* The **retina** projects to multiple areas in the brain. Each area is specialized for different functions
* Dorsal **lateral geniculate nucleus** (dLGN)– located in the thalamus- receives visual info from retina and sends it to the visual cortex. Most important visual projection with respect to visual perception
* **Pretectum**– located at midbrain-thalamus boundary. Responsible for pupillary light reflex
* **Superior colliculus**– in midbrain, coordinates head and eye movements
* **Suprachiasmatic nucleus**– hypothalamus, involved in day/night cycles
The output neurons of the eye-- the retinal ganglion cells-- form synaptic connections in two visual centers the lateral geniculate nucleus and the superior colliculus.
And the geniculate neurons have in turn formed synaptic connections with the visual cortex, thus forming the basic visual pathway from the eye to the cerebral cortex.
* **Optic disc, optic nerve**- All the retinal ganglion cell (RGC) axons exit the eye at the optic disk (results in a blind spot) and form a big myelinated nerve called optic nerve (cranial nerve II).
* **Optic chiasm**- where the optic nerve enters the brain, at the base of the hypothalamus.
* **Optic radiation**- portion of the internal capsule (connection between thalamus and cortex) containing the axons from dLGN that project to the visual cortex
* atropine blocks contraction of the **circular **pupillary constrictor muscles muscle (classified as an anticholinergic drug) by being a competitive inverse agonist for muscarinic ACh receptors
* allows the radial pupillary dilator muscle to contract and dilate the pupil
* Question: Where is the site of injury if shining a light into the left eye causes both eyes to constrict but shining light into the right eye does not cause either eye to constrict? <!-- .element: class="fragment fade-in" -->
Neighboring retinal ganglion cells in the eye detect changes in contrast from similar portions of the visual field, thus forming a 2D map of visual space in the retina. This spatial representation of objects in the retina is then projected onto -->multiple down stream visual areas, so that maps of retinal topography, or retinotopy, are maintained at multiple levels in the visual system.
Other visual functional organization that is present at birth includes maps of ocular dominance, where the responses of neuronal groups is dominated by that of one eye or the other and orientation selectivity where the responses of neighboring neurons is dominated by high contrast edges of particular orientation.
* Objects in the left visual field are seen by the nasal retina of the left eye and the temporal retina of the right eye
* Objects on extreme periphery are seen only by the nasal retina on that side
* Nasal retinal derived axons cross the midline at the optic chiasm (contra lateral) and temporal retinal axons do not cross at the chiasm (ipsilateral)
* Images in the left visual field project onto the nasal retina of the left eye and the temporal retina of the right eye. These go to the same side of the brain. Therefore the left visual field is mapped onto the right side of the brain
* The visual map is maintained all the way to V1. The two halves of the visual fields only merge after getting connections from the other half through the corpus callosum
<figure><figcaption class="big">binocular vision (overlapped color in middle)</figcaption><img src="figs/Neuroscience5e-Fig-12.04-0_d0d4f01.jpg" height="500px"><figcaption>Neuroscience 5e Fig. 12.4</figcaption></figure>
* Contains 6 layers, that are specific with respect to eye (ipsi vs contra) and with respect to type of ganglion cell— magnocellular (detects gross shape and movement) and parvocellular (form and color)
* Layers align in order to align visual fields
* Each dLGN receives input from 1 or 2 RGCs therefore like RGCs there also have center-surround responses that are either on or off
* The projections from the retinal ganglion cells maintain the field of view as it was seen - this is called a retinotopic map. The LGN contains 6 layers of cell bodies; each layer receives input from only one eye. The two most ventral layers receive M (magno) ganglion cell inputs, while the other 4 receive P (parvo) inputs
* Superior and inferior visual fields take different routes to the visual cortex. Meyer’s loop, where superior axons diverge and go into temporal lobe before going to occipital lobe
*Reasons for macular sparing not known. Has been proposed that there is overlap in the pattern of crossed and uncrossed ganglion cells that provide central vision*
* The visual cortex is layered. Each layer has stereotypical inputs and outputs. LGN projects to layer 4. Output layer is layer 5.
