An ON-center retinal ganglion cell is tested with three stimuli: (A) uniform bright light over its entire receptive field, (B) a small bright spot covering only the center, (C) a small bright spot covering only the surround. Which produces the strongest response?
AA — the cell responds to overall brightness, so more light is always better
BB — the center-surround organization means the cell responds best to a bright spot in its center with no surround stimulation
CC — the surround region is larger and therefore drives the cell more strongly
DA and B equally — spot and diffuse illumination produce identical responses
ON-center cells are designed to encode contrast, not absolute brightness. The surround exerts inhibition via horizontal cells; illuminating it suppresses the cell's response. A small bright spot on the center with a dark surround provides maximum excitation from the center with no inhibitory cancellation from the surround — the optimal stimulus. Uniform illumination (stimulus A) is actually a poor driver because the surround inhibition partially cancels the center excitation. This center-surround organization explains why the visual system is so sensitive to edges and borders while being relatively indifferent to overall illumination levels.
Question 2 Multiple Choice
How does simple cell orientation selectivity in V1 arise from inputs with circular, non-oriented receptive fields?
AV1 cells receive orientation information directly from specialized photoreceptors tuned to different angles
BA simple cell receives excitatory input from a row of LGN neurons whose circular center-surround fields are spatially aligned; an edge at that orientation simultaneously activates all of them
COrientation selectivity is created by feedback connections from higher visual areas back to V1
DV1 cells inherit orientation selectivity from the LGN, which already processes oriented edges
This is Hubel and Wiesel's explanation: simple cells in V1 receive convergent input from multiple LGN neurons whose circular center-surround fields happen to be arranged in a straight line across the visual field. An edge oriented along that line simultaneously falls on the center of each LGN neuron, producing a strong combined response. An edge perpendicular to the line stimulates some centers and some surrounds, producing a weak net response. LGN neurons themselves have circular, non-oriented receptive fields — orientation selectivity is an emergent property of the convergent wiring in V1.
Question 3 True / False
The retina functions primarily as a passive sensor that transmits a faithful pixel-by-pixel representation of the visual image to the brain for later processing.
TTrue
FFalse
Answer: False
The retina is a piece of brain tissue that performs substantial computation before any signal leaves the eye. Center-surround receptive fields — set up by lateral inhibition from horizontal and amacrine cells — already encode local contrast rather than raw brightness. The retina doesn't send a picture; it sends a spatial map emphasizing edges, borders, and changes in illumination. By the time signals leave the optic nerve, significant feature extraction has already occurred. The analogy to a camera sensor is misleading: a camera captures raw intensity values, while the retina transmits a processed representation of spatial contrast.
Question 4 True / False
After the partial decussation at the optic chiasm, each cerebral hemisphere primarily receives visual input from the opposite visual hemifield (from both eyes), rather than from the opposite eye.
TTrue
FFalse
Answer: True
At the optic chiasm, axons from the nasal half of each retina (which sees the temporal visual field) cross to the opposite hemisphere, while axons from the temporal half of each retina (which sees the nasal visual field) stay ipsilateral. The result is that the left hemisphere receives all input from the right visual field — from the temporal retina of the right eye and the nasal retina of the left eye — and vice versa. The organizing principle is visual hemifield, not eye. This is why a lesion in the right primary visual cortex causes blindness in the left visual field of both eyes.
Question 5 Short Answer
Why do retinal ganglion cells encode contrast through center-surround receptive fields rather than simply responding to absolute light intensity, and why is this computationally useful?
Think about your answer, then reveal below.
Model answer: Absolute light intensity varies enormously with overall illumination (a white page looks very different in sunlight vs. dim room), but the information needed for object recognition lies in spatial contrasts — the differences between adjacent regions that mark edges, borders, and surfaces. Center-surround organization, created by lateral inhibition from horizontal cells, makes ganglion cells selective for local contrast: they respond strongly when the center and surround receive different illumination levels (as at an edge) but weakly when both are uniformly lit. This makes the retinal output largely invariant to changes in overall illumination while highlighting the spatially structured differences that the brain needs. It is an efficient solution to the dynamic range problem: the visual system operates across a million-fold range of light intensities, and encoding contrast rather than absolute intensity allows meaningful responses throughout that range.
Center-surround organization also contributes to edge enhancement: a boundary between a light and dark region is emphasized because the ganglion cells at the boundary have their center and surround differentially stimulated. This pre-processing in the retina reduces the information the brain must process by discarding redundant (uniform) regions and highlighting informative (boundary) regions — a principle related to efficient coding theory.