Source: MCB C61, UC Berkeley, Ch. 14
Tags: retina, fovea, rods, cones, rhodopsin, cone opsins, retinal, vitamin A, beta-carotene, photoisomerisation, cGMP, GPCR cascade, bipolar cells, ganglion cells, horizontal cells, amacrine cells, optic nerve, blind spot, LGN, visual cortex, V1, V4, V5, scotoma, hemianopia, achromatopsia, prosopagnosia, agnosia, blindsight, superior colliculus, receptive field, contralateral connectivity, trichromatic colour vision
Vision begins when photons are absorbed by retinal molecules in rod and cone photoreceptor cells, triggering a GPCR intracellular cascade with enormous signal amplification. Information flows through layers of retinal cells (photoreceptors → bipolar → ganglion), exits via the optic nerve, crosses at the optic chiasm, and reaches the visual cortex through the lateral geniculate nucleus (LGN). Different cortical areas process different aspects of vision (edges in V1, colour in V4, motion in V5), and lesions in each area produce distinct clinical conditions.
Retina
A structure at the back of the eye consisting of light-sensitive photoreceptor cells and several layers of interconnected nerve cells. Heavily vascularised (photoreceptors require substantial glucose and oxygen). Named from Latin rete ("net").
Fovea
A small region at the centre of the retina where light from the centre of the visual field is focused. Has the highest density of photoreceptors, producing the highest visual acuity. Contains both rods and cones.
Rods
Rod-shaped photoreceptor cells distributed throughout most of the retina. Very numerous (~100 million per retina). Sensitive to even small amounts of light. Operate in dim light or at night. Little to no input in colour perception.
Cones
Cone-shaped photoreceptor cells concentrated at the fovea. Fewer in number (~5 million per retina). Respond to brighter light and mediate colour vision. Three types: S (short-wavelength, blue), M (medium-wavelength, green), and L (long-wavelength, red).
Trichromatic colour vision
Human colour vision based on comparing the activation levels of three cone types (S, M, L). Colour is a mental experience related to how the nervous system responds to electromagnetic radiation of different energies.
Rhodopsin / cone opsins
Photoreceptor proteins (~350 amino acids each) embedded in the lipid bilayer membrane of the outer segments of rods and cones. They absorb light and initiate the transformation of light energy into a neural signal.
Retinal
A molecule derived from vitamin A (retinol) or beta-carotene that is covalently attached to opsin proteins. It is the light-absorbing component. The body cannot make retinal from scratch; it must obtain vitamin A or carotenoids from the diet.
Beta-carotene
The most abundant chemical precursor to retinal and retinol in nature. Gives carrots their orange colour. Found widely in vegetables. The body converts one beta-carotene molecule into two retinal molecules.
Photoisomerisation of retinal
When 11-cis retinal absorbs a photon, it straightens into the all-trans form. This shape change activates the opsin protein and initiates the GPCR intracellular cascade.
cGMP cascade (GPCR intracellular cascade in vision)
Activated opsin binds an intracellular G protein → G protein activates cGMP phosphodiesterase → this enzyme hydrolyses cGMP to noncyclic GMP → reduced cGMP causes ion channels to close → membrane potential changes → altered NT release at the photoreceptor synapse. One photon can decrease more than 10,000 cGMP molecules, producing enormous amplification.
Bipolar cells
Retinal interneurons that receive synaptic input from rods and cones and transmit it to ganglion cells.
Ganglion cells
Retinal neurons whose axons bundle together to form the optic nerve. Each retina has about 1 million ganglion cells. They integrate input from multiple photoreceptors and require coincidence of 5–10 photons before firing an action potential.
Horizontal cells / amacrine cells
Retinal interneurons in the same layer as bipolar cells. They contribute to lateral integration and processing of visual information within the retina.
Blind spot (scotoma)
The region of the retina where ganglion cell axons converge to form the optic nerve and exit the eye. No photoreceptors are present here. Usually unnoticed because the blind spots of the two eyes cover different regions, allowing each to fill in for the other.
Optic nerve
The bundle of ~1 million ganglion cell axons that exits the eye at the blind spot and carries visual information to the brain.
Optic chiasm
The point where optic nerve fibres partially cross. Information from the right visual field goes to the left hemisphere and vice versa.
Contralateral connectivity
The crossing-over of sensory information between the spatial environment and the opposite side of the brain.
LGN (lateral geniculate nucleus)
A relay structure in the thalamus. 90% of optic nerve axons arrive here after the optic chiasm. Right visual field information goes to left LGN, and left visual field to right LGN. LGN cells send axons to V1 in the occipital lobe.
Visual cortex areas (V1, V2, V3, V4, V5)
Regions in the occipital lobe that process different aspects of visual information. All are highly interconnected with bidirectional axonal connections between every pair of areas and back to the LGN.