* Each column of neurons in the vertical plane typically respond to the same part of the visual field and the same orientation.
* Neurons in the horizontal plane respond to neighboring areas of the visual field and change orientation preferences that repeats each milimeter or so.
* Neurons in layer 4 respond to just one eye or the other (monocular cells) but other layers have neurons that can respond from either eye. This sets up ocular dominance columns in the cortex.
Note:
Now let’s go over the structural and functional organization of visual neocortex
## Neurons in the primary visual cortex respond selectively to oriented edges
* David Hubel and Torsten Wiesel— measured responses of neurons in visual cortex. Found not center-surround like RGCs and LGN neurons but found that they respond to bars or lines but only of a particular orientation
* Two types of cells:
* Simple, respond to stimulus only if matches orientation. Spots of light don’t do much, bars or lines make them fire. They also have surround inhibition. Receptive fields can be generated by having 3-4 LGN neurons innervate one simple cell
* Complex cells- bigger receptive fields, not strongly orientation selective, no clear on or off zones, detect movement
<img src="figs/Neuroscience5e-Fig-12.08-2R_05d7b3b.jpg" height="350px"><figcaption>Neuroscience 5e Fig. 12.8</figcaption>
</div>
<div><figcaption class="big">Oriention tuning curve for a single V1 neuron</figcaption><img src="figs/Neuroscience5e-Fig-12.08-3R_8799a25.jpg" height="400px"><figcaption>Neuroscience 5e Fig. 12.8</figcaption></div>
<figure><figcaption class="big">Inputs from several RGC center-surround RFs may be integrated to create a oriented edge RF for V1 neurons</figcaption><img src="figs/hubel-EBB-p74-simple-cell-rf_2_copy_2_3dd95f6.jpg" height="400px"><figcaption>D. Hubel. *Eye, Brain, and Vision* p. 74</figcaption></figure>
complex cells are also all orientation selective and retinotopic, but need moving lines. Do not react to stationary stimuli. Most common functional cell type in striate cortex, maybe 75% of population. Glass slide in field of view was first stimulus.
<figure><img src="figs/Neuroscience5e-Fig-12.12-0_copy_118f490.jpg" height="400px"><figcaption>Neuroscience 5e Fig. 12.12, D. Fitzpatrick (left), Ohki et al. *Nature* 2006 (right)</figcaption></figure>
<div><figcaption class="big">Histological stain of thalamocortical afferents, section through L4 of V1</figcaption><img src="figs/hubel-wiesel-levay-1976_9574797.png" height="400px"><figcaption>Hubel, Wiesel, and Levay 1976</figcaption></div>
If we were to peer at layer 4 only and perform a histological procedure that labels thalamocortical inputs from only one eye we would see a pattern like this in primate cortex, resembling ocular dominance bands or stripes.
* In human retina there are three main types of retinal ganglion cells, called M, P , and K types. M and P types best characterized
* M cells– are bigger, have larger receptive fields, faster conduction velocities, and respond transiently to visual stimulation. P cells smaller, respond in a sustained fashion
* P cells– respond to color. This is because their center and surround are from different cones
* K cells– less understood, but known to transmit some aspects of color vision such information from short wavelength cones
* M cells do not respond well to color because center and surround are from the same type of cones
* M, P, and K RGCs go to different layers in the LGN which in turn project to different layers in V1
“the brain is a complex of widely and reciprocally interconnected systems and that the dynamic interplay of neural activity within and between systems is the very essence of brain function” (V. Mountcastle). And indeed if you look at this—> anatomical wiring diagram for different visual areas represented by different colors you will notice that we use an organized constellation of brain regions to process and route different types of visual information
: long strip of cortex in ventral temporal lobe, tracking along hippocampal gyrus in rostral-caudal extent, but separate from entorhinal or parahippocampal cortex
* However some patients can still "guess" what an object is. Implies that there are other projections from eye to brain (superior colliculus) that can somehow compensate for loss of V1