Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, NEUROSCIENCE ( (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

date: 24 January 2020

Mammalian Visual System Organization

Summary and Keywords

Many mammals, including humans, rely primarily on vision to sense the environment. While a large proportion of the brain is devoted to vision in highly visual animals, there are not enough neurons in the visual system to support a neuron-per-object look-up table. Instead, visual animals evolved ways to rapidly and dynamically encode an enormous diversity of visual information using minimal numbers of neurons (merely hundreds of millions of neurons and billions of connections!). In the mammalian visual system, a visual image is essentially broken down into simple elements that are reconstructed through a series of processing stages, most of which occur beneath consciousness. Importantly, visual information processing is not simply a serial progression along the hierarchy of visual brain structures (e.g., retina to visual thalamus to primary visual cortex to secondary visual cortex, etc.). Instead, connections within and between visual brain structures exist in all possible directions: feedforward, feedback, and lateral. Additionally, many mammalian visual systems are organized into parallel channels, presumably to enable efficient processing of information about different and important features in the visual environment (e.g., color, motion). The overall operations of the mammalian visual system are to: (1) combine unique groups of feature detectors in order to generate object representations and (2) integrate visual sensory information with cognitive and contextual information from the rest of the brain. Together, these operations enable individuals to perceive, plan, and act within their environment.

Keywords: visual system, structure-function relationships, parallel streams, local circuitry, feedforward/feedback connections, sensory/contextual information integration

List of Abbreviations

  • LGN – lateral geniculate nucleus of the thalamus

  • V1 – primary visual cortex

  • V2 – secondary visual cortex

  • Magno – magnocellular

  • Parvo – parvocellular

  • Konio – koniocellular


In biological systems, structure often dictates function. The mammalian visual system is an elegant example of this tenet. In the subset of mammals that rely on vision as their primary means of sensing the environment, including carnivores and primates, significant proportions of the brain are dedicated to processing visual information and the structures involved are precisely organized to enable rapid and dynamic transmission of visual signals. Visual brain areas are locally organized—specific types of neurons and circuits are segregated into discrete structural compartments within each brain area—and the connections between these brain areas are also organized into separate channels that enable parallel processing of information about different features in the visual environment. Thus, highly evolved visual systems are structurally organized to facilitate sophisticated visual functionality such as acuity and color perception (Kaas, 2005, 2013a). Accordingly, the mammalian visual system provides a unique substrate in which to examine how structural organization gives rise to specific function. In order to make these critical links between structure/organization and function, it is essential to understand both the “connectivity map” of the visual system and also the visual function of identified neurons. Fortunately, a number of exciting technological developments have enabled researchers to reconstruct complex connectivity maps or wiring diagrams among visual neurons (Briggman, Helmstaedter, & Denk, 2011; Briggman & Bock, 2012; Callaway & Luo, 2015) and also to record the physiological responses of increasingly large populations of visual neurons in intact brains (Mrsci-Flogel et al., 2007; Murakami, Yoshida, Matsui, & Ohki, 2015; Ohki & Reid, 2014).

Using these innovative methods, visual neuroscientists have recently observed striking differences in the organization of visual systems across mammals with varying visual capabilities. The visual systems of mammals that rely less on vision to sense the environment, including rodents, lack precise structural organization (Ohki & Reid, 2007) and instead are arranged to integrate information across multiple senses (Niell, 2015). It is important to study visual function in a variety of species with varying visual capabilities because vestiges of multi-sensory integration may be present in visual brain areas of highly visual species as well. In primates, including humans, visual areas that are immediately downstream of the retina, that is, “early visual structures,” are modulated by contextual and cognitive signals such as attention (McAlonan, Cavanaugh, & Wurtz, 2008; O’Connor, Fukui, Pinsk, & Kastner, 2002; Saalmann & Kastner, 2009), expectation (Lima, Singer, & Neuenschwander, 2011), and reward contingencies (Stanisor, van der Togt, Pennartz, & Roelfsema, 2013). Thus visual structures across mammalian species have preserved the capacity to integrate information from multiple sources, whether that entails integration of multi-sensory information or integration of visual information with cognitive signals.

The feedforward hierarchy of brain structures that make up the mammalian visual system is the basis for the organization of this chapter. Emphasis is placed on primate and carnivore visual systems with pertinent information about other species included throughout. Additionally, the “early” visual brain structures (retina, visual thalamus, primary visual cortex) receive the most attention because more is known about the structural components and organization of these areas. Three organizational themes are discussed, each of which provides important clues about how visual system structure facilitates visual perception. These themes are: parallel processing pathways; feedforward, feedback, and local circuits; and structural organization to promote rapid and dynamic contextual modulation. Commentary on various techniques and experimental methods that enable mapping the structure, organization, and function of visual areas is also included.

Structures of the Mammalian Visual System


All visual information first enters the mammalian brain at the retina, which is located at the back of the eye. The retina is organized into layers, each of which contain structurally and functionally unique neuronal populations (Ramon y Cajal, 1892). Located at the back of the retina adjacent to the pigment epithelium are the photoreceptors that transduce photons into neuronal signals (Baylor, Nunn, & Schnapf, 1984). There are two main types of photoreceptors: rods, which are optimized for scotopic or low-light-level vision (e.g., night vision), and cones, which are optimized for photopic or high-light-level vision (e.g., daytime vision). Rods and cones transmit visual signals, in the form of graded changes in membrane potential voltage, to bipolar neurons that make up the inner nuclear layer of the retina. Rods are absent from the fovea, the region of the primate retina that contains tightly packed cones and enables high-acuity vision. The retinal circuits downstream of rods differ from those downstream of cones, for example amacrine cells relay rod signals to retinal ganglion neurons that transmit scotopic light information (Nelson, 1982), therefore rods may contribute to a unique visual information channel that is specialized for low-light vision (Euler, Haverkamp, Schubert, & Baden, 2014). Because the vast majority of visual neuroscience focuses on cone-mediated vision, the focus here will be on photopic or cone-mediated vision.

Cones are subdivided into different types determined by the specific wavelengths of light to which they are most sensitive. As far as we know, carnivores such as cats have two cone types, sensitive to short (~420 nm) and long (~560 nm) wavelengths of light, and are thus dichromats. Old world primates, including humans, have three types of cones, sensitive to short, medium (~530 nm), and long wavelengths, and are trichromats. Cone sensitivities are determined by distinct cone opsin genes that translate the light-sensitive proteins expressed in each cone (Lamb, 2013). There are a number of hypotheses regarding the evolution of trichromacy through the duplication of a cone opsin gene in an Old World primate ancestor. Some suggest the evolutionary pressure that generated separate middle and long wavelength cones and their downstream neuronal circuits arose from a need to distinguish red from green in order to select ripe from unripe fruit (Hunt et al., 1998). There are at least three different classes of bipolar cells that receive cone inputs in the primate retina (Field & Chichilnisky, 2007). These bipolar cell classes represent the beginning of the three feedforward parallel information processing streams—the magnocellular (Magno), parvocellular (Parvo), and koniocellular (Konio) streams in primates (Callaway, 2005; Kaplan, 2004). Somewhat homologous parallel processing streams are also present in carnivores, termed the X, Y, and W streams (Sherman & Spear, 1982). Retinal and thalamic neurons with physiological characteristics similar to the Magno/Parvo/Konio or X/Y/W information streams have not been identified in rodents (Scholl, Tan, Corey, & Priebe, 2013; but see Krahe, El-Danaf, Dilger, Henderson, & Guido, 2011), although different types of parallel processing pathways may be present in rodents that are not as specific for vision, but instead optimized for multi-sensory integration (Bickford, Zhou, Krahe, Govindaiah, & Guido, 2015; Niell, 2015). In primates and carnivores, however, the parallel streams begin with distinct bipolar cell classes in the retina and remain strikingly segregated, both anatomically and functionally, through the visual thalamus and into the input layers of the primary visual cortex (Kaplan, 2004; Sherman & Guillery, 2009).

Importantly, neurons in each of the three parallel streams come in two varieties, ON and OFF, with responses to increments and decrements of light, respectively. These ON and OFF cell types first appear at the bipolar cell stage. All cone photoreceptors emit the same output signal: in response to light increments, cones hyperpolarize and decrease glutamate release; in response to light decrements, cones depolarize and increase glutamate release (Devries, 2014). ON and OFF bipolar cells are distinguished by the type of postsynaptic receptor they express: ON bipolar cells express metabotropic glutamate receptors that are sign-reversing, therefore ON bipolar cells are depolarized by light increments and hyperpolarized by light decrements; OFF bipolar cells express ionotropic glutamate receptors that are sign-conserving, therefore OFF bipolar cells are hyperpolarized by light increments and depolarized by light decrements, just like the cones (Attwell, Werblin, Wilson, & Wu, 1983). Bipolar cells receiving input from foveal cones receive input from single cones whereas bipolar cells at parafoveal and peripheral eccentricities receive inputs from multiple cones (Masland, 2001). Regardless of the number of cone inputs, bipolar cell receptive fields are roughly circular with a “center” that is determined by the direct cone input (e.g., ON-center or OFF-center) and a “surround” that is generated through lateral inhibition mediated by horizontal cells that link cone-bipolar circuits of opposite luminance polarity (Dacey, Lee, Stafford, Pokorny, & Smith, 1996; Thoreson & Mangel, 2012). This center/surround structure defines the bipolar cell’s “classical” receptive field, or the region of visual space in which a stimulus drives a response in the neuron. The center/surround structure of retinal neuronal receptive fields (Kuffler, 1953) is the main building block of neuronal receptive fields throughout the visual system. Retinal neurons are also topographically organized such that the receptive fields of neighboring retinal neurons represent neighboring regions of visual space (Masland, 2001). This topographic organization is called retinotopy and it originates in the retina and is maintained throughout most of the brain areas in the visual system.

At least two of the three cone bipolar cell classes, the diffuse and midget bipolar cells, have distinct morphologies that signify whether they convey ON or OFF signals. ON diffuse and midget bipolar cells terminate in the ON subregion of the inner plexiform layer while OFF diffuse and midget bipolar cells terminate in the OFF subregion of the inner plexiform layer of the retina (Figure 1; Field & Chichilnisky, 2007). Diffuse bipolar cells receive mixed inputs from long and middle wavelength sensitive cones and synapse onto parasol retinal ganglion cells in the ganglion cell layer of the retina (Field & Chichilnisky, 2007). The diffuse bipolar-to-parasol retinal ganglion cell connection forms the beginning of the magnocellular (Magno) stream in primates (Nassi & Callaway, 2009). There is an additional, rarer retinal ganglion cell type with smooth and more widely distributed dendrites compared to parasol retinal ganglion cells that projects axons to the superior colliculus and also to the Magno layers of the LGN (Crook et al., 2008). Smooth retinal ganglion cells have a mixture of middle and long wavelength cone inputs, similar to parasol retinal ganglion cells, and their physiology is similar to that of Y retinal ganglion cells in the cat (Crook et al., 2008). Overall, homologies have been made between parasol retinal ganglion cells in the primate and Y retinal ganglion cells in the carnivore because both retinal ganglion cell types have fast-conducting axons, slightly larger receptive fields compared to eccentricity-matched midget or X retinal ganglion cells, transient responses to visual stimuli, nonlinear spatial summation within the receptive field, high contrast sensitivity, and high temporal frequency sensitivity (Nassi & Callaway, 2009). Neither Magno nor Y stream retinal ganglion cells are color opponent; instead, both prefer luminance modulations (e.g., ON center—OFF surround) within their receptive fields. Thus in primates and carnivores, the Magno and Y streams are thought to convey signals that form the basis for motion detection.

Mammalian Visual System Organization

Figure 1. Circuitry of the primate retina. The layers of the retina are outlined by boxes and labeled to the right. Light moves in the direction of ganglion cell layer to photoreceptor layer where it activates rods and cones and is eventually absorbed by the pigment epithelium. Rods (R) and rod bipolar (RB) neurons are illustrated in orange. Long-wavelength-sensitive cone photoreceptors (L) and their downstream circuits are illustrated in red; medium-wavelength-sensitive cones (M) and their downstream circuits are illustrated in green; and short-wavelength-sensitive cones (S) and their downstream circuits are illustrated in blue. Midget bipolar (MB), diffuse bipolar (DB), blue-cone bipolar (BB), midget, parasol, and small-bistratified neurons are labeled and ON and OFF cells are indicated by their terminations in the inner plexiform layer subdivisions. Horizontal cells (H1 and H2) are illustrated in brown and purple and an amacrine cell is illustrated in yellow. The receptive fields of each retinal ganglion cell type are illustrated beneath each retinal ganglion cell. This segment represents the foveal region of the retina in that circuits in the parvocellular stream (midget bipolar and midget retinal ganglion cells) have one-to-one correspondence or private line communication from cone to retinal ganglion cell. Diffuse bipolar to parasol circuits make up the magnocellular stream and circuits downstream of S cones make up the koniocellular stream.

Figure created by F. Briggs.

Midget bipolar cells receive input predominantly from cones with the same wavelength sensitivity and synapse onto midget retinal ganglion cells (Field et al., 2010; Masland, 2001). In the fovea, there are “private lines” or labeled lines of communication from a single cone to a single bipolar cell to a single retinal ganglion cell (Dacey, Crook, & Packer, 2014). Because the center of the receptive field of these private line retinal ganglion cells is driven by light activation of a single cone, these private line circuits set the upper bound on animals’ ability to detect high spatial frequency stimuli and establish acuity limits (although acuity can also be enhanced through local circuitry in the cortex leading to “hyper-acuity”; Masland, 2001; Westheimer, 2008). Non-primate species lack a defined foveal region in the retina; however, carnivores have an area centralis in which cones make private line connections to retinal ganglion cells and establish relatively high acuity vision through similar circuit mechanisms (Rapaport & Stone, 1984). Midget bipolar and retinal ganglion cells that receive input from long or middle wavelength sensitive cones form the beginning of the parvocellular (Parvo) stream in primates (Nassi & Callaway, 2009). The Parvo stream in primates is in some ways homologous to the X stream in carnivores although most carnivores probably lack red/green color vision and their acuity is not as high as that for primates. However, both midget and X retinal ganglion cells have slower conducting axons and smaller receptive fields relative to parasol/Y retinal ganglion cells, sustained responses to visual stimuli, linear spatial summation within the receptive field, and poorer sensitivity to contrast and temporal frequency (Kaplan, 2004; Nassi & Callaway, 2009; Sherman & Spear, 1982). While X and Parvo retinal ganglion cells have similar center/surround receptive field structure, only midget retinal ganglion cells are color opponent (e.g., red [long wavelength] ON center—green [middle wavelength] OFF surround; Nassi & Callaway, 2009). Thus for these reasons, the Parvo and X streams are thought to convey signals about visual form, red/green color (in the Parvo stream), and acuity.

The short wavelength sensitive cones are the evolutionarily oldest cone photoreceptors in mammals (Hunt & Peichl, 2014). Short wavelength cones are less numerous and more widely spaced across the primate retina compared to long and middle wavelength cones (Dacey et al., 2014; Hunt & Peichl, 2014). The retinal circuits downstream of short wavelength cones are less well understood compared to the circuits downstream of long/middle wavelength cones. Blue cone bipolar neurons receive ON inputs from short wavelength cones and synapse onto small bistratified retinal ganglion cells (Dacey et al., 2014; Mariani, 1984). A type of OFF midget bipolar cell that receives short wavelength cone input and conveys short wavelength OFF signals to a specific OFF midget ganglion cell type was subsequently identified (Klug, Herr, Ngo, Sterling, & Schein, 2003). These blue cone bipolar-to-small bistratified and OFF midget circuits likely form the beginning of the koniocellular (Konio) stream in primates (Dacey et al., 2014; Dacey & Packer, 2003; Wassle, 2004). There may be additional retinal ganglion cell types that contribute to the Konio stream, including giant melanopsin-sensitive retinal ganglion cells that are also thought to respond to short wavelength OFF signals (Dacey et al., 2014; Pickard & Sollars, 2012). There may be some homology between the Konio stream in primates and the W stream in carnivores, although both Konio and W stream neurons are less well characterized. Indeed, the W stream in carnivores was originally a catch-all category that contained neurons not classified as participating in X or Y streams (Sherman & Spear, 1982). There is increasing evidence from primates that Konio stream retinal ganglion cells have larger receptive fields and slowly conducting axons relative to Magno and Parvo stream retinal ganglion cells (Dacey et al., 2014). Konio stream retinal ganglion cells are sensitive to contrast and are color opponent (e.g., blue ON center—yellow OFF surround in which the blue ON is derived from short wavelength cone input and the yellow OFF is derived from long plus medium wavelength cone input; Dacey et al., 2014; Nassi & Callaway, 2009). It is likely that neurons making up the Konio and W streams in primates and carnivores are diverse and convey unique visual signals. However, a hallmark of the Konio stream in primates seems to be blue/yellow color information.

In all visual mammals, the feedforward progression of visual signals through the retina begins with the photoreceptors and continues through graded potential signaling to bipolar cells that then transmit graded potential signals to retinal ganglion cells. Retinal ganglion cells integrate graded potential inputs and generate action potentials or spikes that are transmitted out of the retina through the optic tract to the lateral geniculate nucleus of the thalamus (LGN), the next brain structure in the hierarchy of the visual system (Masland, 2001). In addition to the photoreceptor/bipolar/retinal ganglion cell types that make up the feedforward circuits of the retina, there are inhibitory neurons within the retina that connect neighboring photoreceptors, bipolar cells, and retinal ganglion cells in the lateral direction. Horizontal cells make connections with the synaptic terminals of photoreceptors, linking neighboring photoreceptors and providing lateral inhibition that generates the antagonistic center/surround receptive field structure of retinal neurons (Masland, 2001). There are two types of horizontal cells—H1 and H2—characterized based on whether they link long or middle wavelength versus short wavelength sensitive cones, respectively (Masland, 2001). The hyperpolarizing activity of horizontal cells generates center-surround receptive field antagonism and color opponency in bipolar and retinal ganglion cells by providing spatially offset inhibition with opposite luminance polarity (e.g., to generate a green ON center—red OFF surround receptive field; Masland, 2001). In addition to horizontal cells, another class of inhibitory cells in the retina, amacrine cells, links neighboring bipolar and retinal ganglion cells at bipolar-to-retinal ganglion cell synapses (Masland, 2001). There are many morphologically and biochemically distinct amacrine cell types, and they likely contribute to different functional properties of retinal ganglion cells including direction selectivity (sensitivity to the direction of stimulus motion), transient responses (responses to the onset/offset of stimuli), extraclassical surround suppression (or suppression from outside the classical receptive field), and adaptation (Masland, 2001). It is important to note that the retina does not receive any top-down or feedback input from the rest of the brain. While other brain structures control pupillary reflexes and muscles responsible for lens contraction and eye movements, all of which influence retinal responses indirectly, the retina itself is basically a closed system in which an impressive number of computations are achieved through local feedforward and lateral circuits.

