Differences in orientation maps in the mammalian visual brain

ArticleDetailDownload PDF

Professor Michael R Ibbotson and Dr Jason Jung from the National Vision Research Institute, Australia, provide theories regarding the differences of orientation maps in the visual cortex in mammalian brains. In primates, the orientation maps are highly structured in a pinwheel cortical architecture: in rodents, however, they are randomly distributed. Using comparative brain studies, it was theorised that eye divergence, cell densities in the visual cortex, densities of lateral geniculate nucleus axons, and central peripheral ratio of retinal ganglion cells play a role in the differences seen in mammals. Further studies in intermediate animals between primates and rodents are needed to support or contradict these hypotheses.

The cerebral cortex is the outermost and largest part of the brain. It is the brain’s control and information processing centre, responsible for many higher-order brain functions, such as memory, perception, awareness, thought, language, and consciousness. It is the wrinkly grey outer layer of the brain, common to all mammals. All mammalian cerebral cortexes are similarly made up of six distinct layers of neurons, and are known as the brain’s ‘functional architecture’.

Different regions of the cortex are organised into distinct functions. Researchers have aimed to distinguish these regions and their properties by mapping the brain. Three functionally distinct areas of the brain have been identified: the sensory, motor, and associative areas. The main sensory area of the brain includes the primary visual cortex, known as area ‘V1’.

Retinotopy is the process of visual mapping from the retina to the neurons.

The process of visual mapping from the retina to the neurons is known as retinotopy. In mammals, the visual information encoded on the retina is mapped and organised spatially on the cortex in V1. As well as encoding an image’s spatial organisation, neurons in V1 also encode information such as orientation of edges, spatial frequency, direction of motion, and eye of origin in mammals such as primates and cats. In primates and cats, neurons with similar orientation selectivity are clustered together. In terms of location in the visual field, an image is mirrored in the activation of neurons in the visual cortex. However, the cortex also clusters together cells that code particular image features, such as edge orientation.

“It was identified that mammals have two distinct types of organisation in the visual cortex.”

In recent years, researchers have recognised that rodents do not have such specific image feature organisation as primates and cats (who have what is called a ‘pinwheel organisation’). This means that the information encoded in V1 in rodents is distributed in a less organised fashion (known as ‘salt-and-pepper maps’). It is noted, therefore, that mammals have at least two distinct types of organisation in the visual cortex.

Orientation maps

In a recent paper published in Frontiers in Systems Neuroscience, Professor Michael Ibbotson and Dr Jason Jung from the National Vision Research Institute, Australia, investigated the origins of organisation in the visual cortex in mammals, specifically with regard to orientation maps. Why do certain mammals – such as primates and cats – encode information about the orientation of edges and map them across V1 in an organised way, whereas other mammals – such as rodents and rabbits – have a random distribution of orientation maps across V1?

It was previously hypothesised that rodents do not rely so heavily on vision, and therefore do not possess such complex orientation mapping. Even though they possess orientation-selective neurons in V1, rodents do not develop pinwheel orientation maps, possibly due to a small absolute V1 size. They also lack the acuity and contrast sensitivity seen in primates and cats.

Orientation maps could form in larger mammals because neurons with similar orientation preferences make local connections, whereas small mammals do not require this connection for visual processing (perhaps because their synaptic connections are formed easily without additional organisation, owing to smaller distances linked to a smaller V1 size). However, this hypothesis was disproven in a previous study, as squirrels with similar sized V1s to ferrets and tree shrews do not have robust orientation maps, while the latter do. What functional roles, then, does the orientation map play in visual processing?

Theories to explain differences

Professor Ibbotson and his team performed a comparative survey of various mammals based on their visual characteristics, to identify several theories to explain the differences in visual brains in mammals.

Professor Ibbotson and his team performed comparative studies of numerous mammals.

Firstly, eye divergence was investigated. Eye divergence is the simultaneous outward movement of both eyes away from each other to look at a distant object. It can be quantified by photographing the eyes while an animal is viewing a distant light source. The reflection on the corneas in the optical centre of the eyes, and the distance by which the separation of the pupils exceeds the separation of the images, becomes the eye divergence measurement.

It was hypothesised that mammals with highly diverged eyes (pointing sideways) do not have organised orientation maps. Professor Ibbotson concludes, however, that eye divergence alone cannot be used to explain the differences in orientation maps because some species, such as tree shrews and ferrets, have large eye divergence but possess highly organised orientation maps in V1.

