The Thalamus dorsalis

Anatomical terms are often very apt. Not so in the case of the thalamus. The Greek and Latin word “thalamus” actually means any room, a cavity, a living space. However, in modern anatomy, the thalamus is located on either side of a ventricular space – the third ventricle in the diencephalon: it is part of the wall and not the cavity. The discrepancy arose because brain functions were originally sought in these very ventricles, where the human mind was thought to be distilled, comparable perhaps to good rum ... but that's another story.

In any case, this third ventricle of the diencephalon, at the base of which lie the conspicuous optic nerves, was aptly named the optic thalamus. When it was realized that it is the gray matter in the walls of the brain that carries out its functions, the term changed accordingly. And when it was realized that only a small part of the thalamus is optical (the “tail” at its end, see below), the ‘opticus’ was dropped. And when it was further realized that the thalamus consists of an upper and a lower level, the upper level – which will be discussed here – was given the addition “dorsalis.” This clarification is hardly used anymore, because the lower “ventral thalamus” (consisting of the subthalamic nucleus and globus pallidus) has completely different functions. So when anatomists simply say “thalamus,” they mean the thalamus dorsalis.

The two dorsal thalami are impressive structures, each about the size of a thumb. Their shape is a little strange; they resemble clumsy drops with their blunt heads pointing toward the forehead and their short stubby tails at the rear bent downward and outward. The body of the drop is the actual thalamus, the transition region to the stubby tail is called the pulvinar, and the tail itself is called the metathalamus, i.e., the post-thalamus, which consists of two knee bumps, the lateral and medial geniculate bodies (“genu” is Latin for “knee”). The actual thalamus, the drop, consists of many individual nuclei with different functions, making it one of the most complex structures in the brain.

The thalamus is connected to the cortex. “Patchy,” one might say, because certain areas of the thalamus send axons to certain areas of the cortex – and vice versa. The pattern of these connections – i.e., “who with whom” – is extremely stable and the same in all humans (and many animals). Accordingly, they are referred to as “specific thalamic nuclei.” They can also be understood as “cortical agents”: they carry information from the sensory systems, the eyes, the ears, the skin, the muscles and joints, the sense of pain and the sense of taste to the cortex. The only exception is the sense of smell. But even that – if it is to become conscious – must ultimately pass through the thalamus.

The fiber pathways of all these sensory systems converge in the thalamus at specific subnuclei – the visual pathway at the lateral geniculate bodies, the auditory pathway at the medial geniculate bodies. The remaining sensory pathways mainly run to the rear sides of the thalamus itself. In these subnuclei of the thalamus, synaptic switching then takes place to the last neuron of the corresponding sensory pathway. The axon of this neuron then runs to the respective primary sensory cortex area – and this area then sends axons back to “its” thalamic nucleus.

However, these genuinely sensory nuclei of the thalamus make up only a small part of its total mass. The other nuclei are less “agents” of the cortex than “mirrors” of it. They also send axons to specific, circumscribed areas of the cortex – but their main input does not come from the sensory periphery, but from the cortex itself, either directly – because, as mentioned above, every area of the cortex that receives thalamic inputs sends axons back – or indirectly, via loops. It would therefore be an oversimplification to refer to the thalamus – as is often done – merely as the “gateway to consciousness”: it is much more than just the “conveyor” of sensory information.

Let's start at the back with the metathalamus, the “outwardly curved stub tail.” It is not completely smooth, but slightly bumpy, which is why we distinguish between the lateral geniculate nucleus (LGN) and the medial geniculate nucleus (MGN). As already mentioned, the auditory pathway ascends to the CGM, while the optic tract descends to the CGL. And just to illustrate the relationship between sensory and cortical input with an example: the CGL – after all, the main switching point for visual signals from the retina on their way to the primary visual cortex at the back of the head – actually receives only 10 percent of its inputs from the retina. The other 90 percent come from the primary visual cortex. It is assumed that a selection of the transmitted information already takes place at this unconscious level – recent studies show that certain layers of the visual cortex receive thalamic inputs, and other layers of the cortex then provide feedback to the CGL. In this way, the visual cortex modulates its own inputs.

The pulvinar is also involved in visual processing and, in addition to cortical connections to large areas of the temporal and parietal lobes, also has connections to the CGL and the superior colliculi of the midbrain – all centers involved in the processing of visual information.

Things get a little confusing in the thalamus itself, i.e., in the actual body of the “drop.” A distinction is made between an anterior group of nuclei (nuclei anteriores), which projects to the cingulate gyrus, a group of nuclei located at the ventricle (nuclei mediales), which is connected to the frontal lobe, and a large lateral group of nuclei (nuclei laterales), which projects to the frontal and parietal lobes. Of all these nuclei, only one subnucleus of the lateral nucleus group, the posterior ventrolateral nucleus, is genuinely sensory: the pathways of the bodily senses lead to it. Its axons, in turn, ascend to the postcentral gyrus of the parietal lobe. All the other nuclei mentioned are involved in more or less complex feedback loops between the cortex and other areas of the brain.

All of the above-mentioned nuclei maintain massive connections to specific, circumscribed areas of the cortex. They are therefore also called “specific” nuclei.

The “non-specific nuclei” have less pronounced connections to the cortex, or none at all. One of them is the reticular nucleus. It surrounds the lateral surface of the thalamus like a shell and contains inhibitory GABAergic neurons that send their axons into the thalamus itself. It is believed to be involved in attention control.

Other non-specific nuclei are located inside the thalamus itself, in thin lamellae of white matter that run through and divide it. These so-called “intralaminar nuclei” send their axons mainly to the basal ganglia, but also have diffuse, widely scattered cortical projections. Their inputs come primarily from the formation reticularis of the brain stem and – primarily motor – from the basal ganglia and the cerebellum.

In line with the manifold tasks of the thalamus, deficits are the effects of lesions. And some of them are very specific: if, for example, the ventral posterolateral nucleus is affected, this leads to disturbances in surface and deep sensitivity – and the feeling of swelling and heaviness in the extremities can be the result. Pain, motor phenomena, or paralysis are possible. Fortunately, extensive damage to the dorsal thalamus is very rare. But when it does occur, everything goes wrong – there are not only sensory deficits, but also severe motor deficits. And, above all, psychological problems.

First published on August 23, 2011
Last updated on November 28, 2025