It is perhaps convenient to summarize the receptive field structures of the main retinal ganglion cell types in the Magno, Parvo, and Konio streams that project to the LGN (Nassi & Callaway, 2009). All are circular with center and surround components that are luminance or color opponent (Figure 1). The parasol retinal ganglion cells in the Magno stream are either ON-center/OFF-surround or OFF-center/ON-surround because parasol retinal ganglion cells receive a mixture of long and medium wavelength cone input to both center and surround regions of the receptive field. Because they receive direct input from a single cone type (long or medium wavelength), midget retinal ganglion cells in the Parvo stream come in four varieties: green ON-center/red OFF-surround; green OFF-center/red-ON surround; red ON-center/green OFF-surround; and red OFF-center/green ON-surround. Small bistratified and OFF midget retinal ganglion cells that receive direct short wavelength cone inputs may have slightly different receptive field structure in that the “center” and “surround” components of the receptive field are more spatially overlapping (Dacey et al., 2014). Therefore, these Konio stream retinal ganglion cells come in two varieties: blue ON/yellow OFF and yellow OFF/blue ON.

The retinal ganglion cell types that project to the LGN make up the majority of retinal ganglion cells in the primate and carnivore retina; however, there are many additional retinal ganglion cell types that project to the superior colliculus, superchiasmatic nuclei, and other structures in order to convey signals important for eye and head positioning and for circadian functions (Hannibal & Fahrenkrug, 2002). Recent developments in genetic, biochemical, and neurophysiological methods have enabled researchers to sample from more, including rare, neuronal cell types within the retina. The use of genetic markers that drive expression of fluorescent proteins in specific retinal cell types enables visualization of the tiling of each cell class across the retina (Dhande et al., 2013). Similarly, advances in multi-electrode array recordings have enabled recordings from tens to hundreds of retinal neurons simultaneously (Field & Chichilnisky, 2007). These types of population neurophysiological recording methods have enabled researchers to demonstrate higher rates of spiking correlations among retinal ganglion cells of the same type, an indication that the stream-specific circuits within the retina carry independent visual signals (Greschner et al., 2011). High-throughput sectioning and analysis of connectivity among retinal neurons using electron microscopy (EM) has generated beautiful maps of the connections between all cell types within an entire block of retinal tissue (Briggman, Helmstaedter, & Denk, 2011). Because EM enables visualization of actual synaptic connections between neurons, it is possible using this technique to accurately reconstruct complete wiring diagrams of retinal circuits (Helmstaedter et al., 2013). These EM techniques have been utilized to visualize the anisotropy of bipolar-to-amacrine cell synaptic connectivity that gives rise to direction selective responses in amacrine cells in the mouse retina (Briggman et al., 2011). Through combinations of genetic, neurophysiological, and microscopy techniques, we may soon have a complete survey of all the cell types in the mammalian retina and a complete connectivity map through which we can infer how specific circuits give rise to the unique functional response properties of each retinal ganglion cell type.

Visual Thalamus—LGN

While there are many different retinal ganglion cell types in the mammalian retina, in primates and carnivores, the majority of retinal ganglion cells project axons to the LGN. In the primate visual system, midget, parasol, and small bistratified retinal ganglion cell axons target distinct Parvo, Magno, and Konio layers within the LGN, respectively (Nassi & Callaway, 2009). The primate LGN contains six main layers: four Parvo layers with alternating ipsilateral and contralateral eye inputs dorsal to two Magno layers also with alternating ipsilateral and contralateral eye inputs (Figure 2; Nassi & Callaway, 2009). The 6 Konio layers are intercalated in between each adjacent Parvo and Magno layer (Hendry & Yoshioka, 1994). Konio LGN neurons typically have smaller cell bodies compared to Magno and Parvo LGN neurons, and in some species, the Konio layers are quite thin, making Konio cells difficult to target for neurophysiological recording. In the carnivore, X and Y retinogeniculate axons project to the LGN A and A1 layers with eye-specific segregation: A layers receive contralateral eye inputs while A1 layers receive ipsilateral eye inputs (Sherman & Spear, 1982). While X and Y retinogeniculate axons target the same layers, they selectively contact LGN X and Y neurons (Sherman & Guillery, 2014) and in some species, there are separate ON and OFF sub-divisions within the A and A1 layers (Sherman & Guillery, 2014). W retinogeniculate axons target the C lamina, which can also receive Y inputs (Sherman & Spear, 1982). Thus, while carnivores lack anatomically separated layers for X and Y retinogeniculate projections, the parallel streams are segregated through the LGN, as in primates. The structural organization of retinogeniculate circuits and LGN layers preserves a number of physiological properties at the LGN processing stage, including monocular responses (each LGN neuron receives input from one eye, although some Konio cells are binocular, receiving inputs from both eyes; Zeater, Cheong, Solomon, Dreher, & Martin, 2015) and retinotopic organization. Importantly, the combination of eye-specific inputs and retinotopic organization within the LGN produces a complete topographic representation of the contralateral visual hemifield, and this map is preserved in all of the visual cortical areas downstream of the LGN in each cortical hemisphere (Hubel, 1988).

Mammalian Visual System Organization

Figure 2. Retina to LGN to V1. Bottom left: A Nissl-stained coronal section through the LGN illustrates the six layers: two magnocellular (Magno) layers highlighted in grey, four parvocellular (Parvo) layers highlighted in pink, and the intercalated koniocellular (Konio) layers highlighted in blue; “c” and “i” indicate which layers receive input from the contralateral or ipsilateral eyes, respectively. Schematic receptive fields represent retinal ganglion cell input to the Konio, Magno, and Parvo layers. Geniculocortical inputs to V1 are indicated for Magno (black), Parvo (red), and Konio (blue) streams. Thicker lines indicate more robust projections. Upper right: A Nissl-stained coronal section through V1 illustrating the cortical layers, labeled to the right and with lines to delineate laminar boundaries. Magno, Parvi, and Konio-recipient zones are indicated by grey, pink, and blue highlights in addition to the arrows at each geniculocortical termination. Ocular dominance (OD) columns are schematically represented by boxes and labeled beneath the figure.

Figure created by F. Briggs.

The majority of neurons in the LGN are relay neurons that receive direct retinal input and project axons to the primary visual cortex (V1; Sherman & Guillery, 2009). The axons of LGN relay neurons make up the geniculocortical projection that is a component of the optic radiations. The LGN also contains inhibitory interneurons that receive direct retinal input and then in turn inhibit LGN relay neurons (Figure 3; Sherman & Guillery, 2009). Both LGN relay neurons and inhibitory interneurons have similar receptive field structure with center-surround antagonism, ON/OFF luminance polarity, and color opponency in primates (Dubin & Cleland, 1977). In general, the receptive field structure of LGN neurons mimics the receptive field structure of their retinal inputs. This is in large part due to the fact that retinogeniculate synapses are anatomically and physiologically positioned to strongly influence LGN neuronal responses. Retinogeniculate synapses are large, have a high probability of vesicle release, and are often located on the proximal dendrites of LGN relay neurons (Sherman & Guillery, 2009). Based on the ultrastructure of retinogeniculate synapses and their physiological strength, retinogeniculate inputs are described as “driving” and are the main determinants of LGN neuronal responses (Sherman & Guillery, 1998, 2009). There may be private line connectivity between the retina and LGN for circuits downstream of the fovea (or the area centralis in carnivores); however, the precise amount of divergence (one neuron connects to many) and convergence (many neurons converge on one neuron) in retinogeniculate connectivity is not known. It is estimated that LGN relay neurons in the cat receive input from around three retinal geniculate neurons at eccentricities near the area centralis representation (Martinez, Molano-Mazon, Wang, Sommer, & Hirsch, 2014), but even with some divergence in the retinogeniculate projection, connection strength could vary such that a single retinogeniculate input dominates (e.g., private line communication). Retinal geniculate cells with receptive fields in the periphery, far from the fovea, are known to sample information from multiple photoreceptors—their dendritic fields and corresponding receptive field widths are scaled up accordingly (Masland, 2001). It is possible that there is more divergence (and convergence) in retinogeniculate connectivity for circuits representing peripheral eccentricities, but it is experimentally challenging to obtain precise measurements of retinogeniculate circuit divergence and convergence at any eccentricity.

Mammalian Visual System Organization

Figure 3. LGN circuits. Left: Schematic efferent, local connections, and afferent connections of the LGN. Retinal inputs are illustrated in green, LGN relay neurons are illustrated in blue, local LGN inhibitory neurons and RTN inhibitory inputs are illustrated in black, and corticogeniculate feedback from V1 is illustrated in red. Note that LGN relay neurons are inhibited locally within the LGN by LGN inhibitory neurons and also via both feedforward geniculocortical and feedback corticogeniculate collaterals through the RTN. Upper right: Schematic illustration of the connections onto an LGN relay neuron in blue. Receptive fields of each cell type are illustrated and outlined in the representative color. The box outlines the inset shown below: an example of a triadic synapse involving a retinal input and a local LGN inhibitory input onto the LGN relay cell. Note that inhibition within the triad may come from an inhibitory neuron with the same polarity receptive field while non-triad inhibition is opposite receptive field polarity relative to the LGN relay neuron.

Figure created by F. Briggs.

In addition to standard axon-to-dendrite synapses, retinogeniculate axons often form “triadic” synapses onto relay neurons (Figure 3; Pasik, Pasik, Hamori, & Szentagothai, 1973; Winfield, 1980; Wilson, 1989). Triadic synapses include a direct retinogeniculate synapse and an adjacent dendritic process of an inhibitory interneuron packaged within the same glomerular structure (Sherman & Guillery, 1996). The inhibitory dendritic process both receives retinal input and directly inhibits the relay cell within the triad. Thus at a triadic synapse, a relay neuron can be both directly excited and indirectly inhibited by the same retinal input. It is likely that the excitation and inhibition delivered to relay neurons at triadic synapses are the same polarity, for example, a retinal ganglion cell provides ON excitation to an ON LGN relay neuron and an ON inhibitory neuron that then inhibits the LGN relay neuron (Figure 3; Hirsch, Wang, Sommer, & Martinez, 2015). This connectivity is in contrast to the majority of inhibitory-to-relay cell connections that involve inhibitory axonal synapses onto relay neurons (outside of triads) and are opposite polarity, for example, an OFF interneuron inhibits an ON LGN relay neuron (Figure 3; Martinez et al., 2014). Given that LGN inhibitory interneurons can initiate opposite-polarity and same-polarity inhibition of LGN relay neurons, it is likely that the role of inhibitory interneurons in visual processing is multifaceted. An elegant computational model based on LGN circuitry and involving opposite-polarity inhibition of LGN relay neurons predicts that the excitatory/inhibitory circuits within the LGN help sharpen LGN responses to visual stimuli and reduce image blur compared to retinal ganglion cell responses to the same stimuli (Martinez et al., 2014). Local inhibitory circuits within the LGN could also play a role in increasing the efficiency of LGN relay cell responses to incoming visual inputs. The spiking rate of retinal ganglion cells is much higher than that of LGN relay neurons (Usrey, Reppas, & Reid, 1998) and evidence suggests that spikes that are less informative about a given visual stimulus are filtered out at the LGN processing stage (Rathbun, Warland, & Usrey, 2010; Sincich, Horton, & Sharpee, 2009). Whether or not local inhibitory circuits within the LGN play a role in this filtering process remains to be determined.

In addition to retinal and local inhibitory inputs, LGN relay neurons also receive inputs from the cortex and from other thalamic and brainstem structures. In fact, the vast majority of synapses onto LGN relay neurons come from non-retinal sources (Erisir, Van Horn, & Sherman, 1997; Van Horn, Erisir, & Sherman, 2000). An important source of synaptic inputs onto relay neurons is the visual portion of the reticular nucleus (RTN), a large, shell-like thalamic structure that wraps around the entire thalamus and extends for many millimeters in the rostral-caudal axis (Pinault, 2004). Neurons in the RTN are entirely GABAergic (i.e., they are inhibitory neurons), and neurons in the visual portion of the RTN (called the “perigeniculate” nucleus in cats) send axons to the LGN where they contact relay neurons. RTN neurons receive collaterals of afferent (feedforward) geniculocortical axons and collaterals of efferent (feedback) corticogeniculate axons (Figure 3; Sherman & Guillery, 2009). Thus, there are both feedforward and feedback disynaptic inhibitory loops that set up RTN-mediated inhibition of LGN relay neurons in conjunction with relay neuron outputs to the cortex and in conjunction with inputs to relay neurons received from the cortex. Interestingly, RTN neurons are connected to one another via gap-junctions (Landisman et al., 2002), and they can receive inputs from a variety of cortical and subcortical structures (Pinault, 2004), making them a possible integration hub for bottom-up sensory and top-down contextual information. The precise role of RTN-mediated inhibition in the LGN is not known; however, the rhythmic spiking properties of RTN neurons have led some to suggest they are involved in entraining LGN neurons into specific spiking activity patterns that are characteristic of brain states such as sleep or wakefulness (Lewis et al., 2015; Steriade, 2001). RTN neurons fire bursts of spikes that can cause prolonged (~200 msec) hyperpolarization of LGN relay neurons (Sherman, 2001; Steriade & Deschenes, 1984). This prolonged hyperpolarization de-inactivates a T-type calcium channel that, when activated by an incoming depolarization, causes a burst of spikes in the relay neuron (Timofeev & Steriade, 1998). Thus, RTN inputs may control the firing mode of LGN relay cells, switching them from a “tonic” mode of spiking to a “burst” mode (Sherman, 2001). The precise functional correlates of tonic versus burst firing modes are not known; however, LGN relay neurons fire more bursts when animals are asleep or under anesthesia (Weyand, Boudreaux, & Guido, 2001). A somewhat related postulate is that RTN neurons convey signals about attentiveness to the LGN (Crick, 1984). Indeed, there is experimental evidence that RTN and LGN neurons are modulated by spatial attention (McAlonan, Cavanaugh, & Wurtz, 2008; Vanduffel, Tootell, & Orban, 2000). Overall, it is tempting to speculate that RTN-mediated inhibition serves a different purpose than local LGN interneuron-mediated inhibition because whereas local LGN inhibition is probably stream- and cell type–specific, RTN inhibition is likely to be more global and governed by influences from other brain areas.

A second important source of synaptic input to the LGN is the visual cortex. In fact, corticogeniculate synapses are the most numerous in the LGN with corticogeniculate synapses outnumbering retinogeniculate synapses by almost 10 to 1 in cats (Erisir et al., 1997). As described previously, corticogeniculate axon collaterals also target the RTN and create disynaptic feedback inhibition in addition to monosynaptic feedback excitation of LGN neurons (Figure 3; Sherman & Guillery, 2009). Corticogeniculate axons synapse directly onto both LGN relay neurons and LGN inhibitory interneurons (Claps & Casagrande, 1990, Weber, Kalil, & Behan, 1989), thus there are two levels of disynaptic inhibition of relay neurons associated with corticogeniculate feedback (Figure 3). The number and type of relay neurons contacted by a single corticogeniculate axon are not known; however, individual corticogeniculate axonal terminations are mostly confined to single LGN layers with some overlap in neighboring LGN layers (Ichida, Mavity-Hudson, & Casagrande, 2014). Corticogeniculate neurons are mainly pyramidal neurons that reside in layer 6 of V1 and V2 (Briggs, Kiley, Callaway, & Usrey, 2016), and their axonal projections to the LGN make up a significant portion of the optic radiations. Corticogeniculate synapses onto relay neurons are small, have a low probability of vesicle release, and are located on the distal dendrites of relay neurons (Sherman & Guillery, 2009). Accordingly, while retinogeniculate synaptic connections “drive” LGN responses, corticogeniculate synaptic connections merely “modulate” the visual responses of LGN neurons (Sherman & Guillery, 1998, 2009). As previously described, the receptive field properties of LGN neurons mimic their retinal inputs and not their corticogeniculate inputs—for example, while most corticogeniculate neurons are orientation tuned (Briggs & Usrey, 2009; Grieve & Sillito, 1995; Harvey, 1978; Tsumoto & Suda, 1980), they do not confer orientation selectivity on LGN neurons. The corticogeniculate feedback projection is anatomically robust, and yet its functional role in vision has remained a mystery. Manipulations of corticogeniculate feedback (with varying degrees of selectivity or reversibility) have led to suggestions that corticogeniculate feedback modulates the gain (Cudiero, Rivadulla, & Grieve, 2000; Gulyas, Lagae, Eysel, & Orban, 1990; Li, Ye, Song, Yang, & Zhou, 2011; Marrocco, McClurkin, & Alkire, 1996; Olsen, Bortone, Adesnik, & Scanziani, 2012; Przybyszewski, Gaska, Foote, & Pollen, 2000), efficiency, or synchrony (Bal, Debay, & Destexhe, 2000; Eyding, Macklis, Neubacher, Funke, & Worgotter, 2003; Sillito, Jones, Gerstein, & West, 1994) of LGN responses to visual stimuli while other studies have demonstrated little impact of corticogeniculate feedback on LGN responses (Denman & Contreras, 2015). One possible functional role for corticogeniculate feedback is to filter incoming sensory information such that signals that are relevant for perceptual tasks are enhanced while signals that are irrelevant are filtered out. Indeed a filtering role for corticogeniculate feedback is consistent with the notion that retinogeniculate spikes that are less informative about a visual stimulus are culled at the LGN processing stage (Rathbun, Warland, & Usrey, 2010). Using optogenetics—a technique that employs light to selectively and reversibly manipulate the activity of specific populations of neurons based on gene expression of non-endogenous light-sensitive ion channels (Airan, Thompson, Fenno, Bernstein, & Deisseroth, 2009)—a recent study demonstrated that corticogeniculate feedback shortens the response latencies and enhances spike-timing precision among LGN neurons (Hasse & Briggs, in review). These findings suggest that corticogeniculate feedback sharpens the temporal precision of LGN responses to visual stimuli and are consistent with the hypothesis that corticogeniculate feedback filters incoming visual signals. Multiple studies have also demonstrated that corticogeniculate feedback sharpens the spatial resolution of LGN receptive fields (Andolina, Jones, & Sillito, 2013; Hasse & Briggs, in review; Worgotter, Eyding, Macklis, & Funke, 2002), perhaps by enhancing the extraclassical surround suppression of LGN neurons (Andolina, Jones, Wang, & Sillito, 2007).

