Discovery of the Cortical Barrel

If ever a picture launched a thousand scientific investigations, it is Figure 15 of the 1970 paper by Thomas Woolsey and Hendrik Van der Loos entitled "The structural organization of layer IV in the somatosensory region (SI) of mouse cerebral cortex: The description of a cortical field composed of discrete cytoarchitectonic units".

Figure 15 (slightly adapted) from Woolsey and Van der Loos (1970)

This figure, the last in the report, shows a remarkable correspondence between the pattern of the mystacial vibrissae (whiskers) on the face of a mouse and the spatial organization of neuron clusters in its contralateral cerebral cortex. Because of their characteristic three-dimensional shape, Woolsey and Van der Loos called these neuronal structures "barrels". Figure 15 presents the authors’ "one barrel-one vibrissa" hypothesis which proposed the now firmly established one-to-one relationship between individual whiskers and their corresponding cortical barrels. The hypothesis was based in part on work performed by Woolsey several years earlier as a medical student, in which surface evoked potential recording techniques, perfected during the 1930's and 1940's by Dr. Clinton Woosley, Thomas’ father, were used to define the cortical maps representing visual, auditory and somatic sensory peripheries in the mouse. In attempting to interpret his physiological findings in the context of previous anatomical descriptions of the mouse cerebral cortex, Dr. Woolsey realized that the functional representation of the animal’s face was coextensive with regions containing "cell dense nets" described previously by others in what was thought erroneously to be auditory cortex. Examining this cortical area more closely, Woolsey and Van der Loos sectioned the cortical tissue in a tangential plane of section that permitted visualization of a large horizontal expanse of layer IV in a single tissue section. When viewed from this novel perspective, the pattern of barrels - and their correspondence with the facial whiskers - was unambiguously revealed. The symbiosis between anatomical and physiological approaches, so elegantly demonstrated in this early work, has become a hallmark of the experimental investigation of the whisker/barrel system.

Woolsey’s and Van der Loos’ "one barrel-one vibrissa" hypothesis was quickly confirmed. Influenced by classical neuroembryological experimentation, Van der Loos and Woolsey (1973) ablated selected whisker follicles in newborn mice and observed, in the mature animal, an absence of cortical barrels corresponding precisely to the damaged vibrissae. The diameters of neighboring barrels increased and occupied the cortical territory that would have represented the ablated whiskers. Importantly, the effects on the cortical architecture were age-dependent, such that damage to the whisker follicles led to alterations in barrel field organization only if it was produced within the first five days of postnatal life. In addition to supporting the "one barrel-one vibrissa" hypothesis, the neonatal whisker ablation study suggested that the structural and functional development of the somatosensory cortex was strongly influenced by the sensory periphery. Compelling support for this idea was provided a few years later in the form of comparative studies examining the relationship between whiskers and barrels in different species. These investigations showed that, in those species where cortical barrels could be observed, their number and spatial organization was always isomorphic to the pattern of whiskers on the animals’ face, which varied from one species to the next. Concurrently, Herbert Killackey demonstrated, using axonal degeneration methods, that the cortical barrel field received a correspondingly patch-like innervation from the thalamic ventral posterior medial nucleus, the major pathway in rodents as in higher mammalian orders that transmits afferent sensory information to the somatosensory cortex.

Direct functional evidence for the "one barrel-one vibrissa" hypothesis was provided by a series of elegant electrophysiological studies by Carol Welker (1976). Using microelectrodes to record action potentials from small numbers of neighboring layer IV neurons in the primary somatosensory cortex of barbiturate-anesthetized rats, she carefully mapped the representation of the contralateral body and face, making microlesions at selected recording sites. The cortical hemispheres from these specimens were then flattened and sectioned in tangential plane, as done earlier by Woolsey and Van der Loos. Examination of the tissue revealed the presence of barrels or barrel-like structures within the functionally defined representations of different parts of the animal’s body surface, including the fore- and hind-paws. In every case, marker lesions made at physiologically identified recording sites were found in the appropriate location within the anatomical map made by the barrels. Moreover, as in other mammalian sensory systems, the somatosensory representation was grossly distorted, with regions of high peripheral innervation density, like the hand representation in primates and the whisker representation in rats, occupying a disproportionately large extent of cortical space. Shortly thereafter, Simons used more refined electrophysiological techniques to investigate in rats and mice the receptive field properties of neurons located in different cortical laminae. These studies demonstrated that a barrel, in layer IV, was the morphological counterpart of a functional cortical column that extended throughout the cortical depths and that neuronal receptive field properties varied with laminar location in a fashion suggestive of a hierarchical organization of information processing. These early physiological studies thus provided important links to studies in other species, including cats and monkeys, suggesting important commonalities in the general organization of sensory cortex and the nature of information processing within local cortical circuits.