, basal ganglia, insula, hippocampus, cerebellum, and brainstem.

Keywords

afferent; comparative anatomy; connectivity; development; dorsolateral cortex; efferent; evolution; thalamus; ventromedial cortex

Outline

I Introduction

This chapter is devoted to the anatomy and developmental neurobiology of the prefrontal cortex. It begins with the discussion of issues related to the phylogenetic development and comparative anatomy of the neocortex of the frontal lobe. After that, the chapter deals with its ontogenetic development. Then, the chapter deals with the anatomy and microscopic architecture of the prefrontal cortex in the adult organism, and with the morphological changes it undergoes as a result of aging. Finally, the chapter provides an overview of the afferent and efferent connections of the prefrontal cortex in several species. This overview of connectivity of the prefrontal cortex, which is arguably the most richly connected of all cortical regions, opens the way to subsequent chapters, where connectivity is found to be the key to all its functions.

II Evolution and Comparative Anatomy

The prefrontal cortex increases in relative size with phylogenetic development. This can be inferred from the study of existent animals’ brains as well as from paleoneurological data (Papez, 1929; Grünthal, 1948; Ariëns Kappers et al., 1960; Poliakov, 1966b; Radinsky, 1969). It is most apparent in the primate order, where the cortical sector named by Brodmann (1909, 1912) the “regio frontalis,” which approximately corresponds to what we call the prefrontal cortex, constitutes, by his calculations based on cytoarchitectonics, 29% of the total cortex in humans, 17% in the chimpanzee, 11.5% in the gibbon and the macaque, and 8.5% in the lemur (Brodmann, 1912). For the dog and the cat, the figures are, respectively, 7% and 3.5%.

The use of values such as those to estimate differences in evolutionary growth has pitfalls and limitations, however (Passingham, 1973). The old notion that the entirety of the frontal lobe is relatively larger in humans than in other primates has been challenged by the results of brain imaging in several primate species (Semendeferi, 2001). Furthermore, by calculating the volume of the prefrontal cortex and plotting it against the total volume of the brain (in rat, marmoset, macaque, orangutan, and human), some authors have come up with a linear relationship, thus belying the volumetric prefrontal advantage of the human (Uylings and Van Eden, 1990). Others, however, have utilized sound empirical reasons to argue that in the course of evolution the prefrontal region per se, strictly defined cytoarchitectonically, grows more than other cortical regions (review by Preuss, 2000). No one has persuasively denied that in the human, as Brodmann showed, the prefrontal cortex attains the greatest magnitude in comparison with those other regions. That greater relative magnitude of the human prefrontal cortex presumably indicates that this cortex is the substrate for cognitive functions of the highest order, which, as a result of phylogenetic differentiation, have become a distinctive part of the evolutionary patrimony of our species. It has even been proposed that certain cortical areas, such as Broca’s area – which is arguably prefrontal – have developed by natural selection with the development of language, a distinctly human function (Aboitiz and García, 1997).

It is always difficult to draw evolutionary conclusions from neuroanatomical comparisons between contemporaneous species in the absence of common ancestors (Hodos, 1970; Campbell, 1975). Such comparisons commonly fail to establish the homology of brain structures (Campbell and Hodos, 1970), and this is a particularly vexing problem when dealing with cortical areas. Ordinarily, for lack of more reliable guidelines, the neuroanatomist uses structural criteria to determine cortical homology. The principal criteria for defining the prefrontal cortex and for establishing its homology across species are topology, topography, architecture, and fiber connections (hodology). The same criteria have been utilized in attempts to elucidate its evolutionary development.

The neocortex of mammals has emerged and developed between two ancient structures that constitute most of the pallium in non-mammalian vertebrates: the hippocampus and the piriform area or lobe (Figure 2.1). The process is part of what has been generally characterized as the evolutionary “neocorticalization” of the brain (Jerison, 1994). What in the brain of the reptile is a sheet of simple cortex-like structure bridging those two structures is replaced and outgrown by the multilayered neocortex of the mammalian brain (Crosby, 1917; Elliott Smith, 1919; Kuhlenbeck, 1927, 1929; Ariëns Kappers et al., 1960; Nauta and Karten, 1970; Aboitiz et al., 2003). Because the growth of the newer cortex takes place in the dorsal aspect of the cerebral hemisphere, the evolutionary process has been characterized as one of “dorsalization” of pallial development. Strictly speaking, however, it is inaccurate to consider the reptile’s general cortex as the homologous precursor of the mammalian neocortex (Kruger and Berkowitz, 1960). Moreover, there are other plausible alternate theories of neocortical evolution in addition to the above (Northcutt and Kaas, 1995; Butler and Molnar, 2002). In any case, it appears that the mammalian neocortex is phylogenetically preceded by certain homologous subcortical nuclei in the brains of reptiles and birds.

Studies of cortical architecture in aplacental mammals, such as those by Abbie (1940, 1942), have been helpful to trace neocortical development. They reveal that the neocortex is made of two separate components or moieties, one adjoining the hippocampus and the other the piriform area, that develop in opposite directions around the hemisphere and meet on its lateral aspect. Both undergo progressive differentiation, which consists of cortical thickening, sharpening of lamination, and, ultimately, emergence of granular cells. In higher mammals the two primordial structures, the hippocampus and the piriform lobe, have been outflanked, pushed against each other, and buried in ventromedial locations by the vastly expanded cortex (Sanides, 1964, 1970). Around the rostral pole of the hemisphere, the two phylogenetically differentiated moieties form the prefrontal neopallium.

The external morphology of the frontal region varies so much from species to species that it is difficult to ascertain the homology of its landmarks. Within a given order of mammals, certain sulci can be identified as homologous and used as a guide for understanding cortical evolution; across orders, however, all comparisons are hazardous. Nevertheless, some general principles of prefrontal evolution seem sustainable. One such principle is that, like the rest of the neopallium, the frontal cortex becomes not only larger but also more complex, more fissurated and convoluted, as mammalian species evolve. In primates, the process reaches its culmination with the human brain.

We should note, however, that the phyletic increase in gyrification and fissuration can be attributable to mechanical factors and not only to such factors as functional differentiation. The cortex folds and thus gains surface, keeping up with the three-dimensional expansion of subcortical masses (Bok, 1959). Thus, the overall number of gyri and sulci that form with evolution is largely a function of brain size, as stated by the law of Baillarger–Dareste (Ariëns Kappers et al.,...