There are two noteworthy observations about corticogeniculate feedback that are consistent across a variety of mammalian species. First, there are multiple corticogeniculate cell types defined both morphologically and physiologically (Briggs et al., 2016; Briggs & Usrey, 2005, 2007, 2009; Grieve & Sillito, 1995; Harvey, 1978; Swadlow & Weyand, 1987; Tsumoto & Suda, 1980). Second, these distinct corticogeniculate cell types are characterized by different axon conduction latencies (Briggs & Usrey, 2005, 2007, 2009; Grieve & Sillito, 1995; Harvey, 1978; Swadlow & Weyand, 1987; Tsumoto & Suda, 1980). Furthermore, evidence in primates suggests that the corticogeniculate cell types are organized into parallel processing streams that map precisely onto the feedforward Magno, Parvo, and Konio streams (Briggs et al., 2016; Briggs & Usrey, 2009). It is therefore tempting to hypothesize that corticogeniculate feedback modulates LGN activity in a stream-specific manner and this stream-specific modulation occurs on different time scales. In primates and carnivores, corticogeniculate neurons with more Magno- or Y-like characteristics have the most rapidly conducting axons (Briggs & Usrey, 2009). Additionally, Magno-like corticogeniculate neurons receive suprathreshold feedforward geniculocortical inputs, setting up a fast and reciprocal geniculo-cortico-geniculate loop for rapid information exchange between the thalamus and cortex in the Magno stream (Briggs & Usrey, 2007). Parvo- or X-like corticogeniculate neurons have medium conducting axons while Konio-like corticogeniculate neurons have very slowly conducting axons (Briggs & Usrey, 2009). Taken together, recent observations of corticogeniculate circuit organization and physiology suggest that corticogeniculate feedback serves to enhance the temporal precision of LGN neurons in a stream-specific fashion.

The parabrachial region of the brainstem provides cholinergic and some noradrenergic input to the LGN (Sherman & Guillery, 2009). Synapses from the brainstem resemble those from the cortex; however, their terminations may be more widespread and less topographically organized than corticogeniculate synapses (Sherman & Guillery, 2009). In the cat, brainstem inputs mainly contact neurons in the A and A1 layers (Sherman & Guillery, 2009). Other inputs to the A layers include serotonergic inputs from the dorsal raphe nucleus of the midbrain and pons and GABAergic inputs from the nucleus of the optic tract in the midbrain (Sherman & Guillery, 2009). These latter two types of inputs are limited in number compared to other non-cortical inputs (Sherman & Guillery, 2009). Additionally, there is evidence that the superior colliculus provides input to the C layers in carnivores and perhaps also the Konio layers in primates (Sherman & Guillery, 2009, 2014). This projection is consistent with the notion of two broadly defined classes of thalamic channels, termed “core” and “matrix,” whereby core thalamic structures convey primarily sensory inputs from peripheral receptors to the sensory cortex while matrix thalamic structures are involved in multi-sensory or contextual integration (Jones, 1998). In the visual system, the Magno/Parvo and X/Y streams and their associated Magno/Parvo and A LGN layers would be considered core thalamic structures while the Konio and W streams and their associated Konio and C LGN layers would be considered matrix thalamic structures (Jones, 1998). Core and matrix classifications could be evolutionarily conserved, for example rodents lack visual parallel streams homologous to Magno/Parvo or X/Y but may have parallel core and matrix channels associated with visual processing (Niell, 2015).

Taken together, it is clear that the LGN is a site of integration of information from multiple sources via feedforward, feedback, and local circuits. Therefore, LGN relay neurons are in a unique position to integrate contextual information with retinal inputs as they transmit visual signals to the cortex. While the visual responses of LGN neurons appear at first glance to be primarily driven by their retinal inputs and merely modulated by their cortical (and other) inputs, emerging evidence suggests that the LGN is not simply a “passive relay” of visual signals from the retina to the cortex. Visual signals entering and exiting the LGN are organized into structurally and functionally segregated parallel processing streams; however, additional, potentially non-stream-specific inputs from the reticular nucleus and from brainstem structures may provide the LGN with more global or contextual information about brain state and attention.

Other Thalamic Structures Important for Vision

The superior colliculus is an evolutionarily conserved brainstem structure that plays important roles in the integration of visual and motor signals (Krauzlis, 2014). In non-mammals, the equivalent structure is called the optic tectum (Krauzlis, 2014). Similar to the LGN, the superior colliculus is a laminar structure that is topographically organized to represent visual space in the contralateral hemifield, and it receives substantial input from retinal ganglion cells (Krauzlis, 2014). In the mouse, more retinal ganglion cells project to the superior colliculus than to the LGN (Zhang, Kim, Sanes, & Meister, 2012). The superficial layer of the superior colliculus is visual while the intermediate and deep layers are multi-sensory and connect to motor areas throughout the brain including the cerebellum and basal ganglia (Krauzlis, 2014). The superficial superior colliculus receives inputs from the visual cortex, visual nuclei in the midbrain (pretectum), and the parabigeminal nucleus (Krauzlis, 2014). Superficial superior collicular neurons in turn make reciprocal projections to these same structures and also make projections to the LGN and ventral LGN (Krauzlis, 2014). Superior colliculus-to-visual cortex projections are routed through the inferior pulvinar and are more selective for dorsal extrastriate visual cortical areas (Kaas & Lyon, 2007). Deep-layer superior collicular neurons receive input from the frontal eye fields, the lateral interparietal cortex, the supplementary eye fields, and the prefrontal cortex; and they project back to these cortical areas via the medial dorsal thalamic nucleus and the lateral pulvinar (Krauzlis, 2014). These circuits are likely important for orienting and planning eye movements and for updating the visual system on planned saccades or shifts in attention (Krauzlis, 2014; Noudoost, Chang, Steinmetz, & Moore, 2010). Neurons in the superior colliculus across a variety of mammals are strongly modulated by salient and often transient motion stimuli, without regard to the specific direction of stimulus motion, and they are modulated by spatial attention directed toward collicular receptive fields (Krauzlis, 2014).

The pulvinar constitutes a significant portion of the thalamus and has recently received more attention for its role in visual perception. The pulvinar is made up of several distinct subdivisions, many of which are linked to separate sensory modalities (Kaas & Lyon, 2007). The visual subdivisions of the pulvinar receive inputs from all visual cortical areas and make reciprocal projections to all visual cortical areas (Saalmann & Kastner, 2009). Many corticopulvinar synapses display “driver” characteristics, leading to the proposal that the pulvinar provides an alternative route by which visual information progresses up the visual cortical hierarchy (e.g., via cortico-pulvinar-cortical circuits in addition to corticocortical circuits; Sherman & Guillery, 2002). Unlike the LGN or the superior colliculus, the pulvinar does not receive direct retinal input (Sherman & Guillery, 2014). Together, these findings suggest the pulvinar is a higher order thalamic structure relative to first order thalamic structures that receive direct peripheral inputs (Sherman & Guillery, 2002). The visual pulvinar receives subcortical input from the superior colliculus, the RTN, and the brainstem (Sherman & Guillery, 2014). Corticopulvinar neurons originating in layer 5 of the visual cortex make “driving” synaptic connections onto pulvinar neurons and also send collateral axonal projections to brainstem motor areas, perhaps carrying an efference copy to these motor areas (Sherman, 2016; Sherman & Guillery, 2014). Corticopulvinar neurons originating in layer 6 of the visual cortex make “modulatory” synaptic connections onto pulvinar neurons and send axon collaterals to the RTN, similar to corticogeniculate neurons also located in layer 6 of the visual cortex (Sherman & Guillery, 2014). Interestingly, the visual portions of the pulvinar, specifically the inferior and ventrolateral pulvinar subdivisions, are segregated into regions that make projections primarily to dorsal versus ventral extrastriate visual cortical areas (Kaas & Lyon, 2007). The posterior and centromedial regions of the inferior pulvinar receive input from the superior colliculus and project outputs to dorsal cortical areas, while the central lateral inferior and ventrolateral pulvinar regions lack input from the colliculus and project to primary, secondary, and ventral extrastriate visual cortical areas (Kaas & Lyon, 2007). The visual pulvinar is somewhat topographically organized, although not as precisely as other visual areas (Standage & Benevento, 1983); and the receptive fields of visual pulvinar neurons are of intermediate size relative to extrastriate cortical neuronal receptive fields (Bender, 1982; Petersen, Robinson, & Keys, 1985; Zhou, Schafer, & Desimone, 2016). Modulating the activity of visual pulvinar neurons alters the magnitude of visual cortical neuronal responses to visual stimuli and shifts in visual spatial attention (Purushothaman, Marion, Li, & Casagrande, 2012; Zhou et al., 2016). Thus the functional role of the visual pulvinar may be to synchronize communication between visual cortical areas in order to facilitate visual perception and attention (Saalmann, Pinsk, Wang, Li, & Kastner, 2012).

Primary Visual Cortex

In highly visual mammals, most visual information is conveyed from the retina through the LGN to primary visual cortex or V1. This major route for incoming visual information follows the parallel processing streams through core thalamic structures. Alternative routes for visual information to the cortex involve matrix thalamic structures including the superior colliculus and pulvinar and possibly also the Konio layers of the LGN. Inputs from the parallel Magno/Parvo/Konio or X/Y/W streams remain segregated in the input-recipient zones of V1 (Kaplan, 2004; Nassi & Callaway, 2009); however, beyond these zones, it is not clear to what extent information from the three feedforward parallel streams is mixed or remains segregated. Instead, two major parallel pathways for visual information emerge from V1 outputs, forming the basis for the extrastriate dorsal and ventral information channels (Bell, Pasternak, & Ungerleider, 2014; Sincich & Horton, 2005). Dorsal extrastriate cortical areas, such as the medial temporal area (MT) and the medial superior temporal area (MST), compute signals important for motion perception (forming the “where” pathway) while ventral extrastriate cortical areas, such as V4 and inferior temporal (IT) cortex, compute signals important for form perception (forming the “what” pathway; Ungerleider & Haxby, 1994). Thus, parallel processing is a hallmark of the mammalian visual system at all stages. This parallel organization scheme effectively rules out the notion of the “grandmother neuron,” or visual perception based on a neuronal look-up table of specific object detectors (not to mention the problem that the brain doesn’t contain enough neurons to encode for every possible visual object in the world!). Instead, mammalian visual perception involves constructing object representations based on combinations of elementary feature detectors (Serre et al., 2007a; Serre, Oliva, & Poggio, 2007b). At the retinal and LGN processing stages, the visual scene is encoded as a series of pixel maps (Lamb, Collin, & Pugh, 2007; Wassle, 2004). It is at the V1 stage that the first feature detectors really emerge.

V1 is also called the “striate” cortex because of the distinct striations made by myelinated fibers. Various staining techniques (Nissl, cytochrome oxidase) reveal the cortical layers and delineate the geniculocortical input-recipient zones in V1 (Figure 2). Like all cortical areas, V1 contains multiple layers numbered 1 through 6 starting at the pial surface and ending at the white matter (for a recent discussion of defining V1 layers in primates, see Balaram, Young, & Kaas, 2014). Unlike other cortical areas, however, the 6 main layers of V1 are further subdivided based on the pattern of geniculocortical terminations. In primates, geniculocortical inputs from LGN Magno/Parvo/Konio relay neurons synapse in specific sublaminar compartments: Magno inputs target layer 4Cα‎ and project collaterals to the bottom third of layer 6; Parvo inputs target layer 4Cβ‎ and project collaterals to the upper third of layer 6; Konio inputs target layer 1, the cytochrome oxidase-rich blobs in layer 2/3, layer 4A, and probably also send collaterals to layer 6 (Figure 2; Nassi & Callaway, 2009). The only traditionally defined layer that does not receive geniculocortical input is layer 5, a characteristic that is consistent across primates and carnivores. Geniculocortical inputs from the ipsilateral and contralateral eye-specific layers of the LGN also terminate in non-overlapping and adjacent columns, termed ocular dominance columns, within layer 4C (Figure 2; LeVay, Hubel, & Wiesel, 1975). The retinotopic map of visual space is preserved in V1 and there is a significant magnification of cortical surface area devoted to the fovea (Dow, Snyder, Vautin, & Bauer, 1981), consistent with the disproportionally large number of retinal ganglion cells and LGN relay neurons with foveal receptive fields (Wassle, 2004). The exact number of geniculocortical inputs to a single V1 neuron is not known, nor is the amount of convergence/divergence per eccentricity (Adams & Horton, 2003; Alonso, Usrey, & Reid, 2001). As with retinogeniculate synapses in the LGN, the number of geniculocortical synapses onto V1 neurons is small compared to the number of cortical synapses (Ahmed et al., 1994; O’Kusky & Colonnier, 1982; Salin & Bullier, 1995); and the strength of geniculocortical synapses is strong such that the visual responses of geniculocortical-recipient neurons in V1 are driven by their geniculocortical inputs and modulated by their local cortical inputs (Stratford, Tarczy-Hornoch, Martin, Bannister, & Jack, 1996).

A key feature detector emerges at the geniculocortical synapse: geniculocortical-recipient neurons in V1 are sensitive to the orientation of a visual stimulus (with the exception of Parvo-recipient neurons in layer 4Cβ‎ that have center/surround receptive fields and are not orientation selective; Blasdel & Fitzpatrick, 1984; Hubel & Wiesel, 1968; Kagan, Gur, & Snodderly, 2002; McLaughlin, Shapley, Shelley, & Wielaard, 2000; Ringach, Hawken, & Shapley, 1997). Hubel and Wiesel proposed a simple model whereby orientation selectivity emerges based on inputs from LGN neurons with receptive fields aligned along a particular orientation axis (Figure 4; Hubel & Wiesel, 1968). Subsequent physiological measurements confirmed this Hubel and Wiesel model in cat V1 (Reid & Alonso, 1996). LGN relay neurons with the same polarity (ON or OFF) and with receptive fields aligned along a particular orientation axis connect to a recipient neuron in V1 and confer orientation selectivity on that neuron (Alonso et al., 2001; Reid & Alonso, 1996). Geniculocortical-recipient V1 neurons are thus selective for both orientation and luminance polarity—Hubel and Wiesel called these neurons “simple cells”. Simple cell receptive fields have separate subregions of opposite polarity within the classical receptive field, both of which are orientation selective (Figure 4; Hubel & Wiesel, 1968). Because of their separate ON and OFF receptive field subregions, simple cells have linear spatial summation within their receptive field and their spatial frequency selectivity is defined by the widths of the ON and OFF receptive field subregions (Ringach et al., 1997; Reid, Victor, & Shapley, 1997). In carnivores and primates, layer 4 (4C) spiny stellate neurons receive geniculocortical input that drives their visual physiology: they are monocular neurons with center surround receptive fields (in layer 4Cβ‎ in primates) or they are simple cells with spatially offset ON and OFF receptive field subregions that are selective for a preferred orientation (Blasdel & Fitzpatrick, 1984; Lund, 1987). Geniculocortical-recipient neurons are also clustered according to their feature selectivity: neighboring simple cells often prefer the same stimulus orientation, they receive inputs from the same eye, and in cat V1, spiny stellate cells with stronger ON or OFF responses are also clustered (Hirsch et al., 2003; Martinez et al., 2005; Ziskind, Emondi, Kurganski, Rebrik, & Miller, 2014).

Mammalian Visual System Organization

Figure 4. Feature maps, circuits, and receptive fields in V1. An “ice cube” model of V1 with layers and blobs illustrated in brown, orientation domains outlined in orange with preferred orientation illustrated by black bars on the cortical surface, ocular dominance domains illustrated by grey ellipses, LGN input illustrated by the black arrow, outputs illustrated by orange arrows, and local circuits illustrated by red arrows. Example receptive fields for LGN cells, a simple cell, and a complex cell are illustrated. The LGN and simple cells have spatially offset ON and OFF subregions (white, black) while the complex cell responds to light increments and decrements within its receptive field and prefers oriented bars moving in a particular direction, indicated by the arrow. Above the ice cube model are two images of optical imaging/2-photon imaging data illustrating the pinwheel structure of neighboring orientation domains

(modified with permission from Bonhoeffer & Grinvald, 1991; Ohki & Reid, 2007).

Hubel and Wiesel characterized a second functional class of neurons within V1 that they called “complex cells”. Unlike simple cells, complex cells respond equally well to light increments and decrements within the same region of the receptive field, that is, complex cells respond best to a bar or edge of a particular orientation placed anywhere within the receptive field (Figure 4; Hubel & Wiesel, 1968). Complex cells likely receive input from multiple simple cells with opposite spatial phase polarity preferences (analogous to the Hubel and Wiesel simple cell model). Without spatially separated ON or OFF receptive field subregions, complex cells have nonlinear spatial summation within their receptive field (Saleem, Ayaz, Jeffery, Harris, & Carandini, 2013). Since the pioneering experiments of Hubel and Wiesel, visual neuroscientists have developed a sophisticated toolkit of visual stimuli to measure the feature selectivity, tuning, and receptive field properties of V1 neurons. The standard stimulus is the drifting sinusoidal grating that can be modulated along a number of different stimulus feature dimensions, including grating orientation, spatial and temporal frequency, size, contrast, and color. Another useful stimulus is the white noise m-sequence, a grid of pixels that alternate between white and black (or along a spectrum of colors) according to a mathematical m-sequence such that no two pixels in the grid are correlated across space or time (Reid et al., 1997). This stimulus enables mapping of the linear components of neuronal temporal and spatial receptive fields. It is important to note that these stimuli are statistically simple and not natural, which is convenient for data analysis purposes but not necessarily representative of what neurons evolved to “see” in the real world. Movies of natural scenes have also been used to measure the responses of neurons to more realistic stimuli (Vinje & Gallant, 2000; Weliky, Fiser, Hunt, & Wagner, 2003); however, data analyses are complicated by the increase in higher-order correlations that are present in natural images (Karklin & Lewicki, 2003; Simoncelli & Olshausen, 2001).

