Why doesn’t olfaction pass through the thalamus when all the other senses do?

First I’ll explain why the olfactory sense is different from the others, then I will show how it works differently in the limbic system.

Because all the sensory nerves besides olfactory ones traverse the thalamus, we often think of the thalamus as the first stop in sensory processing, but it is really the final stop, or culmination, of the downward pathways of the predictive brain.

Historically, brain science has described the brain as something like a computer, suggesting it works with an input-process-output (IPO) model. In this model, it would make sense to gather all the senses in a place like the thalamus and forward them to the sensory cortex areas for processing into output/muscle commands. We know this is not how the brain works, though it may have been how the very first brains in the mud of ocean worked; we’ll never know because soft bits like brains don’t fossilize.

What we know now is that all brains are predictive. That is, all brains in all animals maintain a model of the world and use it to predict what will happen around them in the next half second or so. An IPO brain can’t do that since it takes hundreds of milliseconds to process and respond to a sensation. Prediction allows animals to respond to sensations in milliseconds instead of hundreds of milliseconds.

Olfaction is an exception, and here are some reasons:

• Olfaction was probably the first sense. We know bacteria primarily sense their surroundings with chemical receptors on their membranes, and so do single cell eukaryotes.
• Olfaction was the primary sense for early multicellular creatures, which were sitting or moving in mud or muddy waters at first. One indication is the relative size of the olfactory bulb in early species, often over half the size of the brain.
• Olfaction continues to be a primary sense for almost all animals, filling the space we think of as the nose and being highly connected into other brain areas.

Evolution was simultaneously maintaining the importance of olfaction and making all the other senses predictive, and this leads to the conundrum of this question.

Olfaction is a rich sense, and the complexity of the olfactory bulb in every animal demonstrates that. There is evidence that humans can distinguish a trillion odors, and evidence that a dog can distinguish 10,000 times as many. I don’t think we really know, but it’s huge. This leads to a number of unique issues with predicting odors:

• Prediction as used for the other senses requires a downward path to meet the incoming olfactory sensations. The number of sensory prediction neurons would have to be about as large as the preliminary resolution of an “olfactory thalamus”.

[Preliminary resolution is a low level early prediction resolution used in the thalamus. For vision, it is less than a million signals matching those from the optic nerve, though our vision processes those one million retinal signals temporally to provide effectively higher resolution in the visual cortex. Audio processing bandwidth is reduced by suppressing some sounds prior to reaching the thalamus. Tactile processing is summarized by combining signals from adjacent areas. Neither evolution nor I have thought of a way to reduce preliminary resolution for olfaction.]

• Olfaction is much slower than any of the other senses, so the need to process it quickly is reduced. The speed of molecular diffusion through the air is much slower than either the speed of light or sound, and it’s even slower than the neuron transport speed of tactile signals from the skin. This speed difference reduces the value of prediction for the olfactory sense.
• Olfaction doesn’t provide useful orientation to larger animals. While smaller animals (snakes) are known to orient based on olfactory cues, animals whose primary interests are more than ten times the distance between its nostrils don’t gain much directionality from olfaction.
• The nature of olfaction as an environmental cue means that we often detect odors related to survival before we process explicit (eye, nose, touch) sensations.
• Olfaction is often an unpredictable clue that something in the environment requires our attention. For this purpose, prediction isn’t particularly useful.

For these and other reasons, the olfactory bulb is considered part of the limbic system, which also includes the thalamus, amygdala, and hippocampus. What it does instead of feeding directly into the predictive processes of the limbic system through the thalamus is establish states and conditions for responding to other senses.

The later processes of olfactory identification are connected to the thalamus and amygdala directly, so olfactory sensations quickly influence predictive behavior, more in the hundreds of milliseconds rather than immediately.

The evidence suggests that the olfactory sense has never passed through the thalamus. Some other senses pass through without being processed by it.

• Proprioceptive neurons pass through the thalamus, but are not processed there. Proprioceptors provide feedback to motor activity and terminate in the motor cortex. This is a clue that proprioceptors were originally (early in evolution) sensory neurons that were recruited to modulate motor activity.
• Nociceptor (pain) signals pass through the thalamus, but are apparently not processed there. Instead, they pass through and continue on to the somatosensory cortex, along with other tactile sensations. Other tactile sensations are predicted in the thalamus as they pass through. By not predicting nociceptors, their nerve speed requirements are reduced, allowing them to avoid the metabolic and conductor costs of myelinization. For example, it can take over a second for a nociceptor signal from the toe to reach the brain.
• All other sensations besides olfaction, including vision, hearing, touch, light touch, heat, cold, etc., appear to be predicted and their predictions are tested in the thalamus.

From this we can infer that while proprioception and nociception were at an earlier evolutionary time processed in the thalamus, olfaction was not.

Humans are not so different from other animals, and most of what we know about the thalamus is learned from other animals before being verified in humans. We find both similarities and differences, providing clues for the path evolution has taken.

If you like my writing, you can read more on Quora with author John Light.