Development of the Cerebral Cortex: Radial Unit Hypothesis, the Embryonic Protomap, and Neural Interactions

Cerebrum plays a major role in recognizing the surrounding objects. The cerebrum is the most significant part of the human brain located in the anterior part of the skull. A medial longitudinal fissure divides the cerebrum into left and right hemispheres, which are again divided into four lobes: frontal, parietal, temporal, and occipital. The Frontal Lobe is connected with emotions, movement, planning, reasoning, parts of speech, and problem-solving. The Parietal Lobe is associated with movement, recognition, orientation, and stimuli perception. The Occipital Lobe is related to visual processing and the Temporal Lobe is associated with perception, memory, and recognition of auditory stimuli, and speech. 

The function of the cerebrum is to integrate neutral information and coordinate voluntary activity. The cerebellum comprises centrally located white matter and an outer layer of grey matter called the cerebral cortex.

The grey matter of the cerebrum is composed of neural cell bodies, and it is a surface layer of cerebral hemispheres. The cerebral cortex is involved in processing and cognition. On the other hand, white matter is made of myelinated axons that join various cerebrum areas together. Only the outer grey matter which is on the surface layer can be called the cerebral cortex. 

Grey matter is the most noticeable visible feature of the human brain. It is only a few millimeters thick but makes up about 50% of the weight of the brain. There are extensive folds in the cerebral cortex’s surface that increase its surface area to make room for more neurons. In the folds, the ridges are called gyri, and valleys are called sulci.

What does the cerebral cortex do?

The cerebral cortex is involved in multitudinous brain functions because of its extensive connections with subcortical areas. It is often characterized as comprising three areas: sensory, motor, and association. The sensory areas receive and process information from different sensory modalities. For example, the primary somatosensory cortex receives information about tactile and touch-related sensations like pain and temperature. Similarly, other cortex areas receive and process information related to vision, taste, olfaction, hearing, and the vestibular senses.

The motor areas are principally found in the frontal lobe and include the premotor cortex, the primary motor cortex (the main pathway for voluntary movement), and the supplementary motor cortex. 

Association areas are involved in the complex process of integrating information from multiple brain regions. These areas handle the complex information processing between input in the primary sensory cortices and the output: generation of behavior. The efficient working of sensory, motor, and association areas is essential for healthy brain function, and it also accounts for most human cognition and behavior.

Layers of cerebral cortex

There are six layers of the cerebral cortex:

Molecular (plexiform) layer: This is the most superficial layer, laying directly under the pia mater. This layer is inferior with cellular components, represented by only a few horizontal cells of Cajal-Retzius. The central portion of this layer is the processes of the neurons lying within the deeper layers and their synapses.

External granular layer: This layer consists mainly of stellate cells and some form of small pyramidal cells. These small cells in this layer give that “granular” appearance to this layer, hence its name.

Cells of external granular layer send their dendrites to the molecular and other layers of cortex. The axons of these cells travel deeper to the cortex for local, intracortical synapsing. The axons of this layer can form the association fibers that travel through the white matter to end in the various structures of the CNS.

External pyramidal layer: This layer predominantly has medium-sized pyramidal cells that serve as an association and commissural corticocortical fibers. The dendrites of these cells extend superficially and reach the molecular layer, whereas the basal processes join the subcortical white matter. 

Internal granular layer: Most of the stimuli from the periphery arrive here, making it the primary input cortical station. It consists largely of the stellate cells and a few pyramidal cells. The axon’s pyramidal cells synapse deeper within the cortex or join the white matter fibers, whereas that of the stellate cells synapse locally. Also, the stellate cells contribute to the formation of specific sensory cortical areas. 

Internal pyramidal layer: This layer consists predominantly of medium-sized and large pyramidal cells. This layer is most prominent within the motor cortex because it is the source of the output or corticofugal fibers that mediate motor activity. 

Multiform (fusiform) layer: This is the deepest layer of the cortex that directly overlies the subcortical white matter. It contains fusiform primarily cells with less dominant pyramidal cells and interneurons.

The axons of this layer’s fusiform and pyramidal cells distribute corticocortical commissural fibers and corticothalamic projection fibers that end in the thalamus.

Development of Cerebral cortex

1. Radial Unit Hypothesis

Neurons are not born in the position that they occupy in the adult brain. Instead, they migrate or travel from the proliferative zone where they are formed to occupy the new position in the mature brain. The cell displacement is of two types: passive or active. The passive cell displacement occurs in the non-layered neural structure where recently born cells push the old cells further away from the proliferative zone. It is also called an outside-to-inside spatiotemporal gradient.

In contrast, active cell migration occurs in the cerebral cortex and subcortical areas with a laminar structure. In this type of migration, neurons cling to long fibers of the radial glia to reach their correct position. It creates an inside-to-outside spatiotemporal gradient as the fibers radiate from the inner to the brain’s outer surface. 

A fantastic feature of the cerebral cortex is its three-dimensional organization of layers and columns of functionally similar neurons. The radial unit hypothesis (Rakic, 1988) postulates a combination of timing and location of the birth of neurons for the three-dimensional organization. Autoradiographic studies of neuronal origin show that neurons within a given radial column originate from several clones (polyclones) that share the same birthplace, migrate along a common pathway, climb up the same radial glial fiber, and settle on top of each other within the same ontogenetic column (Rakic, 1988b). This has been confirmed by retroviral lineage tracing experiments in primates (Kornack & Rakic, 1995) and rodents (Reid et al., 1995), and chimeric and transgenic animals (Nakatsuji et al., 1991; Soriano et al., 1995; Tan & Breen, 1993). 