The clustered organization of spiny stellate neurons in layer 4C based on feature selectivity, such as ON/OFF polarity and ocular dominance, is a part of a more global functional architecture within V1. In other words, different visual features are independently mapped and each feature map is superimposed over the same region of the cortex. Hubel and Wiesel first observed systematic and sequential changes in orientation selectivity across V1 (Hubel & Wiesel, 1968). They made vertical electrode penetrations through the V1 layers and observed that neurons within the same vertical electrode penetration preferred the same stimulus orientation. In contrast, when they made tangential electrode penetrations within a layer, they observed sequential shifts in preferred stimulus orientation. Hubel and Wiesel proposed an “ice cube” model for the orientation map in V1 such that vertical columns of neurons preferred the same orientation and neighboring columns preferred sequentially shifted orientations (Figure 4; Hubel & Wiesel, 1968, 1974). Hubel and Wiesel’s idea of a locally organized orientation map was revealed using more modern imaging techniques including optical imaging of intrinsic signals, which revealed the pinwheel structure of orientation columns (Blasdel & Salama, 1986; Bosking, Crowley, & Fitzpatrick, 2002; Schuett, Bonhoeffer, & Hubener, 2002), and 2-photon imaging, which revealed smooth transitions in orientation preference for neurons in each of the columns surrounding a pinwheel center (Figure 4; Ohki, Chung, Ch’ng, Kara, & Reid, 2005; Ohki et al., 2006; Ohki & Reid, 2007). The increased sophistication of in vivo imaging technology has enabled researchers to define numerous feature selectivity maps in V1 and to describe how each of the maps relate to one another. For example, neurons in the superficial layers of V1 are selective for stimulus direction in addition to stimulus orientation, and direction selectivity is mapped within each orientation column in an orderly fashion with a sharp transition in direction preference across the center of each column and smooth transitions in direction tuning across neighboring columns (Bosking et al., 2002; Van Hooser, Heimel, Chung, & Nelson, 2006). Ocular dominance columns, which originate with the eye-specific geniculocortical inputs to layer 4C, run orthogonal to the orientation columns such that each orientation column contains neurons that respond preferentially to inputs from each eye (Kaskan, Lu, Dillenburger, Roe, & Kaas, 2007; Ramsden, Hung, & Roe, 2001). There is a feature map of neuronal selectivity for spatial frequency that is orthogonal to the orientation map such that each orientation column contains neurons with different spatial frequency preferences (Nauhaus, Nielsen, Disney, & Callaway, 2012). Finally, imaging evidence also suggests that the cytochrome oxidase-rich blobs in layer 2/3 (a zone of Konio geniculocortical terminations, Figure 2) contain neurons that are more color selective compared to neighboring neurons (Lu & Roe, 2008). These color-selective blobs are located at the center of ocular dominance columns, consistent with the notion that color-selective neurons within blobs may be monocular (Livingstone & Hubel, 1984, 1988; Lu & Roe, 2008). And color-selective blobs are also located near the borders of orientation columns (Lu & Roe, 2008).

The feature selectivity maps in V1 are a striking example of functional architecture and they illustrate the variety of visual response properties (feature detectors) that emerge at the V1 stage. Overall, V1 neurons detect elementary features such as lines, contrast edges, and spots of color; however, V1 neurons encode for many important aspects of each of these elementary features. Specifically, V1 neurons are specialized to detect the orientation and direction of movement of contrast edges (Conway, Hubel, & Livingstone, 2002; Gur et al., 2003; Hubel & Wiesel, 1968; Xing, Ringach, Shapley, & Hawken, 2004). Neurons in V1 are also tuned for temporal frequency, akin to the speed at which a stimulus moves across the receptive field, so V1 neurons provide extrastriate visual cortex with valuable information about both the direction and speed of moving stimuli (Nauhaus et al., 2012; Ringach, Hawken, & Shapley, 2003). V1 neurons are also specialized to detect the width of contrast edges or bars based on the variation in spatial frequency tuning among V1 neurons (Xing et al., 2004; Nauhaus et al., 2012). Populations of V1 neurons are specialized to detect edges defined by luminance contrast (e.g., black/white), others are specialized to detect edges defined by color, and many neurons respond to luminance contrast and color (Horwitz, Chichilnisky, & Albright, 2007; Johnson, Hawken, & Shapley, 2008). V1 neurons in the input-recipient layers are monocular (Hubel & Wiesel, 1968). Many other V1 neurons, especially those located in the output-projecting layers, integrate information from both eyes via local circuit inputs and display binocular responses (Bridge & Cumming, 2001; Cumming & Parker, 2000; Kagan, Gur, & Snodderly, 2002). Interestingly, some V1 neurons develop tuning for binocular disparity, or the difference in input to the two eyes based on the distance between objects visual space (Cumming & Parker, 2000). Thus binocular disparity tuning that first emerges in V1 forms the basis for depth perception.

There are a number of additional response properties of V1 neurons that are likely to emerge as a result of combinations of feedforward inputs from the retina/LGN, local circuit inputs within V1, and feedback connections from extrastriate cortical areas. A good example of this type of response property is selectivity for stimulus size. Neurons in the retina and LGN are selective for visual stimulus size in the sense that they respond maximally to stimuli that cover their classical receptive field and are suppressed by stimuli that cover the classical receptive field and the extraclassical surround (Figure 5; Alitto & Usrey, 2008; Bonin, Mante, & Carandini, 2005; Cavanaugh, Bair, & Movshon, 2002; Solomon, Peirce, & Lennie, 2004). Extraclassical surround suppression is also present among V1 neurons (Sceniak, Hawken, & Shapley, 2001; Tailby, Solomon, Peirce, & Metha, 2004), but the scale of the suppressive surround can be large (Angelucci & Bressloff, 2006). Thus, extraclassical surround suppression in V1 neurons may be partially inherited from retinal/LGN inputs (Alitto & Usrey, 2008) and partially arise via combination of local circuits and feedback connections from extrastriate cortex (Angelucci & Bressloff, 2006). Interestingly, some early visual neurons (in the retina, LGN, and layer 6 of V1) are increasingly excited by larger and larger visual stimuli (Gilbert & Wiesel, 1979). Hubel and Wiesel termed these V1 neurons “non-end-stopped,” in contrast to the majority of V1 neurons that are selective for stimulus size and are therefore “end-stopped” (Sherk, 1989). Neurons with such large receptive fields may be responsible for signaling global contrast and may provide important inputs to neighboring neurons to implement contrast normalization (Carandini & Heeger, 2012; Sceniak et al., 2001, 2002). Another interesting property of some V1 neurons that may also emerge as a result of feedback from extrastriate cortex is border ownership. Some V1 neurons respond differently to an identical contrast edge depending on the larger context within which that edge is embedded (Zhou, Friedman, & von der Heydt, 2000). Because V1 receptive fields are typically too small to signal information about both the local edge and the larger context, these border ownership responses must emerge from horizontal connections within V1 or top-down feedback from extrastriate neurons with larger receptive fields.

Mammalian Visual System Organization

Figure 5. Extraclassical receptive field. Top: the classical receptive field (red) of an ON-center LGN neuron and the extraclassical surround of the same LGN neuron. Middle: Grating stimuli increasing in size such that they extend beyond the classical receptive field of the LGN neuron and into its extraclassical surround. Bottom: A size-tuning curve for the LGN neuron illustrating a suppression of the neuron’s response to gratings that encroach into the extraclassical surround. This type of response is typical for neurons that display surround suppression.

Figure created by F. Briggs.

Finally, it is important to note that the responses of V1 neurons to visual stimuli can be augmented or reduced by global contextual factors such as brain state (alertness, sleepiness; Sellers, Bennett, Hutt, Williams, & Frohlich, 2015; Stoelzel, Bereshpolova, & Swadlow, 2009), attention (Briggs, Mangun, & Usrey, 2013; Harris & Thiele, 2011; Hillyard, Vogel, & Luck, 1998; Luck, Chelazzi, Hillyard, & Desimone, 1997; McAdams & Reid, 2005; Sanayei, Herrero, Distler, & Thiele, 2015; Thiele, Pooresmaeili, Delicato, Herrero, & Roelfsema, 2009), anticipation of reward (Stanisor, van der Togt, Pennartz, & Roelfsema, 2013), etc. This phenomenon is perhaps more obvious in V1 of mice where the overall spiking rate and feature selectivity of neurons is greatly enhanced when animals are running (Niell & Stryker, 2010), suggesting that V1 in mice is both multi-sensory (e.g., receives direct motor inputs) and strongly modulated by brain state (e.g., alertness). In primates, spiking activity in V1 neurons is only modestly altered by cognitive influences like attention and reward, compared to the influence of attention on the spiking activity of extrastriate neurons in V4 and MT (Hillyard et al., 1998; Luck et al., 1997; McAdams & Maunsell, 1999, 2000). However, it is important to understand how cognitive factors influence the way in which visual information is encoded by early visual structures, as contextual modulation of sensory inputs is likely to be an evolutionarily conserved process.

Both feature selectivity of V1 neurons and organization of V1 neurons into the intricate functional architecture come about via local circuit connections. Some have proposed that there are sets of canonical circuits (Douglas & Martin, 2004) that are common across cortical areas and are modified to fit the functionality of each cortical area. Some of the key canonical cortical circuits are: (1) connections from input-recipient layer 4 with the superficial layers, (2) reciprocal connections between layer 4 and layer 6, (3) reciprocal connections between the superficial layers and layer 5, and (4) local connections within each layer, including excitatory-excitatory and inhibitory-excitatory connections between neighboring neurons (Figure 4; Callaway, 1998, 2004; Douglas & Martin, 2004). It is challenging to estimate the exact number, type, and strength of local circuit connections within a column or across columns, because such an experiment would require a combination of neurophysiological recording from large numbers of connected neurons followed by EM analysis of the complete connectivity map. However, a number of techniques have been utilized to estimate local circuit connectivity in V1, including paired neurophysiological recordings in slices from nearest-neighboring neurons (Thomson & Bannister, 1998, 2003), reconstructions of the distributions of synaptic boutons originating from neurons in different layers (Binzegger, Douglas, & Martin, 2007), and combination of neurophysiological recordings in slices with functional mapping procedures like photostimulation and optogenetics (Hull, Adesnik, & Scanziani, 2009; Sawatari & Callaway, 2000; Scanziani & Hausser, 2009; Yabuta, Sawatari, & Callaway, 2001). In primate V1, many local circuits appear to loosely preserve Magno/Parvo/Konio stream-specificity. For example, layer 4Cα‎ neurons that receive geniculocortical Magno inputs connect directly to a population of spiny stellate neurons in layer 4B that project to MT (Nassi & Callaway, 2006; Yabuta et al., 2001). Layer 4B neurons are complex cells that are tuned to the direction of stimulus motion in addition to the orientation of the stimulus (Gur & Snodderly, 2007; Ringach, Shapley, & Hawken, 2002). Thus the layer 4Cα‎-to-4B spiny stellate circuit sets up a path for motion signals to rapidly reach the dorsal extrastriate cortex. Layer 6 neurons, which are anatomically diverse, receive Magno or Parvo-stream biased inputs from layers 4B and 2/3, respectively, depending on their dendritic/axonal morphology (e.g., whether their local axons and dendrites target layer 4Cα‎ or 4Cβ‎; Briggs & Callaway, 2001). A subset of these layer 6 neurons, the corticogeniculate neurons, also display physiological response properties that are Mango-, Parvo-, or Konio-stream specific, indicating that the local circuitry involving corticogeniculate neurons limits mixing of signals from the parallel streams (Briggs & Usrey, 2009). While the notion is somewhat controversial, there may be a color-specific pathway that begins in V1 and involves mixing Parvo- and Konio-stream signals. Layer 2/3 neurons within the cytochrome-oxidase rich blobs, which some say are monocular simple cells with color-opponent responses (Livingstone & Hubel, 1984, 1988; O’Keefe Levitt, Kiper, Shapley, & Movshon, 1998; Shapley & Hawken, 2011; Solomon & Lennie, 2007; Ts’o & Gilbert, 1988), could integrate blue/yellow signals from Konio geniculocortical inputs with red/green signals from layer 4Cβ‎ to generate a variety of color selectivities. Importantly, neurons in the blobs also project outputs to the color-responsive thin stripes in secondary visual cortex (V2; Livingstone & Hubel, 1988; Sincich & Horton, 2002a). Together these findings could suggest that color-processing begins with the combination of Parvo- and Konio-stream inputs within the blobs.

While most studies focus on local circuits linking neurons across layers within a column, like those previously described, it is important to acknowledge that neurons throughout the cortical layers of V1 project lateral or “horizontal” axons that target neurons in the same layer, often in neighboring columns with matched feature selectivity (Bosking, Zhang, Schofield, & Fitzpatrick, 1997; Chisum, Mooser, & Fitzpatrick, 2003, Stettler, Das, Bennett, & Gilbert, 2002; Weliky, Kandler, Fitzpatrick, & Katz, 1995). Generally speaking, there are two types of lateral/horizontal circuits: (1) short-range circuits that link neighboring excitatory-to-excitatory and inhibitory-to-excitatory neurons in either direction, and (2) long-range circuits that involve long-range axons of excitatory pyramidal neurons contacting both excitatory and inhibitory neurons hundreds of microns away and often spanning multiple columns (Roerig & Chen, 2002; Roerig, Chen, & Kao, 2003; Weliky et al., 1995). Horizontal circuits have been examined most in layer 2/3 because the superficial layers are easier to image than the deeper layers. However, similar horizontal circuits exist across the cortical layers. Layer 2/3 neurons have sharper orientation and direction tuning compared to layer 4C neurons, perhaps because they are laterally connected to one another. Each “long-range” horizontal axon originates in a layer 2/3 pyramidal neuron with a particular orientation preference and terminates in a nearby orientation column, often with similar orientation preference. Thus, axons often contact layer 2/3 pyramidal neurons with similar orientation preferences; they also contact layer 2/3 inhibitory neurons, which then inhibit neighboring pyramidal neurons (Chisum et al., 2003; Roerig et al., 2003; Stettler et al., 2002).

Inhibitory neurons in V1 are morphologically and functionally diverse—different morphological inhibitory cell types often target different cellular compartments of neighboring excitatory pyramidal and spiny stellate neurons with their axons (Nassi & Callaway, 2009). Many, although not all, inhibitory neurons have different action potential waveform shapes and display different firing patterns relative to excitatory neurons, making it possible to identify putative inhibitory neurons based on spiking activity (McCormick, Connors, Lighthall, & Prince, 1985; Cardin, Palmer, & Contreras, 2007). Additionally, advanced genetic techniques are available to target inhibitory cell types based on specific gene expression (Taniguchi, 2014). Modern circuit mapping techniques combining virus-mediated anatomical tracing with genetic manipulations and optogenetics are yielding new information about the function and circuitry involving specific inhibitory cell types in V1 (Marshel, Mori, Nielsen, & Callaway, 2010). Finally, a large proportion of excitatory pyramidal and spiny stellate neurons in V1 make long-range axonal projections that target extrastriate cortical areas and subcortical structures. Output-projecting neurons in V1 are concentrated in the superficial layers (2/3 and 4B) and in the deep layers (layers 5 and 6; Sincich & Horton, 2005). The majority of V1 corticocortical axons terminate in V2 (Figure 6; Felleman & Van Essen, 1991), although the V1-to-MT connection via 4B neurons is also relatively robust (Nassi & Callaway, 2006, 2009). Layers 5 and 6 contain corticocortical neurons that target V2 and perhaps other extrastriate cortical areas (Felleman & Van Essen, 1991; Sincich & Horton, 2002a, 2005) in addition to corticopulvinar (layer 5; Kaas & Lyon, 2007; Rockland & Virga, 1989), corticocollicular (layer 5; Nhan & Callaway, 2012), corticogeniculate (layer 6; Briggs & Usrey, 2011), and corticoclaustral (layer 6; LeVay & Sherk, 1981) neurons (Figure 4).

Mammalian Visual System Organization

Figure 6. Structures and connections in V1 and V2. Left: A cytochrome oxidase-stained tangential section through V1 (left) and V2 (right) illustrating the blobs in layer 2/3 of V1 and the thick, thin, and pale stripes in layer 2/3 of V2. Image modified with permission from Kaas (2013b). Right: Schematic illustration of the connections from V1 to V2 and beyond. Cytochrome oxidase-rich (CO) zones are illustrated in brown.

Reprinted from Gazzaniga, Michael S. (Ed.), The Cognitive Neurosciences III 3d ed. figure 17.7, p. 240, © 2004 Massachusetts Institute of Technology, by permission of The MIT Press.

In summary, neurons in V1 encode a number of elementary stimulus features that may be limited by the organization of the functional architecture, that is, past a certain number of orthogonal feature maps, it becomes challenging to represent more feature dimensions within a single column (Bednar & Wilson, 2016). Some features detectors are likely established via integration of feedforward, feedback, and local circuit connections. Additionally, information about visual context and cognitive state that arrives in V1 via corticocortical feedback and subcortical input may piggyback onto existing cortical circuits in order to augment or reduce neuronal responses or selectivity. In primates and carnivores, functional architecture, feature maps, and morphologically and physiologically distinct neurons within each of the cortical layers define V1. In less visual species, like mice, V1 lacks a functional architecture (Ohki & Reid, 2007) and neuronal morphology and physiology is more homogenous across the layers (e.g., Briggs, 2010), perhaps because V1 in rodents does not create precise feature detectors but instead integrates contextual information with global visual signals.

