Presentation on theme: "A study by Dr. E.J. Chichilnisky, Systems Neurobiology Laboratories at the Salk Institute, La Jolla, CA Presented by: Sarah-Nicole Bostan Psychology 159."— Presentation transcript:
A study by Dr. E.J. Chichilnisky, Systems Neurobiology Laboratories at the Salk Institute, La Jolla, CA Presented by: Sarah-Nicole Bostan Psychology 159 Receptive Fields in Primate Retina: coordinated to sample visual space more uniformly
Abstract “In the visual system, large ensembles of neurons collectively sample visual space with receptive fields (RFs). A puzzling problem is how neural ensembles provide a uniform, high-resolution visual representation in spite of irregularities in the RFs of individual cells. This problem was approached by simultaneously mapping the RFs of hundreds of primate retinal ganglion cells. As observed in previous studies, RFs exhibited irregular shapes that deviated from standard Gaussian models. Surprisingly, these irregularities were coordinated at a fine spatial scale: RFs interlocked with their neighbors, filling in gaps and avoiding large variations in overlap. RF shapes were coordinated with high spatial precision: the observed uniformity was degraded by angular perturbations as small as 15°, and the observed populations sampled visual space with more than 50% of the theoretical ideal uniformity. These results show that the primate retina encodes light with an exquisitely coordinated array of RF shapes, illustrating a higher degree of functional precision in the neural circuitry than previously appreciated.”
Layman’s terms… Each neuron has its own receptive field, the patch of receptors sending messages to X neuron. Another way to describe this is that all info reaching the brain is sent by retinal ganglion cells, and each of these cells is sensitive to a small region of space (its receptive field). These receptive fields can be irregularly shaped, depending on placement in retina & direction of foveal attention. Hyperfields projecting to hypercolumns exponentially smaller & more densely packed in periphery, hence irregular shape of RF. How, then, does the brain provide us with an uninterrupted, high-resolution visual representation, despite these irregularities?
2 routes by which high-resolution visual information is stored: Magnocellular pathway = parasol retinal ganglion cells Called such because they carry luminance signals to magnocellular layers of lateral geniculate nucleus LGN: primary “relay station” for visual info. from retina, receives input from RGCs LARGE in size Rods, necessary for small differences in brightness (i.e.- stars in peripheral vision), depth perception, movement Parvocellular pathway = midget retinal ganglion cells Called such because they carry luminance signals to parvocellular layers of LGN, also RED/GREEN color opponency (wave lengths) SMALL in size Cones, responsible for color vision & fine detail
Function matches form… right? Formerly thought that parasol & midget cells were circularly symmetric. No longer the case! Passaglia et. al. showed through tests on macaque monkeys & cats (through mathematical formulation) that receptive fields of both types of RGCs are, in fact, elliptical! “by measuring responses to drifting sinusoidal gratings of different spatial frequency and orientation” Midget cells > parasol cells in ellipticity/unevenness/jaggedness. Midget cells have bimodal centers!
Old idea of how receptive fields are structured: ~ 20 diff. types of RGCs Work together to send a complete “picture” to the brain (note- inner screen similarity), with each type forming a “regular lattice covering visual space” To the left: how RGCs were formerly believed to work, their perfectly circular shape lent to a very pristine view of the “inner screen” idea.
New discoveries in structure of receptive fields! Current research upholds that the receptive fields of RGCs aren’t as simple as once theorized… Paradoxically, despite high visual acuity of primates, parasol & midget retinal ganglion cells have inconsistent shapes within the latticework.
Effects of heterogeneous collection of RGCs: (can sometimes be attributed to severe glaucoma)
Method: “To measure directly how ganglion cell populations sample visual space, large- scale simultaneous recordings were obtained from hundreds of identified neurons in patches of peripheral primate retina. Stable recordings over several hours allowed RFs to be mapped at a fine spatial scale. Because hundreds of cells were recorded simultaneously, they could be grouped into clear functional classes defined by physiological properties such as latency, light response polarity, and spike train autocorrelation.” Latency: how long it takes for info to be transmitted from one place to another Light response polarity: Spike trains = sequence of action potentials! While action potentials are believed to be elemental bits of information transmittable by a neuron, temporal structure of a spike train serves as a code for the information transmitted.
Results: Overall visual coverage is NOT patchy or excessively overlapped. Receptive fields may be jagged, but they coordinate with surrounding neighbors & interlock, in fact, enhancing visual acuity. Evidence: “irregular outlines of neighboring cells complemented each other, interlocking like jigsaw puzzle pieces.” This coordination exists among all 4 cell types!
More Results: REJECTED Null Hypothesis: deviation from observed arrangement of RGCs won’t produce more uniform coverage. If RGCs were indeed perfect ellipses, rotating 180 degrees should have no effect, but had significant effect. More overlap & more gaps were created, leading to less effective high- resolution sampling. REJECTED Null Hypothesis: deviation from observed arrangement of RGCs won’t produce more uniform coverage. If RGCs were indeed perfect ellipses, rotating 180 degrees should have no effect, but had significant effect. More overlap & more gaps were created, leading to less effective high- resolution sampling. Uniformity index: a quantitative measure of the regularity of visual coverage. 1= perfect visual coverage, higher #’s = better visual coverage, lower #’s = more poor visual coverage Rotating RF about its axis by even 15 degrees produced significant reduction of visual scene Uniformity index: a quantitative measure of the regularity of visual coverage. 1= perfect visual coverage, higher #’s = better visual coverage, lower #’s = more poor visual coverage Rotating RF about its axis by even 15 degrees produced significant reduction of visual scene
53% improvement in visual area sampled in observed receptive field vs. scrambled RFs, suggesting that receptive fields are randomly individually formed, but exist at specific locations to provide optimal vision
Significance: “How might RF coordination arise during development? Given the diversity of circuit elements that must be arranged to precisely align interlocking RF shapes, one possibility is that RF shapes arise from plasticity driven by visual input. Under this hypothesis, the mechanisms that modify retinal circuitry would be sensitive to the coordination of visual signals in neighboring RFs, as distinct from anatomical growth cues or patterns of spontaneous activity… The present results have surprising implications for how populations of neurons produce an efficient and complete representation. Recorded in isolation, single neurons frequently exhibit irregular response properties, suggesting that large populations must rely on averaging or interpolation to produce accurate sensory performance or behavior. The present results, however, show that in a complete population, irregular features can be integral to a finely coordinated population code. This suggests that the nervous system operates with a higher degree of precision than previously thought, and that irregularities in individual cells may actually reflect an unappreciated aspect of neural population codes.”
Thank you for your time & attention! Herman grid illusion, 1870.
Works cited: Callaway, Edward. Systems Neurobiology Laboratories, Salk Institute, La Jolla, CA. “Structure and function of parallel pathways in the primate early visual system” Feb. 6, 2009. http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1464718/figure/fig01/http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1464718/figure/fig01/ Passaglia, Christopher. Department of Biomedical Engineering & Neuroscience Institute, Northwestern University, Evanston, IL. “Orientation sensitivity of ganglion cells in primate retina.” Jan. 1, 2002. http://www.ncbi.nlm.nih.gov/pubmed/11888534http://www.ncbi.nlm.nih.gov/pubmed/11888534 Frisby, John & Stone, James. “Seeing: A Computational Approach to Biological Vision.” Massachusetts Institute of Technology. Cambridge, MA. 2010.