V1 (primary visual cortex): responds to edges of contrast oriented or moving in a particular direction
V4: responds to specific colours
V5: responds to movement, speed, and direction
Each area contains a topographic map of visual space.
Superior colliculus
A paired midbrain structure. Superficial layers receive direct retinal input and respond to visual stimuli. Deeper layers respond to multiple sensory modalities and contain motor neurons for eye movements. Directs behavioural responses toward specific points in body-centred space. Contains a topographic retinotopic map.
Receptive field
The region of sensory space in which a stimulus will modify the firing of a given neuron. All retinal cells and brain neurons receiving retinal signals have visual receptive fields.
Retinal achromatopsia
A genetic or developmental condition resulting in loss of all functional cone cells. Affected individuals see only in shades of black, white, and grey but may appreciate subtle gradations of contrast, shadow, and texture more than those with normal vision.
Cortical achromatopsia
Caused by a lesion in V4. Disrupts colour vision in affected regions of visual space. Severity ranges from washed-out colour to complete loss of colour awareness. Distinct from retinal achromatopsia because the cause is cortical, not retinal.
Scotoma (cortical)
A region of visual field loss resulting from a lesion in V1. Small lesions may go unnoticed.
Hemianopia
Loss of vision from half of the visual field, resulting from a large lesion in V1. Contralateral to the side of the brain with the lesion.
Motion blindness (akinetopsia)
Caused by a lesion in V5. The person is unaware of movement in some regions of visual space. The world appears as a series of static snapshots. Very disabling.
Prosopagnosia
Difficulty or complete inability to recognise faces, with other aspects of visual perception remaining intact. Caused by a lesion in the inferior and medial temporal lobe where face-selective neurons reside. A specific type of agnosia.
Agnosia
Difficulty recognising visual objects. The person can identify individual details but struggles to integrate them into a meaningful whole. Associated with lesions where the occipital, temporal, and parietal lobes converge.
Blindsight
A condition in which a person who is clinically blind can respond to visual stimuli they cannot consciously see (e.g. guessing orientation above chance). Thought to involve the ~10% of ganglion cell axons that go to the superior colliculus rather than V1.
Cephalopod eye vs. vertebrate eye
The octopus eye has photoreceptors on top and ganglion cells on the bottom, so there is no blind spot. The vertebrate eye has the arrangement inverted (ganglion cells on top, photoreceptors behind), which creates the blind spot.
Outer segments contain rhodopsin (rods) or cone opsins (cones), embedded in lipid bilayer membranes
Inner segments contain the nucleus, mitochondria, and other cellular machinery
Connected by a connecting cilium
The retinal molecule:
Covalently bonded to a lysine amino acid within the opsin protein via a nitrogen atom
In its resting state, retinal is in the 11-cis form (kinked/bent)
Light absorption triggers photoisomerisation: 11-cis retinal straightens to all-trans retinal
This shape change activates the opsin protein
The body obtains retinal from vitamin A (retinol) or beta-carotene from the diet
Different amino acid sequences in the various opsin proteins create different electronic environments around the attached retinal, shifting the wavelength of maximum light absorption. This is how S, M, and L cones respond to different wavelength ranges.
This cascade is the reason we can see in very dim light:
Photon absorbed by 11-cis retinal → isomerises to all-trans → opsin protein activated
Activated opsin binds an intracellular G protein → G protein activates cGMP phosphodiesterase
cGMP phosphodiesterase hydrolyses cGMP (cyclic guanosine monophosphate) to noncyclic GMP
cGMP normally keeps certain ion channels open; as cGMP decreases, channels close
Channel closure changes membrane potential and thus the amount of NT released at the photoreceptor synapse
Amplification: one photon can produce a decrease of more than 10,000 cGMP molecules, closing many ion channels
Three major cell layers: photoreceptors → bipolar cells → ganglion cells
Additional interneurons: horizontal cells and amacrine cells (in the bipolar cell layer) contribute to lateral integration.
Light enters the retina from the "top" (ganglion side) and passes through all layers before reaching photoreceptor outer segments. At the fovea, the upper cell layers are folded back to allow unobstructed light access to photoreceptors.
Ganglion cells require coincidence of 5–10 photons from one or more photoreceptors before generating an action potential, filtering out noise.
