Light, the eye, and the retina Flashcards
Why study vision
Matching
Magnitude estimation and production
Detection and discrimination task
Vision is a dominant sense in humans: cortex that is predominantly visual in function occupies about 27%, by comparison, about 8% is predominantly auditory, 7% somatosensory, and 7% motor (the remaining half, 51%, includes major domains associated with cognition, emotion, and language) (Van Essen in the Visual Neurosciences 2004).
Vision is one of the most widely studied part of the brain: it serves as a model system to understand brain function. As a result, we have a good understanding of the structure and function of the visual system: an example is the detailed connectivity diagram of the macaque visual system (Felleman and Van Essen, Cereb Ctx 1991)
The physics of light
Vision begins with light emitted or reflected by objects in the world. Light is a form of energy propagated by electromagnetic waves, travelling at high velocity (300km/ms), in discrete packets called quanta or photons.
Only a small range of light is visible to humans, ranging from wavelength of ~400 nanometer (violet) to 700 nm (red).
The visual system analyses four components of light: 1) wavelength 2) intensity and how these vary In 3) space and 4) time.
Measuring light
Matching
Magnitude estimation and production Detection and discrimination
The candela (cd) measures rate of light emission (a 60W bulb emits ~100cd). The amount of light received by object per unit area is called illuminance (measured in lux: the degree of illumination of a surface 1m away from a source of 1 candela radiating uniformly in all directions) (full sunlight may provide 100,000 lux)
Objects in world scatter back light: the emitted amount of light per unit area is called luminance (units: cd / m^2)
The ratio of luminance to illuminance is called albedo
The visual system can adapt to a wide range of luminances, and perception scales to the overall illumination. In general, the visual system cares about contrast and less about absolute luminance levels. Phototopic vision is vision under well lit conditions when rods have saturated reliant on cone photoreceptors, scotopic vision is vision during relatively dim lighting reliant on rods (below the threshold for activation of cone photoreceptors), while mesoscopic refers to vision in intermediate lighting conditions reliant on both rods and cones
Photopic and scotopic vision: adaptation to dark
An example of how the visual system adapts to changing luminance is given by dark adaptation experiments, where a subject moves from daylight to a darkened room and absolute detection thresholds are measured demonstrating increased sensitivity to light when the subject adapts. Two components can be observed, an initial fast adaptation that levels off within ~8min that reflects the adaptation of cone photoreceptors and a slower process (over the course of another 30min) that reflects the adaptation of rod photoreceptors (also associated with a shift in sensitivity towards blue light).
Light intensity and contrast
The visual system cares more about contrast than relative luminance. Contrast between 2 surfaces with absolute luminances L1 and L2 can be defined as: C=(L1-L2)/(L1+L2).
The advantage is that contrast is a property of object, not of the lighting conditions.
For example: L1=2 and L2=1 > contrast = 1/3
if luminances increase by 1000, contrast stays the same: 1000/3000 = 1/3
The eye
The eye is used to form a spatial image of light and dark. On the outside of the eye we see the sclera (or white of the eye), the protective outer layer of the eye, and the coloured iris with the pupil as its dark aperture.
Light is focussed by the cornea and lens on the retina through the pupil, creating an inverted image.
Retinal coordinates are typically expressed in angles. One degree is approximately the size of the nail of your thumb at arm’s length, corresponding to ~300um on the retina.
Optics of the eye
When light enters a region with a higher refractive index (e.g. going from air to water), it is bent towards the normal by an amount that depends on the refractive index. If a surface is curved, the further out parts bend parallel rays more, and inner parts less. The power of a refractive surface is described by the reciprocal of the focal length (m) with units dioptres D. If power is measured within a refractive medium, the power should be multiplied by the refractive index.