In contrast, their radial (vertical) position is determined by the time of their origin (Rakic, 1988b). The number of the ontogenetic columns determines the size of the cortical surface, whereas the number of cells within the columns determines the thickness. Therefore, questions about the evolution of cortical size and thickness become translated into questions about developmental regulation of total cell number and how this regulation was modified during evolution. 

2. Protomap hypothesis

A problem with the radial unit hypothesis is how the immature neurons arrive in the mature brain and differentiate into specialized and area-specific functions of the cortex. Traditionally, it has been argued that the embryonic cortical plate initially has equipotent cells that are later specified in their roles depending on the input from subcortical centers. This is also called the tabula rasa hypothesis, as the cells initially do not know their end destination. 

An alternative to the tabula rasa hypothesis is the protomap hypothesis, which postulates that intrinsic molecular markers determine the laminar and areal specificity of cortical neurons at the time of their last cell divisions in the proliferative ventricular zone. 

Cortical patterning determines the generation, size, shape, and spatial pattern of various functional areas of the cerebral cortex. Protomap is patterned through cortical patterning by signaling centers in the embryo which provides the positional and cell fate instruction. These early genetic instructions lead to the mature functional areas of the cortex, for example, the somatosensory, visual, and motor areas. The term protomap was coined by Pasko Rakic. 

According to this theory, the primordial identity of each functional area of the cerebral cortex is encoded within the cortical stem cells before the formation of the cortical layers. By this view, differentiation into cortical regions occurs early in the cortex’s formation, and the proliferative ventricular zone form a mosaic that establishes a species-specific cortical map. The individual cells preserve the information of their laminar and areal positions even during the migration to the cerebral cortex. 

The protomap hypothesis was opposed by the protocortex hypothesis which suggests that an initially undifferentiated protocortex is divided mainly because of input through projections from the thalamus and is activity-dependent. 

3. Neural interaction

Interactive specialization addresses localization and specialization, two issues in cognitive neuroscience. Localization is the computational function that can be associated with an area of the cortex—specifically, the cortex areas that are activated during a task. Specialization refers to the specific function of an area of the cortex where the functions may be finely or broadly tuned. A cortex region can be fine-tuned when only a restricted category of stimulus or narrow range of task demands activates them and or broadly tuned when activated under a wide range of circumstances. 

The interactive specialization view of human functional brain development predicts that the social brain emerges as a result of finely-tuned to relevant stimuli and events in an activity-dependent manner. 

However, according to the interactive specialization view, the issues of localization and specialization are both consequences of the same common underlying mechanisms. This view argues that many areas begin with poorly defined functions early in postnatal development, and a wide variety of sensory tasks and inputs can partially activate them. During development, the activity-dependent interactions between regions lead to modifications of the intra-regional connectivity. These interactions change the cortex area to become restricted to a narrower range of tasks or stimuli.

The functional imaging studies show that the increasing localization of the functional areas makes them increasingly distinct from their surrounding cortical tissue. For example, Nelson 2003 talks about perceptual narrowing in the development of face processing where the behavioral evidence is consistent with the idea of increasingly finely tuned cortical processing of faces. This concept of the ‘narrowing process’ resulted in better recognition of non-human faces by younger infants (Pascalis et al., 2002). 

Some similar findings were reported for localization in the face matching task and language acquisition by Schlagger and McCandliss (2007). Schlagger and McCandliss have used neuroimaging and other sources to argue that the emergence of the cortical area activated explicitly by reading words occurs through the interactive specialization processes of increased specialization and localization, close association with the child learning to read. In another example, evidence that changes in visual orienting abilities as assessed by fMRI revealed multiple sites of change throughout a network of different pathways involved in oculomotor control and not just the activation of one or two “new” functional areas. 

In summary, according to the interactive specialization view, small areas of the cortex become narrow-tuned for certain functions because of a combination of factors, including

  1. The suitability of the biases within the large-scale region (e.g., transmitter types and levels, synaptic density, etc.),
  2. The information within the sensory inputs (sometimes partly determined by other brain systems), and
  3. Competitive interactions with neighboring regions (so that functions are not duplicated).


  1. Cadwell, C. R., Bhaduri, A., Mostajo-Radji, M. A., Keefe, M. G., & Nowakowski, T. J. (2019). Development and arealization of the cerebral cortex. Neuron, 103(6), 980-1004.
  2. Falk, D., & Gibson, K. R. (Eds.). (2001). Evolutionary anatomy of the primate cerebral cortex. Cambridge University Press.
  3. Johnson, M. H., & De Haan, M. (1997). Developmental cognitive neuroscience: An introduction. (No Title).
  4. Rakic, P. (2000, January). Radial unit hypothesis of neocortical expansion. In Novartis Foundation Symposium (pp. 30-52). Chichester; New York; John Wiley; 1999.
  5. Rakic, P. (1988). Specification of cerebral cortical areas. Science, 241(4862), 170-176.

Discover more from Osheen Jain

Subscribe now to keep reading and get access to the full archive.

Continue reading

Scroll to Top