For more than a century, classical sensory physiology operated under the dogmatic presumption of a monolithic visual apparatus. Vision was conceptually flattened into a single, hierarchical pipeline: light struck the retina, ascended through the lateral geniculate nucleus of the thalamus, and terminated in the primary visual cortex (striate cortex or Area 17), where the external world was faithfully reconstructed into a conscious internal percept. Within this Cartesian-inspired, corticocentric paradigm, all visual capabilities—from detecting a looming predatory threat in the peripheral visual field to discerning the intricate contours of an object held within the hands—were assumed to rely on the progressive elaborations of this single geniculostriate pathway. Non-striate visual structures, particularly the ancient midbrain roof known as the optic tectum or superior colliculus, were relegated to the status of vestigial reflex arcs, subordinate waystations whose sole purpose was to point the eyes so that the cortex could perform the authentic work of seeing.
This classical consensus fractured irreparably in the late 1960s. Driven by pioneering empirical breakthroughs across comparative ethology, neuroanatomy, and split-brain neuropsychology, three researchers working largely independently—David Ingle, Gerald Schneider, and Colwyn Trevarthen—demonstrated that vision is not a unified functional entity, but rather a parallel architecture composed of at least two phylogenetically distinct, anatomically segregated, and computationally specialized visual systems. Trevarthen introduced the foundational conceptual dichotomy between focal vision (high-resolution, central, object-oriented, and conscious) and ambient vision (panoramic, peripheral, space-oriented, and unconscious). Simultaneously, Schneider established an empirical double dissociation in golden hamsters, demonstrating that cortical lesions selectively abolish visual pattern discrimination (“what”) while midbrain lesions abolish spatial orientation (“where”). In amphibians, Ingle revealed that visually guided action is fractured into task-specific subcortical networks, with prey capture and barrier avoidance executing across distinct anatomical channels.
The formulation of the Two-Visual-Systems Hypothesis represents one of the most consequential paradigm shifts in cognitive neuroscience and evolutionary neurobiology. It liberated sensory physiology from corticocentrism, elevated the subcortical midbrain from a primitive motor node to an intelligent spatial engine, and provided the direct intellectual lineage for modern dual-stream cortical models, most notably the dorsal (“where” or “how”) and ventral (“what”) processing streams proposed by Mortimer Mishkin, Leslie Ungerleider, Melvyn Goodale, and David Milner. This comprehensive treatise explores the historical roots, experimental foundations, neuroanatomical substrates, evolutionary dynamics, and contemporary clinical and technological ramifications of the focal-ambient architecture pioneered by Ingle, Schneider, and Trevarthen.
1. Historical Foundations and the Genesis of Dual-Stream Visual Processing
1.1 The Classical Monolithic View of Visual Perception
The intellectual landscape of visual neuroscience throughout the nineteenth and early twentieth centuries was dominated by the pursuit of cerebral localization. Following the clinicopathological observations of Salomon Eberhard Henschen and the meticulous battlefield mapping of visual field deficits in wounded soldiers conducted by Gordon Holmes, the primary visual cortex (striate cortex or V1) was crowned as the absolute seat of visual perception. Within this classical framework, vision was conceptualized as a monolithic, unitary capacity. Light entering the eye was transduced by photoreceptors, channeled along the optic nerve, systematically sorted within the dorsal lateral geniculate nucleus (dLGN), and projected via the optic radiations to the calcarine fissure of the occipital lobe. The striate cortex was assumed to function as a photographic plate or an internal projection screen; damage to this structure inevitably produced an absolute, unrecoverable void in the visual field—a homonymous hemianopia or absolute scotoma.
This classical paradigm rested upon the unexamined philosophical premise that visual perception and conscious visual experience were entirely synonymous. If a patient or an experimental animal sustained a complete bilateral ablation of the striate cortex, they were diagnosed as totally blind, and this blindness was assumed to be universal across all sensorimotor and cognitive dimensions. The visual process was envisioned as a linear, unidirectional hierarchy: sensory input was passively registered at the primary cortical level, synthesized into higher-order geometric forms within the adjacent circumstriate or peristriate association areas (Brodmann Areas 18 and 19), and subsequently routed to the frontal lobes for the execution of voluntary behavioral responses. Under this monolithic architecture, spatial localization (determining where an object is located in three-dimensional space) and object recognition (determining the identity, color, and form of that object) were believed to be indivisible operations executed sequentially by the same cortical machinery.