~100 million rods vs. ~5 million cones per retina
Cones are concentrated at the fovea; colour perception is best when an object's image falls on the fovea
Rods dominate the retinal periphery; peripheral colour perception is poor
Ganglion cell axons form the optic nerve
Optic nerves from both eyes partially cross at the optic chiasm (contralateral connectivity)
90% of axons go to the LGN in the thalamus; ~10% go to the superior colliculus in the midbrain
LGN cells project to V1 (primary visual cortex) in the occipital lobe
V1 distributes to V2, V3, V4, V5 with complete bidirectional interconnectivity
V1: edges, contrast orientation, direction of movement
V4: colour
V5: motion, speed, direction
Each area maintains a topographic map of visual space
V1 lesion → scotoma (localised blindness in a region of visual space) or hemianopia (half-field blindness if large enough); contralateral to lesion side
V4 lesion → cortical achromatopsia (disrupted colour vision, ranging from faded to total loss)
V5 lesion → motion blindness/akinetopsia (world as a series of snapshots)
Inferior/medial temporal lobe lesion → prosopagnosia (cannot recognise faces)
Occipital-temporal-parietal junction lesion → agnosia (cannot integrate details into meaningful objects)
Blindsight: the 10% pathway to the superior colliculus may allow unconscious visual responses when V1 is damaged
⚠️ Know the full GPCR cascade in detail: photon → 11-cis to all-trans retinal → opsin activated → G protein → cGMP phosphodiesterase → cGMP decreases → ion channels close → membrane potential changes → NT release changes.
⚠️ One photon can decrease over 10,000 cGMP molecules. This amplification is why rods can detect very dim light.
⚠️ The retina processes information before sending it to the brain. Ganglion cells integrate and filter signals, requiring 5–10 photon coincidences.
⚠️ Match each lesion to its clinical condition: V1 = scotoma/hemianopia, V4 = cortical achromatopsia, V5 = motion blindness, temporal lobe = prosopagnosia.
⚠️ Retinal achromatopsia (no cones, retinal cause) vs. cortical achromatopsia (V4 lesion, cortical cause). Know the difference.
⚠️ Contralateral connectivity: right visual field → left LGN → left V1 (and vice versa).
⚠️ The octopus eye has no blind spot because its photoreceptors face the light. Vertebrate eyes have an inverted arrangement.
Q: Describe the full sequence of the phototransduction cascade starting from a photon hitting rhodopsin.
A: Photon is absorbed by 11-cis retinal in rhodopsin → retinal isomerises to all-trans form → opsin protein changes shape and activates → activated opsin binds a G protein → G protein activates cGMP phosphodiesterase → cGMP is hydrolysed to GMP → decreased cGMP causes ion channels to close → membrane potential changes → amount of NT released at the photoreceptor synapse changes.
Q: Why can rods detect very dim light?
A: Because of enormous signal amplification in the GPCR cascade. A single photon can trigger the breakdown of more than 10,000 cGMP molecules, closing many ion channels and producing a significant change in the cell's electrical state.
Q: What is the difference between retinal achromatopsia and cortical achromatopsia?
A: Retinal achromatopsia is a genetic/developmental loss of all functional cone cells (retinal cause). Cortical achromatopsia is caused by a lesion in V4 (cortical cause) and may affect only certain regions of visual space.
Q: What is blindsight, and what neural pathway is thought to explain it?
A: Blindsight is the ability to respond to visual stimuli without conscious awareness of seeing them. It is thought to be mediated by the ~10% of ganglion cell axons that project to the superior colliculus in the midbrain, bypassing the damaged V1.
Q: How does the vertebrate eye differ from the cephalopod eye in terms of blind spot?
A: In the vertebrate eye, photoreceptors are behind the other retinal layers, and ganglion cell axons must pass through the retina to form the optic nerve, creating a blind spot. In the cephalopod eye, photoreceptors face the light directly, so no blind spot exists.
Q: A patient has a large lesion in the right V1. What visual deficit would you expect?
A: Left hemianopia: loss of vision from the left half of the visual field, because visual information crosses contralaterally (left visual field → right hemisphere).
Q: What are the three types of human cone cells, and what do they detect?
A: S cones (short-wavelength, blue), M cones (medium-wavelength, green), and L cones (long-wavelength, red). Together they enable trichromatic colour vision.
electromagnetic spectrum, visible light, wavelength, frequency, photon, retina, fovea, rods, cones, S cones, M cones, L cones, trichromatic colour vision, rhodopsin, cone opsins, retinal, 11-cis retinal, all-trans retinal, photoisomerisation, vitamin A, retinol, beta-carotene, cGMP, cGMP phosphodiesterase, G protein, GPCR cascade, amplification, bipolar cells, ganglion cells, horizontal cells, amacrine cells, optic nerve, blind spot, optic chiasm, contralateral connectivity, LGN, lateral geniculate nucleus, thalamus, visual cortex, V1, V2, V3, V4, V5, topographic map, retinotopic map, superior colliculus, receptive field, scotoma, hemianopia, retinal achromatopsia, cortical achromatopsia, motion blindness, akinetopsia, prosopagnosia, agnosia, blindsight, cephalopod eye, octopus eye