The distance of the cornea to the retina is about 24mm which gives a power of approximately 43D * 1.34 = 57D of which 48D is due to cornea and only 19D to lens. The cornea is powerful because of the large change of refractive index going from from air (1.0008) to the cornea (1.34)
So, the lens contributes less than the cornea to the refractive power of the eye (contrary to popular belief), but it has an important special function: it can change shape and thereby fine the eye’s focal length, a process called accommodation. The process relies on elastic and highly flexible and transparent cells in the lens: when the ciliary muscle contracts, it relieves pressure on the lens that then becomes rounder with shorter focal length. The range of accommodation are the near and far points where objects can be found. The near point in a young person is around 80mm but this gets further away in old people.
Refractive error
Emmetropia is the state in which an object at infinity is in sharp focus with the ciliary muscle in a relaxed state. Myopia is the case of short sight, where distant objects appear blurry: this can be corrected with a negative lens. Hypermetropia is called long sight: where close objects appear blurry and can be corrected using a positive lens.
A second type of refractive error is caused by the cornea shape being slightly different in different meridians. This will focus a horizontal line at a different focal length from a vertical line. This is revealed in an optician test using a fan pattern with lines ranging from horizontal to vertical. Corrective lenses can be cut from cylinders instead of spheres to adjust for this.
In addition to refractive errors, there are also other optical factors that contribute to errors of image formation. This includes chromatic aberration, spherical aberrations, glare and diffraction.
Chromatic aberration,Spherical aberration
Tips: color, shape of lens
Chromatic aberration: the refractive index decreases with the wavelength of light, so that blue light is refracted more than red causing dispersion of wavelengths. The consequence can be relatively large: for example, visual acuity is improved by ~25% in monochromatic yellow.light. The eye mitigates chromatic aberration in two ways: a) by having yellow pigment over the fovea (the area of the retina with the finest spatial vision) reducing the blue component of light 2) by having very few blue cones in the centre of the fovea.
Spherical aberration: The ideal shape for a lens is an ellipsoid because a sphere bends light too much when you go out to the edges. The eye has some adaptation to reduce spherical aberrations: a) the cornea is not exactly spherical but approaches an ellipsoid shape b) the refractive index of the lens is not constant to partially cancel out some of the bending of light rays
Glare&Diffraction
Glare: when light enters the eye it scatters, in the cornea and lens but also by bouncing off the retina. . This effectively superimposes background on image (with illuminance ~10% of mean illuminance of the retina) target whose actual contrast is 100%, scatter reduces contrast to ~90%. because of progressive opacity of lens, glare gets worse as you get older
Diffraction: caused by pupil. Whenever a light passes through a restricted aperture it spreads out, the smaller the aperture the worse the effect.
Spatial acuity
Spatial acuity is the ability to resolve objects in visual space. The point spread function (PSF) describes the response of an optical system to a point object, quantifying the blur caused by the optical system. The PSF determines what we can and cannot resolve (consider the case with the two distributions in c where we cannot see any ‘dip’ in between the two distributions.
Luminance contrast affects mainly the detection but also the ability to resolve spatial featues.
The interaction between resolution and contrast can be investigated with grating patterns. A grating is simply a regular pattern of stripes (called a square grating if the stripes are uniformly black and white or called a sinusoidal gratings if the intensity profile is sinusoidal). The spatial frequency is the number of black/white or sine cycles that fit within a visual degree (for a example a pattern of 1deg black and 1deg white bars would have 100% contrast and a spatial frequency of 0.5 cycles per degree.
If you plot the threshold contrast as function of spatial frequency, you find that the threshold contrast increases sharply for higher spatial frequencies, because the PSF blurring has a much bigger effect on reducing the perceived contrast of higher spatial frequency patterns (there is also an increased threshold for really low spatial frequency which will be discussed later in the context of lateral inhibition).
At the optician, acuity is usually tested in a simpler manner, by using Snellen or Landolt charts.
The pupil
Size regulation, benefit of each size
The pupil plays an important role in regulating visual actuity and optical properties of the eye.
The size of the pupil in the eye is regulated by two muscles:
1) A constrictor muscle, the sphincter pupillae, that lies circumferentially round the iris and under control of the parasympathetic system. 2) A dilator muscle that lies radially on the eye controlled by the sympathetic system.
Light causes constriction through parasympathetic route from ciliary ganglion, in turn activated by the Edinger–Westphal nucleus in brainstem, close to the oculomotor nucleus, which receives sensory information about overall light level from neurons in pretectum, which themselves receive fibres from the optic nerve.
Vision with a smaller and larger pupil allows for different benefits:
Smaller pupil: Increases depth of field, minimizes optical aberrations & minimizes glare
Larger pupil: Receives more light & minimizes diffraction effects
An overall optimum can be found in between these two extremes.
The interaction between resolution and contrast can be investigated with grating patterns.
The interaction between resolution and contrast can be investigated with grating patterns. A grating is simply a regular pattern of stripes (called a square grating if the stripes are uniformly black and white or called a sinusoidal gratings if the intensity profile is sinusoidal). The spatial frequency is the number of black/white or sine cycles that fit within a visual degree (for a example a pattern of 1deg black and 1deg white bars would have 100% contrast and a spatial frequency of 0.5 cycles per degree. If you plot the threshold contrast as function of spatial frequency, you find that the threshold contrast increases sharply for higher spatial frequencies, because the PSF blurring has a much bigger effect on reducing the perceived contrast of higher spatial frequency patterns (there is also an increased threshold for really low spatial frequency which will be discussed later in the context of lateral inhibition).
At the optician, acuity is usually tested in a simpler manner, by using Snellen or Landolt charts.
The retina
Visual acuity is not just a function of the optics of the eye but depends on the transformation of the neural signals travelling from the photoreceptors in the eye to the brain. One example is that there is no one-to-one mapping of photoreceptors on ganglion and a high degree of convergence of photoreceptors on bipolars and eventually ganglion cells, causing blurring of spatial resolution).
The retina consists of 5 layers
The retina consists of 5 layers (3 nuclear layers containing cell bodies separated by 2 plexiform layers containing neural processes, dendrites and axons.
- The outer nuclear layer (farthest from the centre of the eye) contains the cell bodies of the rod and cone photoreceptors.
- The inner nuclear layer contains the cell bodies of the bipolar, horizontal and amacrine cells
- The ganglion cell layer contains the cell bodies of the ganglion cells whose axons form the optic nerve connecting the retina with the rest of the brain.
The fovea (‘pit’) is the part of the retina with the highest visual acuity. Neural circuitry is displaced in order to reduce light scatter in this location.
There are two two types of photoreceptors; rods and cones (blue or S, green or M and red or L cones). The outer segment of the photoreceptors is where phototransduction takes place (see next lecture). The cones are mainly concentrated in the central parts of the visual fields, while rods are more concentrated in the peripheral part.
In simplified terms, the retinal circuit is organized as follows:
Cone photoreceptors synapse onto bipolar cells
Bipolar cells synapse onto ganglion cells (two major types of ganglion cells are ON and OFF cells)
Axons of ganglion cells exit the eye via the optic nerve
Horizontal and amacrine cells make lateral connections in the retina
Rod photoreceptors synapse onto rod bipolar cells
Rod bipolar cells do not directly synapse onto ganglion cells but take a detour via cone bipolar cells, then onto ganglion cells
There are two two types of photoreceptors
There are two two types of photoreceptors; rods and cones (blue or S, green or M and red or L cones). The outer segment of the photoreceptors is where phototransduction takes place (see next lecture). The cones are mainly concentrated in the central parts of the visual fields, while rods are more concentrated in the peripheral part.
In simplified terms, the retinal circuit is organized as follows:
Cone photoreceptors synapse onto bipolar cells
Bipolar cells synapse onto ganglion cells (two major types of ganglion cells are ON and OFF cells)
Axons of ganglion cells exit the eye via the optic nerve
Horizontal and amacrine cells make lateral connections in the retina
Rod photoreceptors synapse onto rod bipolar cells
Rod bipolar cells do not directly synapse onto ganglion cells but take a detour via cone bipolar cells, then onto ganglion cells
