What is the role of the thalamus in vision?
Prepare for some spectacular oversimplification. This is necessary both because of the inherent complexity of the thalamus and the limitations of my knowledge. I will provide a brief overview of the thalamus because understanding its overall nature makes explaining its relation to vision clearer.
The first simplification is that I will consider the visual and auditory colliculi to be part of the thalamus. Anatomists consider them to be part of the tectum, but they are all very close together and accomplish the same purpose.
For every exterior sense except olfaction and auditory processing, the thalamus is the first destination of neurons carrying a sense from the outside of the body. Olfaction finds its way to the thalamus from the piriform cortex rather than directly from the olfactory bulb. What’s a couple hundred milliseconds latency for detecting an odor? Auditory signals arrive from the superior olivary complex, whence they have been preprocessed for tone and direction, both valuable for thalamic processing.
For all the peripheral senses, the neural signals arrive with coding that indicates where in the outside world the sensation is coming from.
- For touch sensations, each sensory neuron arrives from a particular location on the body surface, reliably indicating where it is coming from.
- For proprioceptive sensations, each sensory neuron is reliably associated with a particular muscle or joint, again reliably indicating its source.
- For taste and smell, the sensations are reliably linked to the mouth and nose, respectively, without any particular spatial resolution.
- For audio information, the preprocessed audio mentioned above arrives in the inferior colliculus to provide a low resolution cue about where in the left/right axis the sound is coming from along with the general nature (amplitude, frequency) of the sound.
- For visual information, the retinal signals arrive coded by a region of the visual field. The right side of the visual field of both eyes goes to the left superior colliculus, and the left side of the the visual field of both eyes goes to the right superior colliculus.
The thalamus carries out two primary functions for all sensations (again, except for olfaction), based on the general nature of the sensation (pain, pitch, intensity, etc.) and its explicit or implicit location relative to the body. The original and primary function is to make immediate judgements about survival needs and inform action centers to respond appropriately. The thalamus is located just above the brain stem, indicating its primordial connection with survival.
The more recent and secondary function of the thalamus is to relay the incoming sensory signals to other parts of the brain, primarily the cerebral cortex. The touch and proprioceptive signals are relayed to the somatosensory cortex in the parietal lobe, located just aft of the motor cortex. This area is often represented by the Cortical homunculus. The inferior colliculus relays sound information to the audio cortex in the temporal lobe. The superior colliculus relays visual information to both the primary visual cortex (V1) in the back of the occipital lobe and the collicular visual cortex in each temporal lobe.
With that background, I will try to answer the original question. The secondary visual role of the thalamus is to relay visual information to two separate visual cortex regions.
The primary visual role of the thalamus is to recognize threats as early as possible. But how can it do that without interpreting the retinal information? The answer is that the V1 visual cortex sends predictions to the superior colliculus, and the colliculus compares the predictions to the received retinal signals. Mismatches between predictions and reality may represent threats of some sort, and one of the primary places these signals indicating missed predictions are sent is the amygdala.
Before I describe those predictions in more detail, I want to point out that this comparing of predictions and sensations is happening simultaneously in the thalamus for all of the senses, with the possible exception of olfaction. Not only are all the senses being tested, but the explicit and implicit location information allows the thalamus to perform cross-sensory analysis. This means the thalamus can recognize that a visual surprise, an auditory surprise, and touch surprise are all happening in the lower right side of the back, perhaps indicating an animal attack there. All of this happens without any conscious cognition, or any real understanding of what the nature of the threat is. The key is that the threat can be recognized by the thalamus in less than a dozen milliseconds after it appears, valuable because the slower cognitive process that can take hundreds of milliseconds aren’t involved.
Finally we can look at the nature of thalamic visual processing. The first important information is that there are more neural projections from V1 visual cortex to the thalamus than there are from the thalamus to V1. So while the superior colliculus (SC) is acting as a forwarding agent, it is obviously doing much more than that, or it wouldn’t need all those back connections.
To understand what the SC is likely doing, we have to consider the highly heterogeneous nature of human vision. One part of our vision is very high resolution, the fovea, but that part represents about one tenth of one percent of our field of view. For the purpose of understanding the survival purpose of the thalamus, what really matters is our peripheral vision. The retina condenses our entire peripheral vision into about a half million retinal ganglion cells (RGCs), half of the total million that pass through the optic nerve. It does this by combining swaths of the periphery into summarized regions, which is why our peripheral vision has low resolution.
Here’s an image of human peripheral vision from Peripheral vision — Wikipedia.
This low resolution information is what the thalamus cares about. It doesn’t care about the high resolution foveal center, because important survival issues are unlikely to occur in that minuscule region. I suspect it treats the fovea as just one of the half million regions it processes. Furthermore, the peripheral regions aren’t homogeneous, since ones closer to the fovea are smaller (providing higher resolution) than those near the edge of the periphery.
What the V2 cortex is predicting and sending to the thalamus is a map of the lighting levels in each of the half million regions. This map is a prediction of how bright/dark various parts of the field of view should be. So a tree may cause a vertical band of darkness, but a sudden darkness from an incoming tiger, would be a surprise. The thalamus would not be recognizing the tiger, only the lighting level changes it produces. From a survival point of view, the difference between a tiger and a human is minimal, and the brain will respond to the perceived threat while the rest of the visual system takes the hundreds of milliseconds necessary to determine the nature of the threat.
In summary, the thalamus is the first line of defense against surprises that can ruin our day. Most of the time, our brains correctly predict the lighting changes the thalamus “sees”, but occasionally our higher cognitive functions have to sort out the nature of the threat.