In addition, it was thought that more cells in V1 suggests a higher likelihood of the brain organising itself into columns, or into a specific orientation map structure. Primates and cats have high cortical cell populations, at more than 30 million, and they possess a pinwheel organisation. In contrast, a rabbit has approximately 6 million neurons and possesses a salt-and-pepper organisation. The ferret and tree shrew have 7.6 million and 8 million neurons, respectively, and as stated before, possess a highly organised orientation map. These neuron population numbers may represent the threshold needed to acquire the pinwheel organisation in V1, but further studies with more animals, such as the agouti – a rodent with a large cortex – should be conducted to garner further support for this idea.

Location and visuotopic organisation of marmoset primary visual cortex (V1).

Interestingly, it was shown that species with high densities of axons in V1 (ie, more than 2,000 axons per mm²) in the lateral geniculate nucleus (LGN, a relay centre for the visual pathway, located in the thalamus), have a random orientation map organisation. In contrast, species with low densities of LGN axons, such as primates, carnivores, and scandentia (ie, tree shrews), have pinwheel maps. Species with lower densities of LGN axons have a higher total number of LGN neurons spread out over a larger area and they are also associated with larger numbers of central retinal ganglion cells.

“Different species organise their retinal inputs differently to accommodate their visual needs.”

The centroperipheral ratio

Different species organise their retinal inputs differently to accommodate their visual needs. Rabbits utilise their peripheral vision more, enabling them to consistently look out for predators. Primates have a strong retinal bias towards central vision for high acuity. Professor Ibbotson and Dr Jung used the centroperipheral ratio (CP) – which is acquired by dividing the ganglion cell density in the centre of the retina with the peripheral cell density – to determine if it could explain the differences in orientation map structures.

Their findings show that species with high central retinal densities versus peripheral retina are more likely to possess a pinwheel organised map, as seen in primates. Primates require this feature for very high visual acuity in the central visual field. A greater use of central vision increases the number of retinal inputs to the thalamus, increasing the size of V1 and providing a greater space between thalamic axons, so they can cluster in different cortical areas. This enables the highly organised visual map structure seen in primates.

Based on the comparative study, it was hypothesised that animals with a CP ratio lower than 4 – such as rats, grey squirrels, mice, and rabbits – will possess a random orientation map. If the CP ratio is more than 7, such as in tree shrews (CP<8) and ferrets (CP= 11), their orientation map will be highly organised into pinwheel structures.

At present, the literature lacks the study of orientation maps in animals that are intermediate between small rodents and primates, such as the large rodent agouti, fruit bats, sheep, and marsupials. Their visual pathways and brain structures are also not fully elucidated to allow for specific estimation regarding structural and visual properties. A broader study into these animals may reveal the importance of orientation maps in V1 in mammals, and shed light on this mystery of comparative neuroscience. Studying these interspecies variations can also highlight the multiple factors that control retinal design and cortical map structures.

How do you intend to gather more information about the intermediate species to further support your hypotheses?
We are at present conducting research on the visual cortex of several marsupial species. As these animals split from eutherian mammals 160 million years ago, if they have pinwheel structures it suggests that such cortical organisation was present prior to the split. This will give important genetic clues to the origins of cortical organisation. We also plan to expand the range of eutherian mammals that have been studied to see what type of cortical organisation they have, and how it fits with their visual environments.

References

Ibbotson, M, and Jung, YJ, (2020). Origins of functional organization in the visual cortex. Frontiers in Systems Neuroscience, 14, 10. doi.org/10.3389/fnsys.2020.00010
Van Hooser, SD, Heimel, JAF, Chung, S, Nelson, SB, et al, (2005). Orientation selectivity without orientation maps in visual cortex of a highly visual mammal. The Journal of Neuroscience, 25 (1), 19–28.k doi.org/10.1523/JNEUROSCI.4042-04.2005

DOI
10.26904/RF-138-1821809224

Research Objectives

Professor Michael R Ibbotson studies variations in the orientation maps in the visual cortex of mammalian brains.

Funding

Australian Research Council Centre of Excellence for Integrative Brain Function

Collaborators

Dr Jason Jung

Bio

Professor Michael Ibbotson has a comparative interest in how natural visual systems generate behaviour and perception. He investigates the influence of electrical stimulation to develop visual prosthetics, aimed at returning sight to the blind. In the mammalian visual cortex, he studies receptive field mechanisms, plasticity, and functional maps.