The mammalian visual system is one of the most explored parts of the brain. Neuroscientists have studied for decades how information captured by our eyes is processed in the brain, eventually leading to what we experience as vision. The “primary visual cortex” (V1), a thin patch of nervous tissue located in the posterior part of the brain, is considered the primary entry point of visual information from the periphery to the cerebral cortex. According to the classical model of visual processing, V1 extracts basic features of the visual world, such as lines and edges, and relays visual information to “higher visual cortices” specialized in processing more complex visual features, such as shapes, objects or motion. It is generally believed that visual responses in higher visual cortices depend on V1. In fact, lesions of V1 strongly reduce visual responses throughout the cortex.
In contrast to this hierarchical model of cortical organization, we discovered that one of these allegedly “higher” visual areas, the post-rhinal cortex (POR), does not depend on V1. Using an optogenetic approach to silence V1 while recording from POR of awake head-restrained mice, we found that POR kept responding to visual stimuli even in the absence of input from V1.
Which structure, if not V1, could relay visual information to POR? It is known that a phylogenetically ancient structure located at the base of the brain, called the superior colliculus (SC), can indirectly transmit visual input to the cortex. However, despite this input, no cortical area had been described whose visual responses depended entirely on the SC, as SC lesions have little or no impact on the activity of all previously tested visual cortices. Surprisingly, we found that visual responses in POR were abolished upon silencing the SC. To track down the pathway that connects the SC to POR we used a combination of retrograde and anterograde trans-synaptic viral approaches. We found that the SC is di-synaptically connected to POR through the “pulvinar” nucleus of the thalamus, a relay station located in the center of the brain.
These experiments demonstrate that two parallel visual pathways link the eye to the cerebral cortex, one entering the cortex in V1 and the other one in POR. Do these two cortical areas capture distinct properties of the visual world? Inspired by the fact that the SC is particularly sensitive to the linear motion of small objects, we tested whether POR, which is driven by collicular input, was better than V1 in detecting small moving dots. Consistent with this hypothesis, we found that POR is significantly more sensitive to linear motion than V1.
Figure: Two “primary” visual cortices. The classical primary visual cortex (V1) receives retinal input from the dorsolateral geniculate nucleus of the thalamus (dLGN). In parallel, and independently of V1, the post-rhinal cortex (POR) gets visual information through the superior colliculus and the pulvinar nucleus of the thalamus.
The identification of a cortical area that responds to visual stimuli in the absence of the primary visual cortex could be related to an interesting clinical phenomenon called “blindsight”. It has been reported that some human patients, who are considered “cortically blind” due to lesions in their V1, are still able to respond to visual stimuli that they do not consciously see. Intriguingly, these patients seem to identify the position of moving objects even in the absence of conscious perception of them. This clinical condition is referred to as blindsight, and its neural basis is largely unknown. Interestingly, studies in primates have suggested that blindsight requires the SC and our current research is aimed at understanding whether POR is also involved in some aspects of this phenomenon.
Collicular cortex watches movies (Commentary by Natasha Bray, Nature Neuroscience Reviews)