Extrastriate Visual Cortex

Homologs for some of the primate extrastriate cortical areas have been identified in carnivores, but their structural and functional similarities are not established or well-studied. Visual areas beyond V1 have been identified in the mouse based on retinotopy (Garrett, Nauhaus, Marshel, & Callaway, 2014; Marshel, Garrett, Nauhaus, & Callaway, 2011), but these areas are likely specialized for multi-sensory integration to a large extent and therefore functionally distinct from extrastriate visual areas in primates. There is a rich anatomical literature spanning many decades and involving a number of anatomical tracing tools that describes the corticocortical connectivity within the dorsal and ventral pathways of primate extrastriate visual cortex (Figure 7; Felleman, Burkhalter, & Van Essen, 1997; Felleman & Van Essen, 1991; Gegenfurtner, 2003; Kravitz, Saleem, Baker, & Mishkin, 2011; Lyon & Kaas, 2002; Nassi & Callaway, 2009; Stepniewska & Kaas, 1996; Ungerleider & Desimone, 1986). Aside from these corticocortical connectivity maps, the structural organization within each extrastriate cortical area is largely unknown. Anatomical, imaging, and neurophysiological evidence suggests that V2 and MT contain functional columns (or stripes in V2), but whether other extrastriate cortical areas have functional architecture and feature maps remain open questions. As such, a different experimental approach is required when studying the function of extrastriate cortex because the cell/circuit-centric approach used to study V1 is not obviously applicable. Much of our current knowledge of extrastriate cortical areas is based on averaging neurophysiological responses of many neurons recorded within a cortical area, without knowledge of the morphologies, laminar locations, or circuit identities of recorded neurons. This technique provides estimates of the shared feature selectivity of large populations of neurons and is likely representative of the global functionality of a given cortical area, but also underestimates the number and quality of computations that are achieved by the diverse populations of neurons and circuits within each area. Vision scientists are well aware of the limitations of estimating function based on averaging responses across diverse neuronal populations and many have developed innovative methods to combine neurophysiology with computational modeling in order to estimate how neuronal circuits within and across striate and extrastriate cortex can generate complex functionality like object recognition (Freeman & Simoncelli, 2011; Riesenhuber & Poggio, 1999; Vintch, Movshon, & Simoncelli, 2015; Wallis & Rolls, 1997). Additionally, advances in multi-electrode array design and imaging techniques enable simultaneous recordings from large populations of neurons and it is likely that these new approaches will help elucidate the structural and functional organization of extrastriate cortex.

Mammalian Visual System Organization

Figure 7. Dorsal and ventral streams in the primate. Left: Lateral (top) and medial (bottom) views of the macaque brain and a flat-mounted illustration of the visual areas of the macaque. Warm colors illustrate dorsal stream areas, cool colors illustrate the ventral stream areas. Figure 2 from Felleman and Van Essen (1991), reproduced with permission. Right: Wiring diagram of visual areas in the macaque where warm colors illustrate dorsal stream areas and cool colors illustrate ventral stream areas. Size of boxes approximate cortical territory for each area and line thickness approximates anatomical connection strength. Only feedforward connections are represented.

Reprinted from Neuron, 60(2), Pascal Wallisch and J. Anthony Movshon, “Structure and Function Come Unglued in the Visual Cortex,” pp. 195–197, Copyright (2008), with permission from Elsevier.

V1 corticocortical projections target extrastriate areas V2, V4, and MT (Figures 6 and 7; Felleman & Van Essen, 1991; Nassi & Callaway, 2009; Sincich & Horton, 2002a, 2005), terminating in layer 4. The majority of V1 corticocortical neurons target V2 (Sincich & Horton, 2002b). V2 is easily distinguished from V1 based on the pattern of cytochrome oxidase reactivity—while cytochrome oxidase staining of the superficial layers of V1 reveals the blobs, cytochrome oxidase staining of the superficial layers in V2 reveals alternating thin, pale, and thick stripes (Figure 6; Shipp & Zeki, 2002a, 2002b; Sincich & Horton, 2002a). The thin, pale, and thick stripes contain functionally distinct neurons that receive and send distinct sets of visual information (Figure 6). Neurons in the thin stripes receive input from corticocortical neurons within the blobs in V1 (originating mostly in layers 2/3 and 4B, but also in layers 5 and 6), are color selective, and project to V4 forming a component of the ventral pathway (Sincich & Horton, 2005; Tanagawa, Lu, & Roe, 2010). Neurons in the pale and thick stripes receive input from corticocortical neurons in between the blobs in V1 (again spanning layers 2/3, 4B, 5, and 6) and are more responsive to orientation and direction of stimulus motion and perhaps also binocular disparity (Sincich & Horton, 2005). Pale stripe corticocortical neurons project to V4, contributing to ventral pathway processing, while thick stripe corticocortical neurons project to MT and contribute to dorsal pathway processing (Sincich & Horton, 2005). In fact, V2 provides feedforward input to almost all extrastriate visual cortical areas in both dorsal and ventral pathways, making V2 the main gateway to the dorsal and ventral information processing channels (Figure 7; Felleman & Van Essen, 1991). V2 corticocortical feedback neurons originate in the superficial layers and layer 6 and project axons to the superficial layers and layer 5 of V1 (Sincich & Horton, 2005). There are corticogeniculate neurons in layer 6 of V2 that project to the LGN, similar to V1 corticogeniculate neurons (Briggs, Kiley, Callaway, & Usrey, 2016). Additionally, V2 contains corticothalamic neurons in layer 5 that project to the pulvinar (Kaas & Lyon, 2007; Lyon, Jain, & Kaas, 2003). It is possible that all extrastriate visual cortical areas participate in similar reciprocal connectivity loops, projecting back to the cortical areas from which they receive feedforward input and also projecting to important subcortical structures like the pulvinar, which could serve as an alternative feedforward cortico-pulvinar-cortical route for visual signals, as previously described (Sherman, 2007). Progress has been made in determining whether V2 neurons compute new information about visual features that is distinct from that computed by V1 feature detectors. First, the receptive fields of V2 neurons are approximately twice the size of V1 receptive fields at a given eccentricity (Shushruth, Ichida, Levitt, & Angelucci, 2009), so it is clear that V2 neurons integrate visual inputs over larger regions of visual space. Second, neurons in V2 respond better to visual stimuli that have more natural texture compared to V1 neurons (Freeman, Ziemba, Heeger, Simoncelli, & Movshon, 2013; Freeman, Ziemba, Movshon, Simoncelli, & Heeger, 2014; Smith, Kohn, & Movshon, 2007).

Area MT receives inputs from neurons in layer 4B of V1 (Nassi & Callaway, 2009, 2006) and neurons in the thick stripes of V2 (Sincich & Horton, 2002a, 2005). The hallmark of MT neuronal responses is strong selectivity for stimulus motion (Britten, Shadlen, Newsome, & Movshon, 1993; Newsome & Pare, 1988; Snowden, Treue, Erickson, & Anderson, 1991). The receptive fields of MT neurons are often quite large, but also scale with eccentricity, similar to the lower visual areas V1 and V2 (Felleman & Kaas, 1984). MT neurons are organized into functional columns based on preference for the direction of stimulus motion (Albright, Desimone, & Gross, 1984). MT neurons are also sensitive to pattern motion, such as a plaid pattern with upward motion that is made up of two overlapping gratings with component motion vector motions plus/minus 45 degrees from upward (Movshon & Newsome, 1996; Rust, Mante, Simoncelli, & Movshon, 2006; Smith, Majaj, & Movshon, 2005). While V1 neurons respond to each component motion vector and not the summed plaid pattern motion, MT neurons respond to the pattern motion (Movshon & Newsome, 1996; Rust et al., 2006; Smith et al., 2005). Human observers generally perceive the pattern motion and not the component motion vectors; thus, at the MT stage, individual neurons begin to respond in a manner that is consistent with perception. Correlations between individual MT neuronal responses and psychophysical behavior in primates have been observed such that activity in specific motion-selective MT neuronal populations is somewhat predictive of whether animals correctly or incorrectly guess the direction of motion in an ambiguous stimulus (Britten, Newsome, Shadlen, Celebrini, & Movshon, 1996; Salzman, Murasugi, Britten, & Newsome, 1992). MT has also been a popular cortical area in which to explore the impact of spatial attention on visual responses because MT neuronal spiking rates are substantially increased when animals direct attention toward neuronal receptive fields (Cook & Maunsell, 2002; Treue & Martinez-Trujillo, 1999; Treue & Maunsell, 1996, 1999; Womelsdorf, Anton-Erxleben, Pieper, & Treue, 2006). MT may also be a key gateway cortical area within the dorsal pathway as MT corticocortical neurons target a number of extrastriate cortical areas (Felleman & Van Essen, 1991; Maunsell & Van Essen, 1983) that are involved in perceptual decision-making and planning of eye movements (e.g., the lateral interparietal cortex; Gnadt & Mays, 1995; Gold & Shadlen, 2007; Hanks, Ditterich, & Shadlen, 2006). In general, dorsal pathway extrastriate visual areas are probably performing a number of important functions and feature detections related to the perception of moving stimuli and the planning of orienting eye and body movements that are guided by attention-demanding or salient visual objects.

Area V4 may be for the ventral pathway what area MT is for the dorsal pathway, in that it receives substantial input from both V1 and V2 (Figure 7; Felleman & Van Essen, 1991; Nakamura, Gattas, Desimone, & Ungerleider, 1993; Sincich & Horton, 2005). New feature detectors emerge in V4 including selective responses to a variety of hues and selective responses to contours and more complex shapes (Conway, 2009; Conway, Moeller, & Tsao, 2007; Kotake, Morimoto, Okazaki, Fujita, & Tamura, 2009; Kusunoki, Moutoussis, & Zeki, 2006; Pasupathy & Connor, 1999, 2001, 2002). V4 neurons are also orientation and direction selective and have binocular disparity leading some to speculate that V4 is involved in figure/ground discrimination (Roe et al., 2012). Also analogous to MT, corticocortical neurons originating in V4 project to a variety of ventral pathway extrastriate areas that are important for object recognition (e.g., inferior temporal cortex; DiCarlo & Cox, 2007; Felleman & Van Essen, 1991; Tanaka, 1993). Additionally, V4 neurons are strongly modulated by spatial attention in a manner similar to MT neurons (McAdams & Maunsell, 1999, 2000; Reynolds, Chelazzi, & Desimone, 1999; Reynolds & Desimone, 2003; Sundberg, Mitchell, Gawne, & Reynolds, 2012; Treue & Martinez-Trujillo, 1999). There is some debate as to whether V4 contains a functional architecture defined by feature maps or clusters of neurons selective for similar features (Roe et al., 2012). Imaging data suggest that neurons with hue selectivity are clustered in an orderly fashion within V4 (Conway & Tsao, 2009) and that orientation/direction selectivity is also mapped within V4 (Mountcastle, Motter, Steinmetz, & Sestokas, 1987; Zeki, 1983). Theoretical models predict that local circuits in extrastriate areas like MT and V4 must be wired such that necessary feature detectors emerge in each area, but that these circuits must also be flexible in order to encode for all possible object forms and movements (Roe et al., 2012; Singer & Gray, 1995). Additionally, these circuits must be able to adapt rapidly depending on the allocation of attention.

In the primate visual system, significant cortical territory, mainly in the inferior temporal cortical region, is devoted to the perception of faces (Desimone, 1991; Gross, Rocha-Miranda, & Bender, 1972). Functional MRI in humans has revealed brain areas specialized for processing information about human faces, bodies, etc., that are anatomically and functionally distinct from the brain areas specialized for processing other objects or natural scenes (Kanwisher, McDermott, & Chun, 1997; Puse, Allison, Asgari, Gore, & McCarthy, 1996). In primates, similar specific face processing patches in the inferior temporal cortex have also been identified (Freiwald, Tsao, & Livingstone, 2009; Tsao, Freiwald, Knutsen, Mandeville, & Tootell, 2003). Interestingly, these face patches in primates seem to have their own hierarchical organization such that the “earlier” face patches respond to cartoon faces as long as a small number of key features are present (two eyes, a nose, and a month appropriately arranged, for example; Freiwald et al., 2009). Later face patches are more selective for monkey faces, and neurons within these patches respond to the same face regardless of its position (Freiwald & Tsao, 2010). These findings suggest that similar principles guide visual perception throughout the hierarchy of visual brain structures. While much remains to be explored, especially in cortical areas outside of V1, it is perhaps comforting to realize that principles like parallel processing and integration of information from feedforward, feedback, and local circuits govern the emergence of feature detectors at each processing stage.


The visual systems of different mammals are specialized to process visual information in a manner that is most useful for the goals of the species. In highly visual mammals, the visual system is organized into parallel processing channels to enable efficient encoding of information about specific features of the visual world. In less visual mammals, the visual system is organized to integrate signals from across the brain to combine visual information with other relevant sensory and cognitive signals. Across species, similar governing principles are evident. These include parallel processing, whether that is parallel processing of visual information streams through core channels or parallel processing of sensory and contextual information through matrix channels; combination of feedforward, feedback, and local circuits to generate functional properties or control neuronal activity and timing; and integration of cognitive and contextual signals at all processing stages. It is important to combine knowledge gained from a variety of species, and using a variety of experimental tools, in order to develop a comprehensive understanding of the mammalian visual system.

Further Reading

Callaway, E. M. (2014). Cell types and local circuits in primary visual cortex of the macaque monkey. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences (pp. 353–365). Cambridge, MA: MIT Press.Find this resource:

Saalmann, Y. B., & Kastner, S. (2011). Cognitive and perceptual functions of the visual thalamus. Neuron, 71, 209–223.Find this resource:


Adams, D. L., & Horton, J. C. (2003). A precise retinotopic map of primate striate cortex generated from the representation of angioscotomas. Journal of Neuroscience, 23, 3771–3789.Find this resource:

Ahmed, B., Anderson, J. C., Douglas, R. J., Martin, K. A. C., & Nelson, J. C. (1994). Polyneuronal innervation of spiny stellate neurons in cat visual cortex. Journal of Comparative Neurology, 341, 39–49.Find this resource:

Airan, R. D., Thompson, K. R., Fenno, L. E., Bernstein, H., & Deisseroth, K. (2009). Temporally precise in vivo control of intracellular signaling. Nature, 458, 1025–1029.Find this resource:

Albright, T. D., Desimone, R., & Gross, C. G. (1984). Columnar organization of directionally selective cells in visual area MT of the macaque. Journal of Neurophysiology, 51, 16–31.Find this resource:

Alitto, H. J., & Usrey, W. M. (2008). Origin and dynamics of extraclassical suppression in the lateral geniculate nucleus of the macaque monkey. Neuron, 57, 135–146.Find this resource:

Alonso, J. M., Usrey, W. M., & Reid, R. C. (2001). Rules of connectivity between geniculate cells and simple cells in cat primary visual cortex. Journal of Neuroscience, 21, 4002–4015.Find this resource:

Andolina, I. M., Jones, H. E., & Sillito, A. M. (2013). Effects of cortical feedback on the spatial properties of relay cells in the lateral geniculate nucleus. Journal of Neurophysiology, 109, 889–899.Find this resource:

Andolina, I. M., Jones, H. E., Wang, W., & Sillito, A. M. (2007). Corticothalamic feedback enhances stimulus response precision in the visual system. Proceedings of the National Academy of Science USA, 104, 1685–1690.Find this resource:

Angelucci, A., & Bressloff, P. C. (2006). Contribution of feedforward, lateral and feedback connections to the classical receptive field center and extra-classical receptive field surround of primate V1 neurons. Progress in Brain Research, 154, 93–120.Find this resource:

Attwell, D., Werblin, F. S., Wilson, M., & Wu, S. M. (1983). A sign-reversing pathway from rods to double and single cones in the retina of the tiger salamander. Journal of Physiology, 336, 313–333.Find this resource:

Bal, T., Debay, D., & Destexhe, A. (2000). Cortical feedback controls the frequency and synchrony of oscillations in the visual thalamus. Journal of Neuroscience, 20, 7478–7488.Find this resource:

Balaram, P., Young, N. A., & Kaas, J. H. (2014). Histological features of layers and sublayers in cortical visual areas V1 and V2 of chimpanzees, macaque monkeys, and humans. Eye Brain, 6, 5–18.Find this resource:

Baylor, D. A., Nunn, B. J., & Schnapf, J. L. (1984). The photocurrent, noise, and spectral sensitivity of rods of the monkey macaca fascicularis. Journal of Physiology, 357, 575–607.Find this resource:

Bednar, J. A., & Wilson, S. P. (2016). Cortical maps. Neuroscientist, 22(6), 604–617.Find this resource:

Bell, A. H., Pasternak, T., & Ungerleider, L. G. (2014). Ventral and dorsal cortical processing streams. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences (pp. 227–242). Cambridge, MA: MIT Press.Find this resource:

Bender, D. B. (1982). Receptive-field properties of neurons in the macaque inferior pulvinar. Journal of Neurophysiology, 48, 1–17.Find this resource:

Bickford, M. E., Zhou, N., Krahe, T. E., Govindaiah, G., & Guido, W. (2015). Retinal and tectal “driver-like” inputs converge in the shell of the mouse dorsal lateral geniculate nucleus. Journal of Neuroscience, 35, 10523–10534.Find this resource:

Binzegger, T., Douglas, R. J., & Martin, K. A. C. (2007). Stereotypical bouton clustering of individual neurons in cat primary visual cortex. Journal of Neuroscience, 27, 12242–12254.Find this resource:

Blasdel, G. G., & Fitzpatrick, D. (1984). Physiological organization of layer 4 in macaque striate cortex. Journal of Neuroscience, 4, 880–895.Find this resource:

Blasdel, G. G., & Salama, G. (1986). Voltage-sensitive dyes reveal a modular organization in monkey striate cortex. Nature, 321, 579–585.Find this resource:

Bosking, W. H., Crowley, J. C., & Fitzpatrick, D. (2002). Spatial coding of position and orientation in primary visual cortex. Nature Neuroscience, 5, 874–882.Find this resource:

Bosking, W. H., Zhang, Y., Schofield, B., & Fitzpatrick, D. (1997). Orientation selectivity and the arrangement of horizontal connections in tree shrew striate cortex. Journal of Neuroscience, 17, 2112–2127.Find this resource:

Bonhoeffer, T., & Grinvald, A. (1991). Iso-orientation domains in cat visual cortex are arranged in pinwheel-like patterns. Nature, 353, 429–431.Find this resource:

Bonhoeffer, T., & Grinvald, A. (1993). The layout of iso-orientation domains in area 18 of cat visual cortex: Optical imaging reveals a pinwheel-like organization. Journal of Neuroscience, 13, 4157–4180.Find this resource:

Bonin, V., Mante, V., & Carandini, M. (2005). The suppressive field of neurons in lateral geniculate nucleus. Journal of Neuroscience, 25, 10844–10856.Find this resource:

Bridge, J., & Cumming, B. G. (2001). Responses of macaque V1 neurons to binocular orientation differences. Journal of Neuroscience, 21, 7293–7302.Find this resource:

Briggman, K. L., & Bock, D. D. (2012). Volume electron microscopy for neuronal circuit reconstruction. Current Opinion in Neurobiology, 22, 154–161.Find this resource:

Briggman, K. L., Helmstaedter, M., & Denk, W. (2011). Wiring specificity in the direction-selectivity circuit of the retina. Nature, 471, 183–188.Find this resource:

Briggs, F. (2010). Organizing principles of cortical layer 6. Frontiers in Neural Circuits, 4(3).Find this resource:

Briggs, F., & Callaway, E. M. (2001). Layer-specific input to distinct cell types in layer 6 of monkey primary visual cortex. Journal of Neuroscience, 21, 3600–3608.Find this resource:

Briggs, F., Kiley, C. W., Callaway, E. M., & Usrey, W. M. (2016). Morphological substrates for parallel streams of corticogeniculate feedback originating in both V1 and V2 of the macaque monkey. Neuron, 90(2), 388–399.Find this resource:

Briggs, F., Mangun, G. R., & Usrey, W. M. (2013). Attention enhances synaptic efficacy and the signal-to-noise ratio in neural circuits. Nature, 499, 476–480.Find this resource:

Briggs, F., & Usrey, W. M. (2005). Temporal properties of feedforward and feedback pathways between thalamus and visual cortex in the ferret. Thalamus and Related Systems, 3, 133–139.Find this resource:

Briggs, F., & Usrey, W. M. (2007). A fast, reciprocal pathway between the lateral geniculate nucleus and visual cortex in the macaque monkey. Journal of Neuroscience, 27, 5431–5436.Find this resource:

Briggs, F., & Usrey, W. M. (2009). Parallel processing in the corticogeniculate pathway of the macaque monkey. Neuron, 62, 135–146.Find this resource:

Briggs, F., & Usrey, W. M. (2011). Corticogeniculate feedback and parallel processing in the primate visual system. Journal of Physiology, 589, 33–40.Find this resource:

Britten, K. H., Newsome, W., Shadlen, M., Celebrini, S., & Movshon, J. A. (1996). A relationship between behavioral choice and the visual response of neurons in macaque MT. Visual Neuroscience, 13, 87–100.Find this resource:

Britten, K. H., Shadlen, M. N., Newsome, W. T., & Movshon, J. A. (1993). Responses of neurons in macaque MT to stochastic motion signals. Visual Neurosciences, 10, 1157–1169.Find this resource:

Callaway, E. M. (1998). Local circuits in primary visual cortex of the macaque monkey. Annual Review of Neuroscience, 21, 47–74.Find this resource:

Callaway, E. M. (2004). Cell types and local circuits in primary visual cortex of the macaque monkey. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences. Cambridge, MA: MIT Press.Find this resource:

Callaway, E. M. (2005). Structure and function of parallel pathways in the primate early visual system. Journal of Physiology, 566, 13–19.Find this resource:

Callaway, E. M., & Luo, L. (2015). Monosynaptic circuit tracing with glycoprotein-deleted rabies viruses. Journal of Neuroscience, 35, 8979–8985.Find this resource:

Carandini, M., & Heeger, D. J. (2012). Normalization as a canonical neural computation. Nature Reviews Neuroscience, 13, 51–62.Find this resource:

Cardin, J. A., Palmer, L. A., & Contreras, D. (2007). Stimulus feature selectivity in excitatory and inhibitory neurons in primary visual cortex. Journal of Neuroscience, 27, 10333–10344.Find this resource:

Cavanaugh, J. R., Bair, W., & Movshon, J. A. (2002). Nature and interaction of signals from the receptive field center and surround in macaque V1 neurons. Journal of Neurophysiology, 88, 2530–2546.Find this resource:

Chisum, H. J., Mooser, F., & Fitzpatrick, D. (2003). Emergent properties of layer 2/3 neurons reflect the collinear arrangement of horizontal connections in tree shrew visual cortex. Journal of Neuroscience, 23, 2947–2960.Find this resource:

Claps, A., & Casagrande, V. A. (1990). The distribution and morphology of corticogeniculate axons in ferrets. Brain Research, 530, 126–129.Find this resource:

Conway, B. R. (2009). Color vision, cones, and color-coding in the cortex. Neuroscientist, 15, 274–290.Find this resource:

Conway, B. R., Hubel, D. H., & Livingstone, M. S. (2002). Color contrast in macaque V1. Cerebral Cortex, 12, 915–925.Find this resource:

Conway, B. R., Moeller, S., & Tsao, D. Y. (2007). Specialized color modules in macaque extrastriate cortex. Neuron, 56, 560–573.Find this resource:

Conway, B. R., & Tsao, D. Y. (2009). Color-tuned neurons are spatially clustered according to color preference within alert macaque posterior inferior temporal cortex. Proceedings of the National Academy of Sciences USA, 106, 18034–18039.Find this resource:

Cook, E. P., & Maunsell, J. H. R. (2002). Attentional modulation of behavioral performance and neuronal responses in middle temporal and ventral intraparietal areas of macaque monkey. Journal of Neuroscience, 22, 1994–2004.Find this resource:

Crick, F. (1984). Function of the thalamic reticular complex: The searchlight hypothesis. Proceedings of the National Academy of Sciences USA, 81, 4586–4590.Find this resource:

Crook, J. D., Peterson, B. B., Packer, O. S., Robinson, F. R., Gamlin, P. D., Troy, J. B., et al. (2008). The smooth monostratified ganglion cell: Evidence for spatial diversity in the Y-cell pathway to the lateral geniculate nucleus and superior colliculus in the macaque monkey. Journal of Neuroscience, 28, 12654–12671.Find this resource:

Cudiero, J., Rivadulla, C., & Grieve, K. L. (2000). Visual response augmentation in cat (and macaque) LGN: Potentiation by corticofugally mediated gain control in the temporal domain. European Journal of Neuroscience, 12, 1135–1144.Find this resource:

Cumming, B. G., & Parker, A. J. (2000). Local disparity not perceived depth is signaled by binocular neurons in cortical area V1 of the macaque. Journal of Neuroscience, 20, 4758–4767.Find this resource:

Dacey, D. M. (2000). Parallel pathways for spectral coding in primate retina. Annual Review of Neuroscience, 23, 743–775.Find this resource:

Dacey, D. M., Crook, J. D., & Packer, O. S. (2014). Distinct synaptic mechanisms create parallel S-ON and S-OFF color opponent pathways in the primate retina. Visual Neuroscience, 31, 139–151.Find this resource:

Dacey, D. M., Lee, B. B., Stafford, D. K., Pokorny, J., & Smith, V. C. (1996). Horizontal cells of the primate retina: cone specificity without spectral opponency. Science, 271, 656–659.Find this resource:

Dacey, D. M., & Packer, O. S. (2003). Colour coding in the primate retina: Diverse cell types and cone-specific circuitry. Current Opinion in Neurobiology, 13, 421–427.Find this resource:

Denman, D. J., & Contreras, D. (2015). Complex effects on in vivo visual responses by specific projections from mouse cortical layer 6 to dorsal lateral geniculate nucleus. Journal of Neuroscience, 35, 9265–9280.Find this resource:

Desimone, R. (1991). Face-selective cells in the temporal cortex of monkeys. Journal of Cognitive Neuroscience, 3, 1–8.Find this resource:

Devries, S. H. (2014). Cone bipolar cells: ON and OFF pathways in the outer retina. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences (pp. 53–62). Cambridge, MA: MIT Press.Find this resource:

Dhande, O. S., Estevez, M. E., Quattrochi, L. E., El-Danaf, R. N., Nguyen, P. L., Berson, D. M., et al. (2013). Genetic dissection of retinal inputs to brainstem nuclei controlling image stabilization. Journal of Neuroscience, 33, 17797–17813.Find this resource:

DiCarlo, J. J., & Cox, D. D. (2007). Untangling invariant object recognition. Trends in Cognitive Sciences, 11, 333–341.Find this resource:

Douglas, R. J., & Martin, K. A. C. (2004). Neuronal circuits of the neocortex. Annual Review of Neuroscience, 27, 419–451.Find this resource:

Dow, B. M., Snyder, A. Z., Vautin, R. G., & Bauer, R. (1981). Magnification factor and receptive field size in foveal striate cortex of the monkey. Experimental Brain Research, 44, 213–228.Find this resource:

Dubin, M. W., & Cleland, B. G. (1977). Organization of visual inputs to interneurons of lateral geniculate nucleus of the cat. Journal of Neurophysiology, 40, 410–427.Find this resource:

Erisir, A., Van Horn, S. C., & Sherman, S. M. (1997). Relative numbers of cortical and brainstem inputs to the lateral geniculate nucleus. Proceedings of the National Academy of Science USA, 94, 1517–1520.Find this resource:

Euler, T., Haverkamp, S., Schubert, T., & Baden, T. (2014). Retinal bipolar cells: Elementary building blocks of vision. Nature Reviews Neuroscience, 15, 507–519.Find this resource:

Eyding, D., Macklis, J. D., Neubacher, U., Funke, K., & Worgotter, F. (2003). Selective elimination of corticogeniculate feedback abolishes the electroencephalogram dependence of primary visual cortical receptive fields and reduces their spatial specificity. Journal of Neuroscience, 23, 7021–7033.Find this resource:

Felleman, D. J., Burkhalter, A., & Van Essen, D. C. (1997). Cortical connections of areas V3 and VP of macaque monkey extrastriate cortex. Journal of Comparative Neurology, 379, 21–47.Find this resource:

Felleman, D. J., & Kaas, J. H. (1984). Receptive-field properties of neurons in middle temporal visual area (MT) of owl monkeys. Journal of Neurophysiology, 52, 488–513.Find this resource:

Felleman, D. J., & Van Essen, D. C. (1991). Distributed hierarchical processing in the primate cerebral cortex. Cerebral Cortex, 1, 1–47.Find this resource:

Field, G. D., & Chichilnisky, E. J. (2007). Information processing in the primate retina: Circuitry and coding. Annual Review of Neuroscience, 30, 1–30.Find this resource:

Field, G. D., Gauthier, J. L., Sher, A., Greschner, M., Machado, T. A., Jepson, L. H., et al. (2010). Functional connectivity in the retina at the resolution of photoreceptors. Nature, 467, 673–677.Find this resource:

Freeman, J., & Simoncelli, E. P. (2011). Metamers of the ventral stream. Nature Neuroscience, 14, 1195–1201.Find this resource:

Freeman, J., Ziemba, C. M., Heeger, D. J., Simoncelli, E. P., & Movshon, J. A. (2013). A functional and perceptual signature of the second visual area in primates. Nature Neuroscience, 16, 974–981.Find this resource:

Freeman, J., Ziemba, C. M., Movshon, J. A., Simoncelli, E. P., & Heeger, D. J. (2014). Carving up the ventral stream with controlled naturalistic stimuli. Journal of Vision, 14.Find this resource:

Freiwald, W. A., & Tsao, D. Y. (2010). Functional compartmentalization and viewpoint generalization within the macaque face-processing system. Science, 330, 845–851.Find this resource:

Freiwald, W. A., Tsao, D. Y., & Livingstone, M. S. (2009). A face feature space in the macaque temporal lobe. Nature Neuroscience, 12, 1187–1196.Find this resource:

Garrett, M. E., Nauhaus, I., Marshel, J. H., & Callaway, E. M. (2014). Topography and area organization of mouse visual cortex. Journal of Neuroscience, 34, 12587–12600.Find this resource:

Gegenfurtner, K. R. (2003). Cortical mechanisms of colour vision. Nature Reviews Neuroscience, 4, 563–572.Find this resource:

Gilbert, C. D., & Wiesel, T. N. (1979). Morphology and intracortical projections of functionally characterized neurons in the cat visual cortex. Nature, 280, 120–125.Find this resource:

Gnadt, J. W., & Mays, L. E. (1995). Neurons in monkey parietal area LIP are tuned for eye-movement parameters in three-dimensional space. Journal of Neurophysiology, 73, 280–297.Find this resource:

Gold, J. I., & Shadlen, M. N. (2007). The neural basis of decision making. Annual Review of Neuroscience, 30, 535–574.Find this resource:

Greschner, M., Shlens, J., Bakolitsa, C., Field, G. D., Gauthier, J. L., Jepson, L. H., et al. (2011). Correlated firing among major ganglion cell types in primate retina. Journal of Physiology, 589, 75–86.Find this resource:

Grieve, K. L., & Sillito, A. M. (1995). Differential properties of cells in the feline primary visual cortex providing the corticofugal feedback to the lateral geniculate nucleus and visual claustrum. Journal of Neuroscience, 15, 4868–4874.Find this resource:

Gross, C. G., Rocha-Miranda, C. E., & Bender, D. B. (1972). Visual properties of neurons in inferotemporal cortex of the macaque. Journal of Neurophysiology, 35, 96–111.Find this resource:

Gulyas, B., Lagae, L., Eysel, U. T., & Orban, G. A. (1990). Corticofugal feedback influences the responses of geniculate neurons to moving stimuli. Experimental Brain Research, 79, 441–446.Find this resource:

Gur, M., Kagan, I., & Snodderly, M. D. (2003). Orientation selectivity in V1 of alert monkeys. Journal of Vision, 3(21).Find this resource:

Gur, M., & Snodderly, M. D. (2007). Direction selectivity in V1 of alert monkeys: Evidence for parallel pathways for motion processing. Journal of Physiology, 585, 383–400.Find this resource:

Hanks, T. D., Ditterich, J., & Shadlen, M. N. (2006). Microstimulation of macaque area LIP affects decision-making in a motion discrimination task. Nature Neuroscience, 9, 682–689.Find this resource:

Hannibal, J., & Fahrenkrug, J. (2002). Melanopsin: A novel photopigment involved in the photoentrainment of the brain’s biological clock? Annals of Medicine, 34, 401–407.Find this resource:

Harris, K. D., & Thiele, A. (2011). Cortical state and attention. Nature Reviews Neuroscience, 12, 509–522.Find this resource:

Harvey, A. R. (1978). Characteristics of corticothalamic neurons in area 17 of the cat. Neuroscience Letters, 7, 177–181.Find this resource:

Hasse, J. M., & Briggs, F. (in review). Corticogeniculate feedback sharpens the temporal precision and spatial resolution of visual signals.Find this resource:

Helmstaedter, M., Briggman, K. L., Turaga, S., Jain, V., Seung, H. S., & Denk, W. (2013). Connectomic reconstruction of the inner plexiform layer in the mouse retina. Nature, 500, 168–174.Find this resource:

Hendry S. H., & Yoshioka, T. A. (1994). A neurochemically distinct third channel in the macaque dorsal lateral geniculate nucleus. Science, 264, 575–577.Find this resource:

Hillyard, S. A., Vogel, E. K., & Luck, S. J. (1998). Sensory gain control (amplification) as a mechanism of selective attention: electrophysiological and neuroimaging evidence. Philosophical Transactions of the Royal Society B, 353, 1257–1270.Find this resource:

Hirsch, J. A., Martinez, L. M., Pillai, C., Alonso, J. M., Wang, Q., & Sommer, F. T. (2003). Functionally distinct inhibitory neurons at the first stage of visual cortical processing. Nature Neuroscience, 6, 1300–1308.Find this resource:

Hirsch, J. A., Wang, X., Sommer, F. T., & Martinez, L. M. (2015). How inhibitory circuits in the thalamus serve vision. Annual Review of Neuroscience, 38, 309–329.Find this resource:

Horton, J. C., & Sincich, L. C. (2004). A new foundation for the visual cortical hierarch. In M. Gazzaniga (Ed.), The cognitive neurosciences (3d ed.). Cambridge, MA: MIT Press.Find this resource:

Horwitz, G. D., Chichilnisky, E. J., & Albright, T. D. (2007). Cone inputs to simple and complex cells in V1 of awake macaque. Journal of Neurophysiology, 97, 3070–3081.Find this resource:

Hubel, D. H. (1988). Eye, brain, and vision (2d ed.). New York: Scientific American Library.Find this resource:

Hubel, D. H., & Wiesel, T. N. (1968). Receptive fields and functional architecture of monkey striate cortex. Journal of Physiology, 195, 215–243.Find this resource:

Hubel, D. H., & Wiesel, T. N. (1974). Sequence regularity and geometry of orientation columns in the monkey striate cortex. Journal of Comparative Neurology, 158, 267–293.Find this resource:

Hull, C., Adesnik, H., & Scanziani, M. (2009). Neocortical disynaptic inhibition requires somatodendritic integration in interneurons. Journal of Neuroscience, 29, 8991–8995.Find this resource:

Hunt, D. M., Dulai, K. S., Cowing, J. A., Julliot, C., Mollon, J. D., Bowmaker, J. K., et al. (1998). Molecular evolution of trichromacy in primates. Vision Research, 38, 3299–3306.Find this resource:

Hunt, D. M., & Peichl, L. (2014). S cones: Evolution, retinal distribution, development, and spectral sensitivity. Visual Neuroscience, 31, 115–138.Find this resource:

Ichida, J. M., Mavity-Hudson, J. A., & Casagrande, V. A. (2014). Distinct patterns of corticogeniculate feedback to different layers of the lateral geniculate nucleus. Eye and Brain, 6, 57–73.Find this resource:

Johnson, E. N., Hawken, M. J., & Shapley, R. (2008). The orientation selectivity of color-responsive neurons in macaque V1. Journal of Neuroscience, 28, 8096–8106.Find this resource:

Jones, E. G. (1998). Viewpoint: The core and matrix of thalamic organization. Neuroscience, 85, 331–345.Find this resource:

Kaas, J. H. (2005). The evolution of visual cortex in primates. In J. Kremers (Ed.), The structure, function and evolution of the primate visual system. Chichester, U.K.: John Wiley and Sons.Find this resource:

Kaas, J. H. (2013a). The evolution of brains from early mammals to humans. Wiley Interdisciplinary Review Cognitive Science, 4, 33–45.Find this resource:

Kaas, J. H. (2013b). Evolution of columns, modules, and domains in the neocortex of primates In G. F. Streidter, J. C. Avise, & F. J. Ayala (Eds.), In the light of evolution. Vol. 6: Brain and behavior. Washington, DC: National Academy Press.Find this resource:

Kaas, J. H., & Lyon, D. C. (2007). Pulvinar contributions to the dorsal and ventral streams of visual processing in primates. Brain Research Reviews, 55, 285–296.Find this resource:

Kagan, I., Gur, M., & Snodderly, D. M. (2002). Spatial organization of receptive fields of V1 neurons of alert monkeys: Comparison with responses to gratings. Journal of Neurophysiology, 88, 2557–2574.Find this resource:

Kanwisher, N., McDermott, J., & Chun, M. M. (1997). The fusiform face area: A module in human extrastriate cortex specialized for face perception. Journal of Neuroscience, 17, 4302–4311.Find this resource:

Kaplan, E. (2004). The M, P, and Konio pathways in the primate visual system. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences. Cambridge, MA: MIT Press.Find this resource:

Karklin, Y., & Lewicki, M. S. (2003). Learning higher-order structures in natural images. Network, 14, 483–499.Find this resource:

Kaskan, P. M., Lu, H. D., Dillenburger, B. C., Roe, A. W., & Kaas, J. H. (2007). Intrinsic-signal optical imaging reveals cryptic ocular dominance columns in primary visual cortex of New World owl monkeys. Frontiers in Neuroscience, 1, 67–75.Find this resource:

Klug, K., Herr, S., Ngo, I. T., Sterling, P., & Schein, S. (2003). Macaque retina contains an S-cone OFF midget pathway. Journal of Neuroscience, 23, 9881–9887.Find this resource:

Kotake, Y., Morimoto, H., Okazaki, Y., Fujita, I., & Tamura, H. (2009). Organization of color-selective neurons in macaque visual area V4. Journal of Neurophysiology, 102, 15–27.Find this resource:

Krahe, T. E., El-Danaf, R. N., Dilger, E. K., Henderson, S. C., & Guido, W. (2011). Morphologically distinct classes of relay cells exhibit regional preferences in the dorsal lateral geniculate nucleus of the mouse. Journal of Neuroscience, 31, 17437–17448.Find this resource:

Krauzlis, R. J. (2014). Superior colliculus and visual attention. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences (pp. 323–337). Cambridsge, MA: MIT Press.Find this resource:

Kravitz, D. J., Saleem, K. S., Baker, C. I., & Mishkin, M. (2011). A new neural framework for visuospatial processing. Nature Reviews Neuroscience, 12, 217–230.Find this resource:

Kuffler, S. W. (1953). Discharge patterns and functional organization of mammalian retina. Journal of Neurophysiology, 16, 37–68.Find this resource:

Kusunoki, M., Moutoussis, K., & Zeki, S. (2006). Effect of background colors on the tuning of color-selective cells in monkey area V4. Journal of Neurophysiology, 95, 3047–3059.Find this resource:

Lamb, T. D. (2013). Evolution of phototransduction, vertebrate photoreceptors and retina. Progress Retina Eye Research, 36, 52–119.Find this resource:

Lamb, T. D., Collin, S. P., & Pugh, E. N. (2007). Evolution of the vertebrate eye: Opsins, photoreceptors, retina and eye cup. Nature Reviews Neuroscience, 8, 960–976.Find this resource:

Landisman, C. E., Long, M. A., Beierlein, M., Deans, M. R., Paul, D. L., & Connors, B. W. (2002). Electrical synapses in the thalamic reticular nucleus. Journal of Neuroscience, 22, 1002–1009.Find this resource:

LeVay, S., Hubel, D. H., & Wiesel, T. N. (1975). The pattern of ocular dominance columns in macaque visual cortex revealed by a reduced silver stain. Journal of Comparative Neurology, 159, 559–575.Find this resource:

LeVay, S., & Sherk, H. (1981). The visual claustrum of the cat I: Structure and connections. Journal of Neuroscience, 1, 956–980.Find this resource:

Lewis, L. D., Voigts, J., Flores, F. J., Schmitt, L. I., Wilson, M. A., Halassa, M. M., et al. (2015). Thalamic reticular nucleus induces fast and local modulation of arousal state. eLife, 4: e08760.Find this resource:

Li, G., Ye, X., Song, T., Yang, Y., & Zhou, Y. (2011). Contrast adaptation in cat lateral geniculate nucleus and influence of corticothalamic feedback. European Journal of Neuroscience, 34, 622–631.Find this resource:

Lima, B., Singer, W., & Neuenschwander, S. (2011). Gamma responses correlate with temporal expectation in monkey primary visual cortex. Journal of Neuroscience, 31, 15919–15931.Find this resource:

Livingstone, M. S., & Hubel, D. H. (1984). Anatomy and physiology of a color system in the primate visual cortex. Journal of Neuroscience, 4, 309–356.Find this resource:

Livingstone, M. S., & Hubel, D. H. (1988). Segregation of form, color, movements, and depth: Anatomy, physiology, and perception. Science, 240, 740–749.Find this resource:

Lu, H. D., & Roe, A. W. (2008). Functional organization of color domains in V1 and V2 of macaque monkey revealed by optical imaging. Cerebral Cortex, 18, 516–533.Find this resource:

Luck, S. J., Chelazzi, L., Hillyard, S. A., & Desimone, R. (1997). Neural mechanisms of spatial selective attention in area V1, V2, and V4 of macaque visual cortex. Journal of Neurophysiology, 77, 24–42.Find this resource:

Lund, J. S. (1987). Local circuit neurons of macaque monkey striate cortex: I. Neurons of laminae 4C and 5A. Journal of Comparative Neurology, 257, 60–92.Find this resource:

Lyon, D. C., Jain, N., & Kaas, J. H. (2003). The visual pulvinar in tree shrews. II: Projections of four nuclei to areas of visual cortex. Journal of Comparative Neurology, 467, 607–627.Find this resource:

Lyon, D. C., & Kaas, J. H. (2002). Evidence for a modified V3 with dorsal and ventral halves in macaque monkeys. Neuron, 33, 453–461.Find this resource:

Mariani, A. P. (1984). Bipolar cells in monkey retina selective for the cones likely to be blue-sensitive. Nature, 308, 184–186.Find this resource:

Marrocco, R. T., McClurkin, J. W., & Alkire, M. T. (1996). The influence of the visual cortex on the spatiotemporal response properties of lateral geniculate nucleus cells. Brain Research, 737, 110–118.Find this resource:

Marshel, J. H., Garrett, M. E., Nauhaus, I., & Callaway, E. M. (2011). Functional specialization of seven mouse visual cortical areas. Neuron, 72, 1040–1054.Find this resource:

Marshel, J. H., Mori, T., Nielsen, K. J., & Callaway, E. M. (2010). Targeting signal neuronal networks for gene expression and cell labeling in vivo. Neuron, 67, 562–574.Find this resource:

Martinez, L. M., Molano-Mazon, M., Wang, X., Sommer, F. T., & Hirsch, J. A. (2014). Statistical wiring of thalamic receptive fields optimizes spatial sampling of the retinal image. Neuron, 81, 943–956.Find this resource:

Martinez, L. M., Wang, Q., Reid, R. C., Pillai, C., Alonso, J. M., Sommer, F. T., et al. (2005). Receptive field structure varies with layer in the primary visual cortex. Nature Neuroscience, 8, 372–379.Find this resource:

Masland, R. H. (2001). The fundamental plan of the retina. Nature Neuroscience, 4, 877–886.Find this resource:

Maunsell, J. H., & Van Essen, D. C. (1983). The connections of the middle temporal visual area (MT) and their relationship to cortical hierarchy in the macaque monkey. Journal of Neuroscience, 3, 2563–2586.Find this resource:

McAdams, C. J., & Maunsell, J. H. R. (1999). Effects of attention on orientation-tuning functions of single neurons in macaque cortical area V4. Journal of Neuroscience, 19, 431–441.Find this resource:

McAdams, C. J., & Maunsell, J. H. R. (2000). Attention to both space and feature modulates neuronal responses in macaque area V4. Journal of Neurophysiology, 83, 1751–1755.Find this resource:

McAdams, C. J., & Reid, R. C. (2005). Attention modulates the responses of simple cells in monkey primary visual cortex. Journal of Neuroscience, 25, 11023–11033.Find this resource:

McAlonan, K., Cavanaugh, J. R., & Wurtz, R. H. (2008). Guarding the gateway to cortex with attention in visual thalamus. Nature, 456, 391–394.Find this resource:

McCormick, D. A., Connors, B. W., Lighthall, J. W., & Prince, D. A. (1985). Comparative electrophysiology of pyramidal and sparsely spiny stellate neurons of the neocortex. Journal of Neurophysiology, 54, 782–806.Find this resource:

McLaughlin, D., Shapley, R., Shelley, M., & Wielaard, D. J. (2000). A neuronal network model of macaque primary visual cortex (V1): Orientation selectivity and dynamics in the input layer 4Calpha. Proceedings of the National Academy of Science USA, 97, 8087–8092.Find this resource:

Mountcastle, V. B., Motter, B. C., Steinmetz, M. A., & Sestokas, A. K. (1987). Common and differential effects of attentive fixation on the excitability of parietal and prestriate (V4) cortical visual neurons in the macaque monkey. Journal of Neuroscience, 7, 2239–2255.Find this resource:

Movshon, J. A., & Newsome, W. T. (1996). Visual response properties of striate cortical neurons projecting to area MT in macaque monkeys. Journal of Neuroscience, 16, 7733–7741.Find this resource:

Mrsic-Flogel, T. D., Hofer, S. B., Ohki, K., Reid, R. C., Bonhoeffer, T., & Hubener, M. (2007). Homeostatic regulation of eye-specific responses in visual cortex during ocular dominance plasticity. Neuron, 54, 961–972.Find this resource:

Murakami, T., Yoshida, T., Matsui, T., & Ohki, K. (2015). Wide-field Ca2+ imaging reveals visually evoked activity in the retrosplenial area. Frontiers in Molecular Neuroscience, 8, 20.Find this resource:

Nakamura, H., Gattas, R., Desimone, R., & Ungerleider, L. G. (1993). The modular organization of projections from areas V1 and V2 to areas V4 and TEO in macaques. Journal of Neuroscience, 13, 3681–3691.Find this resource:

Nassi, J. J., & Callaway, E. M. (2006). Multiple circuits relaying primate parallel visual pathways to the middle temporal area. Journal of Neuroscience, 26, 12789–12798.Find this resource:

Nassi, J. J., & Callaway, E. M. (2009). Parallel processing strategies of the primate visual system. Nature Reviews Neuroscience, 10, 360–372.Find this resource:

Nauhaus, I., Nielsen, K. J., Disney, A. A., & Callaway, E. M. (2012). Orthogonal micro-organization of orientation and spatial frequency in primate primary visual cortex. Nature Neuroscience, 15, 1683–1690.Find this resource:

Nelson, R. (1982). AII amacrine cells quicken time course of rod signals in the cat retina. Journal of Neurophysiology, 47, 928–947.Find this resource:

Newsome, W. T., & Pare, E. B. (1988). A selective impairment of motion perception following lesions of the middle temporal visual area (MT). Journal of Neuroscience, 8, 2201–2211.Find this resource:

Nhan, H. L., & Callaway, E. M. (2012). Morphology of superior colliculus- and middle temporal area-projecting neurons in primate primary visual cortex. Journal of Comparative Neurology, 520, 52–80.Find this resource:

Niell, C. M. (2015). Cell types, circuits, and receptive fields in the mouse visual cortex. Annual Reviews of Neuroscience, 38, 413–431.Find this resource:

Niell, C. M., & Stryker, M. P. (2010). Modulation of visual responses by behavioral state in mouse visual cortex. Neuron, 65, 472–479.Find this resource:

Noudoost, B., Chang, M. H., Steinmetz, N. A., & Moore, T. (2010). Top-down control of visual attention. Current Opinion in Neurobiology, 20, 183–190.Find this resource:

O’Connor, D. H., Fukui, M. M., Pinsk, M. A., & Kastner, S. (2002). Attention modulates responses in the human lateral geniculate nucleus. Nature Neuroscience, 5, 1203–1209.Find this resource:

Ohki, K., Chung, S., Ch’ng, Y. H., Kara, P., & Reid, R. C. (2005). Functional imaging with cellular resolution reveals precise micro-architecture in visual cortex. Nature, 433, 597–603.Find this resource:

Ohki, K., Chung S., Kara, P., Hubener, M., Bonhoeffer, T., & Reid, R. C. (2006). Highly ordered arrangement of single neurons in orientation pinwheels. Nature, 442, 925–928.Find this resource:

Ohki, K., & Reid, R. C. (2007). Specificity and randomness in the visual cortex. Current Opinion in Neurobiology, 17, 401–407.Find this resource:

Ohki, K., & Reid, R. C. (2014). In vivo two-photon calcium imaging in the visual system. Cold Spring Harbor Protocols, 2014, 402–416.Find this resource:

O’Keefe, L. P., Levitt, J. B., Kiper, D. C., Shapley, R. M., & Movshon, J. A. (1998). Functional organization of owl monkey lateral geniculate nucleus and visual cortex. Journal of Neurophysiology, 80, 594–609.Find this resource:

O’Kusky, J., & Colonnier, M. (1982). A laminar analysis of the number of neurons, glia, and synapses in the visual cortex (area 17) of adult macaque monkeys. Journal of Comparative Neurology, 210, 278–290.Find this resource:

Olsen, S. R., Bortone, D. S., Adesnik, H., & Scanziani, M. (2012). Gain control by layer six in cortical circuits of vision. Nature, 483, 47–52.Find this resource:

Pasik, T., Pasik, P., Hamori, J., & Szentagothai, J. (1973). “Triadic” synapses and other articulations of interneurons in the lateral geniculate nucleus of rhesus monkeys. Transactions of the American Neurological Association, 98, 293–295.Find this resource:

Pasupathy, A., & Connor, C. E. (1999). Responses to contour features in macaque area V4. Journal of Neurophysiology, 82, 2490–2502.Find this resource:

Pasupathy, A., & Connor, C. E. (2001). Shape representation in area V4: Position-specific tuning for boundary conformation. Journal of Neurophysiology, 86, 2505–2519.Find this resource:

Pasupathy, A., & Connor, C. E. (2002). Population coding of shape in area V4. Nature Neuroscience, 5, 1332–1338.Find this resource:

Petersen, S. E., Robinson, D. L., & Keys, W. (1985). Pulvinar nuclei of the behaving rhesus monkey: Visual responses and their modulation. Journal of Neurophysiology, 54, 867–886.Find this resource:

Pickard, G. E., & Sollars, P. J. (2012). Intrinsically photosensitive retinal ganglion cells. Review Physiological Biochemistry Pharmacology, 162, 59–90.Find this resource:

Pinault, D. (2004). The thalamic reticular nucleus: Structure, function and concept. Brain Research Reviews, 46, 1–31.Find this resource:

Przybyszewski, A. W., Gaska, J. P., Foote, W., & Pollen, D. A. (2000). Striate cortex increases contrast gain of macaque LGN neurons. Visual Neuroscience, 17, 485–494.Find this resource:

Purushothaman, G., Marion, R., Li, K., & Casagrande, V. A. (2012). Gating and control of primary visual cortex by pulvinar. Nature Neuroscience, 15, 905–912.Find this resource:

Puse, A., Allison, T., Asgari, M., Gore, J. C., & McCarthy, G. (1996). Differential sensitivity of human visual cortex to faces, letterstrings, and textures: A functional magnetic resonance imaging study. Journal of Neuroscience, 16, 5205–5215.Find this resource:

Ramon y Cajal, S. (1892). La retine des vertebres. Lierre, France: Typ. de Joseph van In & Cie.Find this resource:

Ramsden, B. M., Hung, C. P., & Roe, A. W. (2001). Real and illusory contour processing in area V1 of the primate: A cortical balancing act. Cerebral Cortex, 11, 648–665.Find this resource:

Rapaport, D. H., & Stone, J. (1984). The area centralis of the retina in the cat and other mammals: Focal point for function and development of the visual system. Neuroscience, 11, 289–301.Find this resource:

Rathbun, D. L., Warland, D. K., & Usrey, W. M. (2010). Spike timing and information transmission at retinogeniculate synapses. Journal of Neuroscience, 30, 13558–13566.Find this resource:

Reid, R. C., & Alonso, J. M. (1996). The processing and encoding of information in the visual cortex. Current Opinion in Neurobiology, 6, 475–480.Find this resource:

Reid, R. C., Victor, J. D., & Shapley, R. M. (1997). The use of m-sequences in the analysis of visual neurons: linear receptive field properties. Visual Neurosciences, 14, 1015–1027.Find this resource:

Reynolds, J. H., Chelazzi, L., & Desimone, R. (1999). Competitive mechanisms subserve attention in macaque areas V2 and V4. Journal of Neuroscience, 19, 1736–1753.Find this resource:

Reynolds, J. H., & Desimone, R. (2003). Interacting roles of attention and visual salience in V4. Neuron, 37, 853–863.Find this resource:

Riesenhuber, M., & Poggio, T. (1999). Hierarchical models of object recognition in cortex. Nature Neuroscience, 2, 1019–1025.Find this resource:

Ringach, D., Hawken, M. J., & Shapley, R. (1997). Dynamics of orientation tuning in macaque primary visual cortex. Nature, 387, 281–284.Find this resource:

Ringach, D., Hawken, M. J., & Shapley, R. (2003). Dynamics of orientation tuning in macaque V1: The role of global and tuned suppression. Journal of Neurophysiology, 90, 342–352.Find this resource:

Ringach, D. L., Shapley, R. M., & Hawken, M. J. (2002). Orientation selectivity in macaque V1: diversity and laminar dependence. Journal of Neuroscience, 22, 5639–5651.Find this resource:

Rockland, K. S., & Virga, A. (1989). Terminal arbors of individual “feedback” axons projecting from area V2 to V1 in the macaque monkey: A study using immunohistochemistry of anterogradely transported Phaseolus vulgaris-leucoagglutinin. Journal of Comparative Neurology, 285, 54–72.Find this resource:

Roe, A. W., Chelazzi, L., Connor, C. E., Conway, B. R., Fujita, I., Gallant, J. L., et al. (2012). Toward a unified theory of visual area V4. Neuron, 74, 12–29.Find this resource:

Roerig, B., & Chen, B. (2002). Relationships of local inhibitory and excitatory circuits to orientation preference maps in ferret visual cortex. Cerebral Cortex, 12, 187–198.Find this resource:

Roerig, B., Chen, B., & Kao, J. P. Y. (2003). Different inhibitory synaptic input patterns in excitatory and inhibitory layer 4 neurons of ferret visual cortex. Cerebral Cortex, 13, 350–363.Find this resource:

Rust, N. C., Mante, V., Simoncelli, E. P., & Movshon, J. A. (2006). How MT cells analyze the motion of visual patterns. Nature Neuroscience, 9, 1421–1431.Find this resource:

Saalmann, Y. B., & Kastner, S. (2009). Gain control in the visual thalamus during perception and cognition. Current Opinion Neurobiology, 19, 408–414.Find this resource:

Saalmann, Y. B., Pinsk, M. A., Wang, L., Li, X., & Kastner, S. (2012). The pulvinar regulates information transmission between cortical areas based on attention demands. Science, 337, 753–756.Find this resource:

Saleem, A. B., Ayaz, A., Jeffery, K. J., Harris, K. D. & Carandini, M. (2013). Integration of visual motion and locomotion in mouse visual cortex. Nature Neuroscience, 16, 1864–1869.Find this resource:

Salin, P. A., & Bullier, J. (1995). Corticocortical connections in the visual system: Structure and function. Physiological Reviews, 75, 107–154.Find this resource:

Salzman, C. D., Murasugi, C. M., Britten, K. H., & Newsome, W. T. (1992). Microstimulation in visual area MT: effects on direction discrimination performance. Journal of Neuroscience, 12, 2331–2355.Find this resource:

Sanayei, M., Herrero, J. L., Distler, C., & Thiele, A. (2015). Attention and normalization circuits in macaque V1. European Journal of Neuroscience, 41, 947–962.Find this resource:

Sawatari, A., & Callaway, E. M. (2000). Diversity and cell type specificity of local excitatory connections to neurons in layer 3B of monkey primary visual cortex. Neuron, 25, 459–471.Find this resource:

Scanziani, M., & Hausser, M. (2009). Electrophysiology in the age of light. Nature, 461, 930–939.Find this resource:

Sceniak, M. P., Hawken, M. J., & Shapley, R. (2001). Visual spatial characterization of macaque V1 neurons. Journal of Neurophysiology, 85, 1873–1887.Find this resource:

Sceniak, M. P., Hawken, M. J., & Shapley, R. (2002). Contrast-dependent changes in spatial frequency tuning of macaque V1 neurons: Effects of a changing receptive field size. Journal of Neurophysiology, 88, 1363–1373.Find this resource:

Scholl, B., Tan, A. Y. Y., Corey, J., & Priebe, N. J. (2013). Emergence of orientation selectivity in the mammalian visual pathway. Journal of Neuroscience, 33, 10616–10624.Find this resource:

Schuett, S., Bonhoeffer, T., & Hubener, M. (2002). Mapping retinotopic structure in mouse visual cortex with optical imaging. Journal of Neuroscience, 22, 6549–6559.Find this resource:

Sellers, K. K., Bennett, D. V., Hutt, A., Williams, J. H., & Frohlich, F. (2015). Awake vs. anesthetized: Layer-specific sensory processing in visual cortex and functional connectivity between cortical areas. Journal of Neurophysiology, 113, 3798–3815.Find this resource:

Serre, T., Kreiman, G., Kouh, M., Cadieu, C., Knoblich, U., & Poggio, T. (2007a). A quantitative theory of immediate visual recognition. Progress in Brain Research, 165, 33–56.Find this resource:

Serre, T., Oliva, A., & Poggio, T. (2007b). A feedforward architecture accounts for rapid categorization. Proceedings of the National Academy of Science USA, 104, 6424–6429.Find this resource:

Shapley, R., & Hawken, M. J. (2011). Color in the cortex: single- and double-opponent cells. Vision Research, 51, 701–717.Find this resource:

Sherk, H. (1989). Visual response properties of cortical inputs to an extrastriate cortical area in the cat. Visual Neuroscience, 3, 249–265.Find this resource:

Sherman, S. (2001). Tonic and burst firing: Dual modes of thalamocortical relay. Trends in Neurosciences, 24, 122–126.Find this resource:

Sherman, S. M. (2007). The thalamus is more than just a relay. Current Opinion in Neurobiology, 17, 1–6.Find this resource:

Sherman, S. M. (2016). Thalamus plays a central role in ongoing cortical functioning. Nature Neuroscience, 19, 533–541.Find this resource:

Sherman, S. M., & Guillery, R. W. (1996). Functional organization of thalamocortical relays. Journal of Neurophysiology, 76, 1367–1395.Find this resource:

Sherman, S. M., & Guillery, R. W. (1998). On the actions that one nerve cell can have on another: distinguishing “drivers” from “modulators.” Proceedings of the National Academy of Sciences USA, 95, 7121–7126.Find this resource:

Sherman, S. M., & Guillery, R. W. (2002). The role of the thalamus in the flow of information to the cortex. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 357, 1695–1708.Find this resource:

Sherman, S. M., & Guillery, R. W. (2009). Exploring the thalamus and its role in cortical function (2d ed.). Cambridge, MA: MIT Press.Find this resource:

Sherman, S. M., & Guillery, R. W. (2014). The lateral geniculate nucleus and pulvinar. In L. Chalupa & J. Werner (Eds.), The new visual neurosciences (pp. 257–283). Cambridge, MA: MIT Press.Find this resource:

Sherman, S. M., & Spear, P. D. (1982). Organization of visual pathways in normal and visually deprived cats. Physiological Reviews, 62, 738–850.Find this resource:

Shipp, S., & Zeki, S. (2002a). The functional organization of area V2. I: Specialization across stripes and layers. Visual Neuroscience, 19, 187–210.Find this resource:

Shipp, S., & Zeki, S. (2002b). The functional organization of area V2. II: The impact of stripes on visual topography. Visual Neuroscience, 19, 211–231.Find this resource:

Shushruth, S., Ichida, J. M., Levitt, J. B., & Angelucci, A. (2009). Comparison of spatial summation properties of neurons in macaque V1 and V2. Journal of Neurophysiology, 102, 2069–2083.Find this resource:

Sillito, A. M., Jones, H. E., Gerstein, G. L., & West, D. C. (1994). Feature-linked synchronization of thalamic relay cell firing induced by feedback from the visual cortex. Nature, 369, 479–482.Find this resource:

Simoncelli, E. P., & Olshausen, B. A. (2001). Natural image statistics and neural representation. Annual Review of Neuroscience, 24, 1193–1216.Find this resource:

Sincich, L. C., & Horton, J. C. (2002a). Divided by cytochrome oxidase: A map of the projections from V1 to V2 in macaques. Science, 295, 1734–1737.Find this resource:

Sincich, L. C., & Horton, J. C. (2002b). Pale cytochrome oxidase stripes in V2 receive the richest projection from macaque striate cortex. Journal of Comparative Neurology, 447, 18–33.Find this resource:

Sincich, L. C., & Horton, J. C. (2005). The circuitry of V1 and V2: Intergration of color, form, and motion. Annual Review of Neuroscience, 28, 303–326.Find this resource:

Sincich, L. C., Horton, J. C., & Sharpee, T. O. (2009). Preserving information in neural transmission. Journal of Neuroscience, 29, 6207–6216.Find this resource:

Singer, W., & Gray, C. (1995). Visual feature integration and the temporal correlation hypothesis. Annual Review of Neuroscience, 18, 555–586.Find this resource:

Smith, M. A., Kohn, A., & Movshon, J. A. (2007). Glass pattern responses in macaque V2 neurons. Journal of Vision, 7(3), 5.Find this resource:

Smith, M. A., Majaj, N. J., & Movshon, J. A. (2005). Dynamics of motion signaling by neurons in macaque area MT. Nature Neuroscience, 8, 220–228.Find this resource:

Snowden, R. J., Treue, S., Erickson, R. G., & Anderson, R. A. (1991). The response of area MT and V1 neurons to transparent motion. Journal of Neuroscience, 11, 2768–2785.Find this resource:

Solomon, S. G., Peirce, J. W., & Lennie, P. (2004). The impact of suppressive surrounds on chromatic properties of cortical neurons. Journal of Neuroscience, 24, 148–160.Find this resource:

Solomon, S. G., & Lennie, P. (2007). The machinery of colour vision. Nature Reviews Neuroscience, 8, 276–286.Find this resource:

Standage, G. P., & Benevento, L. A. (1983). The organization of connections between the pulvinar and visual area MT in the macaque monkey. Brain Research, 262, 288–294.Find this resource:

Stanisor, L., van der Togt, C., Pennartz, C. M. A., & Roelfsema, P. R. (2013). A unified selection signal for attention and reward in primary visual cortex. Proceedings of the National Academy of Science USA, 110, 9136–9141.Find this resource:

Stepniewska, I., & Kaas, J. H. (1996). Topographic patterns of V2 cortical connections in macaque monkeys. Journal of Comparative Neurology, 371, 129–152.Find this resource:

Steriade, M. (2001). Impact of network activities on neuronal properties in corticothalamic systems. Journal of Neurophysiology, 86, 1–39.Find this resource:

Steriade, M., & Deschenes, M. (1984). The thalamus as a neuronal oscillator. Brain Research, 320, 1–63.Find this resource:

Stettler, D. D., Das, A., Bennett, J., & Gilbert, C. D. (2002). Lateral connectivity and contextual interactions in macaque primary visual cortex. Neuron, 36, 739–750.Find this resource:

Stoelzel, C. R., Bereshpolova, Y., & Swadlow, H. A. (2009). Stability of thalamocortical synaptic transmission across awake brain states. Journal of Neuroscience, 29, 6851–6859.Find this resource:

Stratford, K. J., Tarczy-Hornoch, K., Martin, K. A. C., Bannister, N. J., & Jack, J. J. B. (1996). Excitatory synaptic inputs to spiny stellate cells in cat visual cortex. Nature, 382, 258–261.Find this resource:

Sundberg, K. A., Mitchell, J. F., Gawne, T. J., & Reynolds, J. H. (2012). Attention influences single unit and local field potential response latencies in visual cortical area V4. Journal of Neuroscience, 32, 16040–16050.Find this resource:

Swadlow, H. A., & Weyand, T. G. (1987). Corticogeniculate neurons, corticotectal neurons, and suspected interneurons in visual cortex of awake rabbits: receptive-field properties, axonal properties, and effects of EEG arousal. Journal of Neurophysiology, 57, 977–1001.Find this resource:

Tailby, C., Solomon, S. G., Peirce, J. W., & Metha, A. B. (2004). Two expressions of “surround suppression” in V1 that arise independent of cortical mechanisms of suppression. Visual Neuroscience, 24, 99–109.Find this resource:

Tanagawa, H., Lu, H. D., & Roe, A. W. (2010). Functional organization of color and orientation in macaque V4. Nature Neuroscience, 13, 1542–1548.Find this resource:

Tanaka, K. (1993). Neuronal mechanisms of object recognition. Science, 262, 685–688.Find this resource:

Taniguchi, H. (2014). Genetic dissection of GABAergic neural circuits in mouse neocortex. Frontiers in Cellular Neuroscience, 8.Find this resource:

Thiele, A., Pooresmaeili, A., Delicato, L. S., Herrero, J. L., & Roelfsema, P. R. (2009). Additive effects of attention and stimulus contrast in primary visual cortex. Cerebral Cortex, 19, 2970–2981.Find this resource:

Thomson, A. M., & Bannister, A. P. (1998). Postsynaptic pyramidal target selection by descending layer 3 pyramidal axons: Dual intracellular recordings and biocytin filling in slices of rat neocortex. Neuroscience, 84, 669–683.Find this resource:

Thomson, A. M., & Bannister, A. P. (2003). Interlaminar connections in the neocortex. Cerebral Cortex, 13, 5–14.Find this resource:

Thoreson, W. M., & Mangel, S. C. (2012). Lateral intteractions in the outer retina. Progress in Retinal Eye Research, 31, 407–441.Find this resource:

Timofeev, I., & Steriade, M. (1998). Cellular mechanisms underlying intrathalamic augmenting responses of reticular and relay neurons. Journal of Neurophysiology, 79, 2716–2729.Find this resource:

Treue, S., & Martinez-Trujillo, J. (1999). Feature-based attention influences motion processing gain in macaque visual cortex. Nature, 399, 575–579.Find this resource:

Treue, S., & Maunsell, J. H. (1996). Attentional modulation of visual motion processing in cortical areas MT and MST. Nature, 382, 539–541.Find this resource:

Treue, S., & Maunsell, J. H. (1999). Effects of attention on the processing of motion in macaque middle temporal and medial superior temporal visual cortical areas. Journal of Neuroscience, 19, 7591–7602.Find this resource:

Tsao, D. Y., Freiwald, W. A., Knutsen, T. A., Mandeville, J. B., & Tootell, R. B. H. (2003). Faces and objects in macaque cerebral cortex. Nature Neuroscience, 6, 989–995.Find this resource:

Ts’o, D. Y., & Gilbert, C. D. (1988). The organization of chromatic and spatial interactions in the primate striate cortex. Journal of Neuroscience, 8, 1712–1727.Find this resource:

Tsumoto, T., & Suda, K. (1980). Three groups of cortico-geniculate neurons and their distribution in binocular and monocular segments of cat striate cortex. Journal of Comparative Neurology, 193, 223–236.Find this resource:

Ungerleider, L. G., & Desimone, R. (1986). Cortical connections of visual area MT in the macaque. Journal of Comparative Neurology, 248, 190–222.Find this resource:

Ungerleider, L. G., & Haxby, J. V. (1994). “What” and “where” in the human brain. Current Opinion in Neurobiology, 4, 157–165.Find this resource:

Usrey, W. M., Reppas, J., & Reid, R. C. (1998). Paired-spike interactions and synaptic efficacy of retinal inputs to the thalamus. Nature, 395, 384–387.Find this resource:

Vanduffel, W., Tootell, R. B. H., & Orban, G. (2000). Attention-dependent suppression of metabolic activity in the early stages of the macaque visual system. Cerebral Cortex, 10, 109–126.Find this resource:

Van Hooser, S. D., Heimel, J. A., Chung, S., & Nelson, S. B. (2006). Lack of patchy horizontal connectivity in primary visual cortex of a mammal without orientation maps. Journal of Neuroscience, 26, 7680–7692.Find this resource:

Van Horn, S. C., Erisir, A., & Sherman, S. M. (2000). Relative distribution of synapses in the A-laminae of the lateral geniculate nucleus of the cat. Journal of Comparative Neurology, 416, 509–520.Find this resource:

Vinje, W. E., & Gallant, J. L. (2000). Sparse coding and decorrelation in primary visual cortex during natural vision. Science, 287, 1273–1276.Find this resource:

Vintch, B., Movshon, J. A., & Simoncelli, E. P. (2015). A convolutional subunit model for neuronal responses in macaque V1. Journal of Neuroscience, 35, 14829–14841.Find this resource:

Wallis, G., & Rolls, E. T. (1997). Invariant face and object recognition in the visual system. Progress in Neurobiology, 51, 167–194.Find this resource:

Wallisch, P., & Movshon, J. A. (2008). Structure and function come unglued in the visual cortex. Neuron, 60, 195–197.Find this resource:

Wassle, H. (2004). Parallel processing in the mammalian retina. Nature Reviews Neuroscience, 5, 747–757.Find this resource:

Weber, A. J., Kalil, R. E., & Behan, M. (1989). Synaptic connections between corticogeniculate axons and interneurons in the dorsal lateral geniculate nucleus of the cat. Journal of Comparative Neurology, 289, 156–164.Find this resource:

Weliky, M., Fiser, J., Hunt, R. H., & Wagner, D. N. (2003). Coding of natural scenes in primary visual cortex. Neuron, 37, 703–718.Find this resource:

Weliky, M., Kandler, K., Fitzpatrick, D., & Katz, L. C. (1995). Patterns of excitation and inhibition evoked by horizontal connections in visual cortex share a common relationship to orientation columns. Neuron, 15, 541–552.Find this resource:

Westheimer, G. (2008). Illusions in the spatial sense of the eye: Geometrical-optical illusions and the neural representations of space. Vision Research, 48, 2128–2142.Find this resource:

Weyand, T. G., Boudreaux, M., & Guido, W. (2001). Burst and tonic response modes in thalamic neurons during sleep and wakefulness. Journal of Neurophysiology, 85, 1107–1118.Find this resource:

Wilson, J. R. (1989). Synaptic organization of individual neurons in the macaque lateral geniculate nucleus. Journal of Neuroscience, 9, 2931–2953.Find this resource:

Winfield, D. A. (1980). The synaptic organization of glomeruli in the magnocellular and parvocellular laminae of the lateral geniculate nucleus in the monkey. Brain Research, 198, 55–62.Find this resource:

Womelsdorf, T., Anton-Erxleben, K., Pieper, F., & Treue, S. (2006). Dynamic shifts of visual receptive fields in cortical area MT by spatial attention. Nature Neuroscience, 9, 1156–1160.Find this resource:

Worgotter, F., Eyding, D., Macklis, J. D., & Funke, K. (2002). The influence of the corticothalamic projection on responses in the thalamus and cortex. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 357, 1823–1834.Find this resource:

Xing, D., Ringach, D. L., Shapley, R., & Hawken, M. J. (2004). Correlation of local and global orientation and spatial frequency tuning in macaque V1. Journal of Physiology, 557, 923–933.Find this resource:

Yabuta, N. H., Sawatari, A., & Callaway, E. M. (2001). Two functional channels from primary visual cortex to dorsal visual cortical areas. Science, 292, 297–300.Find this resource:

Zeater, N., Cheong, S. K., Solomon, S. G., Dreher, B., & Martin, P. R. (2015). Binocular visual responses in the primate lateral geniculate nucleus. Current Biology, 25(24), 3190–3195.Find this resource:

Zeki, S. (1983). The distribution of wavelength and orientation selective cells in different areas of monkey visual cortex. Proceedings of the Royal Society B, 217, 449–470.Find this resource:

Zhang, Y., Kim, I. J., Sanes, J. R., & Meister, M. (2012). The most numerous ganglion cell type of the mouse retina is a selective feature detector. Proceedings of the National Academy of Science USA, 109, 2391–2398.Find this resource:

Zhou, H., Friedman, H. S., & von der Heydt, R. (2000). Coding of border ownership in monkey visual cortex. Journal of Neuroscience, 20, 6594–6611.Find this resource:

Zhou, H., Schafer, R. J., & Desimone, R. (2016). Pulvinar-cortex interactions in vision and attention. Neuron, 89, 209–220.Find this resource:

Ziskind A. J., Emondi, A. A., Kurganski, A. V., Rebrik, S. P., & Miller, K. D. (2014). Neurons in cat V1 show significant clustering by degree of tuning. Journal of Neurophysiology, 113, 2555–2581.Find this resource: