Retina

The Eye’s Jungle


Vision could be defined as telling what is where by looking. But what exactly happens when we “look”? In order to truly understand the complex process that decodes, for example, the words that we see on this page from the photons streaming into our eyes, we must explore the complex tangle of cells known as the retina. Read on to begin your journey to understanding visual perception! Although every part of the eye is important for vision, the part that is absolutely critical for transforming light into neural signals is called the retina. It is located at the very back of the eye, and its output travels through the optic nerve to the brain. The retina consists of several distinct cell layers, which can be labeled as in the diagram below. Click the picture to enlarge:



The retinal layers shown in the image above are known to exist across different mammalian species, and scientists have made progress in identifying major cell types in the neuronal population of a generic mammalian retina: the layers are made up of cell types that have been broadly identified as photoreceptors, amacrine cells, bipolar cells, horizontal cells, and ganglion cells.

Photoreceptors


Photoreceptors consist of two broad classes of cells: rods and cones. Rods are concentrated at the outer edges of the retina and are used in peripheral vision. They are more sensitive to light than cones, and are almost entirely responsible for night vision (also called scotopic vision). Cones are more concentrated in the center of the retina, and are the only photoreceptor type found in the center of the retina (the fovea). Cones are responsible for color vision (also called photopic vision). Mammals usually have either two or three different types of cone cells, because in order to specify the wavelength of a stimulus (i.e., its color), the outputs of at least two cone types must be compared.

Horizontal Cells


Horizontal cells are thought to exist in two types, each with a distinct shape, which together provide feedback to all photoreceptor cells. Despite the number of cells with which they form synapses, horizontal cells represent a relatively small population of the retina’s cells (less than 5% of cells of the inner nuclear layer). The specific reason for the existence of the two classes of horizontal cells is not yet known; it potentially involves detection of color differences in the red-green system.

Amacrine Cells


Amacrine cells appear to allow for ganglion cells to send temporally correlated signals to the brain: input to two separate ganglion cells from the same amacrine cell will tend to make those ganglion cells send signals at the same time. The amacrine cells whose behaviors are well understood have been shown to have very specific functions.

More Details

Bipolar Cells


Bipolar cells connect photoreceptors and ganglion cells. Their function is to transmit signals from photoreceptors to ganglion cells, either directly or indirectly. Bipolar cells get their name from their shape — they have a central cell body from which two different sets of neurites (axons or dendrites) extend. They can make connections with either rods or cones (but not both simultaneously), and they also form connections with horizontal cells. Unlike most neurons, which communicate with one another using action potentials, bipolar cells “talk” with other cells using graded potentials.

More Details

Ganglion Cells


Ganglion cells are the output cells of the retina. Their axons leave the eye and travel through the optic nerve to the brain, sending the processed visual stimulus to the lateral geniculate nucleus, forming synapses onto neurons that project to the primary visual cortex, where the stimulus can be further interpreted.

More Details

Current Research
According to many neuroscience textbooks, retinal ganglion cells can be categorized into two different types according to a property that is known as their receptive field. Neurons with receptive fields have been found in the auditory (hearing) system, the somatosensory (feeling) system and the visual system, and the receptive field of a particular neuron can generally be defined as a region of space in which the presence of a stimulus will alter the firing of that neuron.

In the visual system, a receptive field of a particular retinal ganglion cell is defined as the region of the photoreceptor cell layer in the retina that alters the firing (signal-sending) of that ganglion cell when it is stimulated with light. According to textbook accounts, retinal ganglion cells either have ON-center, OFF-surround or OFF-center, ON-surround receptive field. An ON-center, OFF-surround ganglion cell will send a signal when the center of its receptive field detects light, but will be inhibited from firing when the area surrounding the center (the surround) of its receptive field detects light; OFF-center, ON-surround cells have the exact opposite response to light stimulation.

Click on the image below to see how neurons with different receptive fields respond to stimulation with light:



Within the last decade, it has become increasingly clear that the notion that only two types of receptive fields exist in photoreceptors is a gross oversimplification. Scientists now know that ganglion cells come in at least 15 or 20 types, each of which has a distinct shape and physiological function, and which correspondingly has connections with different types of cells in the rest of the retina.

At the Max Planck Institute (MPI) for Medical Research in Heidelberg, Germany, a dataset was obtained from a mouse retina in order to investigate this diversity in retinal ganglion cells – by applying two imaging techniques one after the other (two-photon microscopy (2P) and serial block face scanning electron microscopy (SBEM)), scientists have been able to obtain images that show both neural activity and connectivity in retinal ganglion cells. However, the images are very difficult to analyze and interpret, and doing so is a very time-consuming process. Computer scientists at MIT are working on developing software to help with retinal image analysis, but computational analysis is currently much less accurate and reliable than that performed by humans.

Navigating the Jungle
The ultimate goal motivating the research on the retina that is being done at places like MPI and MIT is to use 2P and SBEM images in order to identify specific cell types within the broad classes of retinal cells that were described earlier, and further to understand connectivity between these cells. Only once the different cell types have been comprehensively catalogued will researchers be able to investigate their specific functions.

This is where YOU come in! In order to fully understand retinal computation, it is necessary to map all of the connections that converge onto ganglion cells, as this diversity of connections generates the diversity of visual signals that are sent to the brain. The challenge now is to refine the coarse knowledge about retinal connectivity in order to gain a much more in-depth understanding of the specific functions of each and every cell type in the retina. Currently, scientists think that there are at least between fifty and sixty types, so there is much work to be done!

Want to learn more about the retinal cells that you’re helping to construct here on Eyewire? Click on the following links to access publications on the topics below:

Basics on the retina


 * Remind yourselves with any standard textbook most familiar to you (e.g., Chapter 9 of Bear, Connors, and Paradiso, Chapter 26 of Kandel, Schwartz, and Jessell, etc.)
 * R. H. Masland. The fundamental plan of the retina. Nature Neuroscience 4: 877-886 (2001). pdf
 * R. H. Masland. Neuronal diversity in the retina. Current Opinion in Neurobiology 11: 431-436 (2001).

Classification of retinal neurons: morphological and genetic methods


 * S. Siegert, B. G. Scherf, K. Del Punta, N. Didkovsky, N. Heintz, B. Roska. Genetic address book for retinal cell types. Nat Neurosci. 12: 1197-1204 (2009).

Ganglion cells


 * J.-H. Kong, D. R. Fish, R. L. Rockhill, and R. H. Masland. Diversity of ganglion cells in the mouse retina: unsupervised morphological classification and its limits. J. Comp. Neurol 489: 293-310 (2005).
 * I.-J. Kim, Y. Zhang, M. Meister, and J. R. Sanes. Laminar Restriction of Retinal Ganglion Cell Dendrites and Axons: Subtype-Specific Developmental Patterns Revealed with Transgenic Markers. J. Neurosci. 30(4): 1452-1462 (2010).

Amacrine cells


 * M. A. MacNeil, J. K. Heussy, R. F. Dacheux, E. Raviola, R. H. Masland. The shapes and numbers of amacrine cells: matching of photofilled with Golgi-stained cells in the rabbit retina and comparison with other mammalian species. J. Comp. Neurol. 413(2): 305-326 (1999).

Bipolar cells


 * K. K. Ghosh, S. Bujan, S. Haverkamp, A. Feigenspan, and H. Wässle. Types of bipolar cells in the mouse retina. J. Comp. Neurol 469(1): 70-82 (2004).
 * H. Wässle, C. Puller, F. Müller, and S. Haverkamp. Cone Contacts, Mosaics, and Territories of Bipolar Cells in the Mouse Retina. J. Neurosci 29(1): 106-117 (2009).

Neural circuits and computation models


 * I.-J. Kim, Y. Zhang, M. Yamagata, M. Meister, and J. R. Sanes. Molecular identification of a retinal cell type that responds to upward motion. Nature 452: 478-82 (2008).
 * K. L. Briggman, M. Helmstaedter, and W. Denk. Wiring specificity in the direction-selectivity circuit of the retina. Nature 471: 183-188 (2011).
 * S. A. Baccus, B. P. Ölveczky, M. Manu, and M. Meister. A Retinal Circuit That Computes Object Motion. J. Neurosci. 28(27): 6807-6817 (2008).
 * S. Venkataramani and W. R. Taylor. Orientation Selectivity in Rabbit Retinal Ganglion Cells Is Mediated by Presynaptic Inhibition. J. Neurosci. 30(46): 15664-15676 (2010).

Retina