The profound limitation of this historical model lay in its methodological and conceptual reliance on blunt human clinical lesions and crude animal behavioral assays. Traditional neuropsychological testing paradigms were heavily biased toward verbal report, conscious recognition, and high-acuity form discrimination. Patients with striate cortex lesions were asked whether they could consciously identify letters, recognize faces, or perceive illuminated targets. When they invariably answered in the negative, the clinical verdict of absolute blindness was rendered, effectively blinding the scientific community to the existence of preserved, non-conscious visual capabilities operating through alternative, non-striate neural pathways.
1.2 The Emergence of Subcortical Functional Independence
Cracks in this corticocentric monolithic dogma began to appear when comparative physiologists and experimental neurologists ventured beyond traditional primate models and refined their surgical ablation methodologies. Early experimentalists, including Flourens and later nineteenth-century investigators such as David Ferrier and Hermann Munk, had occasionally noted that decorticated mammals—animals whose entire neocortex had been surgically removed or functionally deactivated—did not behave as completely blind organisms. Despite the catastrophic loss of cortical tissue, decorticated cats, dogs, and rodents retained an uncanny ability to navigate around large obstacles, adjust their posture in response to visual motion, and orient their heads and eyes toward sudden, bright flashes of light in their environment.
These persistent, non-cortical visual abilities directed scientific attention downward toward the mesencephalon, specifically to the optic tectum (known in mammals as the superior colliculus). The optic tectum represented the primary visual center of the vertebrate brain throughout hundreds of millions of years of evolutionary history, reaching an extraordinary degree of structural complexity and laminar differentiation in teleost fishes, amphibians, and reptiles. In these anamniote and sauropsid lineages, which completely lack a multi-layered neocortex, the optic tectum successfully orchestrated the totality of visual life: detecting prey, tracking mates, evading predators, and guiding whole-body locomotion through complex three-dimensional terrain.
The realization that these ancestral subcortical visual structures were preserved throughout mammalian evolution—persisting beneath the massive mantle of the expanding telencephalon—prompted a radical re-evaluation of vertebrate neuroanatomy. Rather than serving as passive reflex centers subordinate to cortical command, the superior colliculus and its associated pretectal and tegmental nuclei were increasingly suspected of maintaining an autonomous functional capacity. Comparative neuroethologists began to hypothesize that vertebrate encephalization had not replaced midbrain visual mechanisms, but had instead layered an advanced, high-resolution analytical system on top of a resilient, evolutionarily ancient sensorimotor foundation designed for spatial orientation and physical survival.
1.3 The Convergence of Ingle, Schneider, and Trevarthen in the Late 1960s
Between 1967 and 1969, sensory physiology and cognitive neuroscience experienced an unprecedented intellectual convergence. Working independently across completely different vertebrate models, institutional settings, and experimental paradigms, David Ingle, Gerald Schneider, and Colwyn Trevarthen simultaneously published findings that shattered the classical monolithic view of vision and crystallized the Two-Visual-Systems Hypothesis. This period marked a profound epistemological transition within neurobiology, migrating from passive, descriptive lesion studies toward rigorous, ethologically anchored behavioral dissociations designed to isolate distinct computational domains within the visual brain.
At the Massachusetts Institute of Technology (MIT), Gerald Schneider was conducting precise, double-dissociation ablation studies on the golden hamster (Mesocricetus auratus). Schneider demonstrated that an animal could be rendered completely incapable of distinguishing a pattern (such as horizontal versus vertical stripes) while retaining flawless visual orientation toward a sunflower seed, or conversely, could retain pattern discrimination while becoming entirely unable to direct its head toward the location of that same seed, depending on whether the visual cortex or the superior colliculus was lesioned. Schneider formulated this distinction in a landmark 1969 paper in Science as the functional dissociation between identifying “what” a stimulus is and determining “where” it is located in space.
Concurrently, at the California Institute of Technology (Caltech), Colwyn Trevarthen was investigating split-brain monkeys and human commissurotomy patients in the laboratory of Roger Sperry. Trevarthen discovered that while the surgical division of the forebrain commissures (the corpus callosum and anterior commissure) completely severed the interhemispheric transfer of fine pattern vision and conscious perceptual judgments, it left the subjects’ spatial orientation, postural stability, and ambient movement perception remarkably unified across the midline. In a seminal 1968 paper, Trevarthen coined the enduring terminology of the focal visual system and the ambient visual system, formalizing the functional split between high-acuity, foveal object inspection and panoramic, peripheral spatial navigation.
Simultaneously, at McLean Hospital and Harvard University, David Ingle was conducting brilliant neuroethological experiments on anuran amphibians (frogs and toads). Ingle revealed that anurans process visual information through modular, task-specific action systems rather than a centralized spatial map. By selectively lesioning the optic tectum or the pretectal area, Ingle proved that visually guided prey-snapping and visually guided barrier avoidance were mediated by entirely separate anatomical circuits within the midbrain. When synthesized, the revolutionary work of Ingle, Schneider, and Trevarthen demonstrated conclusively that the vertebrate visual brain is fundamentally dualistic: a rapid, evolutionarily ancient, subcortical/ambient system continuously anchors the organism within spatial reality, while an advanced, neocortical/focal system meticulously analyzes the identity and intrinsic properties of isolated visual entities.
2. Colwyn Trevarthen and the Duality of Focal and Ambient Vision
2.1 Formulation of the Ambient Visual System
Colwyn Trevarthen’s primary contribution to theoretical neuroscience was the formalization of the ambient visual system as a specialized, neurofunctional entity distinct from conscious object vision. Trevarthen posited that the ambient visual system is characterized by an extensive, large-field, peripheral retinal architecture designed to survey the totality of the external environment. Unlike the narrow, high-resolution visual beam associated with foveal inspection, ambient vision operates across wide visual angles, capturing global scene dynamics, geometric boundaries, and spatial gradients. It represents a continuous sensory dialogue between the organism and its physical surroundings, functioning as an internal perceptual scaffolding upon which spatial awareness is erected.
The primary functional objective of the ambient visual system is the maintenance of postural equilibrium, the coordination of axial and appendicular musculature during locomotion, and whole-body navigation through complex three-dimensional terrain. Trevarthen recognized that ambient visual processing is intrinsically tied to the registration of what James J. Gibson termed the optical flow field—the continuous patterns of motion and geometric transformation across the retina generated by an organism’s own self-motion through space. By monitoring global velocity vectors, motion parallax, and spatial gradients across the visual periphery, the ambient system effortlessly informs the central nervous system of its instantaneous heading, rate of acceleration, and orientation relative to gravitational vertical.
Crucially, Trevarthen emphasized that ambient visual operations take place entirely outside the realm of conscious awareness and deliberative cognition. Ambient vision functions as an automatic, non-symbolic sensorimotor control system. It does not pause to dissect the color, texture, or semantic categorization of an obstacle; instead, it immediately translates the geometric distribution of surfaces and impending boundaries into instantaneous motor adjustments. It is an unconscious spatial surveillance mechanism that runs perpetually in the background of primate experience, stabilizing the visual platform and protecting the organism from environmental hazards while conscious attention is directed elsewhere.
2.2 Formulation of the Focal Visual System
In direct structural and functional contrast to the ambient system, Trevarthen delineated the focal visual system. The focal system is centered almost exclusively upon the macula and fovea centralis, regions characterized by an extraordinary packing density of cone photoreceptors and a dedicated, minimal-convergence neural wiring to downstream ganglion cells. Focal vision is computationally optimized for high spatial resolution, fine detail discrimination, chromatic contrast, and meticulous feature extraction. It represents the conscious, analytical aperture of the primate brain, selectively directed toward isolated fragments of the broader visual world for detailed interrogation.
The operational mechanics of focal vision are fundamentally active and exploratory. Unlike the ambient system, which passively absorbs the panoramic optical flow field across the periphery, focal vision depends upon the coordinated deployment of high-velocity saccadic eye movements, smooth pursuit tracking, and prolonged, deliberate foveal fixations. The visual system actively points its high-resolution foveal sensor at specific targets of interest, sweeping across scenes to sample informative edges, intersections, and surfaces. This process is inherently bottlenecked by selective spatial and feature-based attention; focal vision cannot process the entirety of a scene at once, requiring sequential, serial sampling to assemble a coherent cognitive model of an object.
Trevarthen identified focal vision as the neural engine of object identification, semantic categorization, and conscious perceptual judgment. It is through the focal visual channel that a human or non-human primate recognizes the identity of a conspecific, deciphers facial expressions, inspects fine tools, assesses the ripeness of fruit based on chromatic nuances, or reads text. Focal vision isolates an object from its spatial context, converting retinal luminance profiles into invariant, object-centered cognitive representations. It is the form of vision that directly populates phenomenal consciousness—the subjective experience of seeing, recognizing, and knowing what is present within the center of gaze.
2.3 Empirical Evidence from Split-Brain Primate Preparations
Trevarthen’s theoretical dichotomy was not merely an abstract conceptual deduction; it was directly forged through experimental investigations of commissurotomy in non-human primates and human split-brain patients conducted in Roger Sperry’s laboratory at Caltech during the 1960s. Surgical section of the corpus callosum and anterior commissure effectively isolates the two cerebral neocortical hemispheres, eliminating the massive transverse axonal tracts responsible for interhemispheric communication. Trevarthen designed sophisticated behavioral apparatuses that presented visual stimuli selectively to one or both visual hemifields while simultaneously monitoring manual responses, eye movements, and spatial coordination.
Trevarthen observed that when split-brain baboons (Papio) and macaque monkeys were tested on complex focal visual discrimination tasks—such as learning to differentiate between two geometric patterns or abstract symbols presented to opposing hemifields—the two divided cerebral hemispheres acted as two entirely independent learning systems. The left hemisphere could learn that a cross predicted a reward while a circle predicted an aversive stimulus, while the right hemisphere learned the precise inverse, with no cross-talk, transfer of training, or interhemispheric interference. Focal vision was thoroughly lateralized and bisected by the callosal transection, confirming that high-resolution pattern discrimination and cognitive decision-making were strictly confined to the isolated cerebral cortices.
However, when Trevarthen examined ambient visual functions within these same split-brain preparations, the outcome was profoundly different. When the monkeys were required to execute coordinated bilateral manual reaches toward spatial locations in the peripheral visual field, maintain postural balance on tilting perches, or track large-field optical movements across the visual midline, the subjects demonstrated seamless, unified performance. The two halves of the body remained perfectly coordinated within a single, continuous, coherent three-dimensional space. Trevarthen deduced that because the forebrain commissures had been completely severed, this unified spatial competence was being sustained by intact subcortical commissures and bilateral midbrain architecture—specifically within the tectum, pretectum, and brainstem reticular formation. The midbrain ambient system formed a continuous, undivided spatial foundation beneath the bifurcated focal processing of the cerebral hemispheres.
3. Gerald Schneider’s Cortical vs. Tectal Dissociation: What Versus Where
3.1 The Classic Golden Hamster (Mesocricetus auratus) Ablation Paradigms
While Trevarthen was teasing apart visual processing through commissurotomy in California, Gerald Schneider was pioneering an equally revolutionary series of experimental ablations in Cambridge, Massachusetts. Working at MIT, Schneider selected the golden hamster (Mesocricetus auratus) as his experimental model. Hamsters possessed several distinct advantages for neuroethological investigation: they possessed well-developed visual systems with high nocturnal and diurnal behavioral responsiveness, demonstrated robust spatial orienting instincts, and possessed a highly accessible, lissencephalic visual cortex alongside a large, anatomically prominent superior colliculus that could be surgically manipulated with exquisite precision.
Schneider developed two rigorously segregated behavioral testing paradigms designed to isolate distinct cognitive and sensorimotor requirements. The first was a two-choice pattern discrimination task, often executed within a visual discrimination chamber or Y-maze. The hamster was trained to approach two adjacent visual panels or doors displaying contrasting geometric patterns—such as horizontal versus vertical alternating black-and-white stripes, or patterns differing in spatial frequency. Pushing through the correct door yielded a food reward (a sunflower seed), while pushing the incorrect door resulted in a locked barrier. This task required the animal to hold its body relatively stationary, visually resolve the intrinsic structural properties of the two stimuli, and execute a choice based strictly upon the identity of the visual pattern—an operationalization of “what.”
The second paradigm was a spatial localization and orienting assay. Hamsters were placed in an open arena, and a sunflower seed was introduced into various locations within their peripheral visual field via a fine mechanical arm, without producing any auditory or olfactory localization cues. The dependent measure in this paradigm was the animal’s latency, accuracy, and kinematics in executing an immediate, ballistic head turn and body orientation directly toward the spatial locus of the seed. This task required no pattern analysis; the seed was a simple point-like visual target, and the animal merely had to compute the exact azimuth and elevation of the visual event relative to its head and execute a precise motor orienting vector—an operationalization of “where.”
3.2 The Double Dissociation: Identifying ‘What’ and Locating ‘Where’
Schneider’s experimental findings provided the first classic, unequivocal double dissociation in the neurobiology of vision. When Schneider performed bilateral surgical ablations of the visual cortex (specifically targeting the striate cortex and adjacent extrastriate visual areas) while leaving the superior colliculus completely intact, the hamsters exhibited a profound, catastrophic deficit on the door pattern discrimination task. Even after hundreds of training trials, the cortically ablated animals were entirely incapable of learning to distinguish horizontal from vertical stripes. Their pattern vision was effectively destroyed, confirming the indispensable role of the visual cortex in high-resolution spatial frequency analysis and geometric form perception.
However, when these same cortically blind hamsters were tested in the spatial orienting paradigm, their behavior was indistinguishable from normal, unoperated control animals. The moment a sunflower seed appeared in their peripheral visual field, they executed an instantaneous, highly accurate head turn, snapped their jaws onto the seed, and consumed it. They could effortlessly localize targets throughout their entire visual field. Their spatial localization mechanism was fully operational, completely unaffected by the total absence of the visual cortex.
The inverse lesion produced the mirrored outcome. When Schneider performed precise bilateral surgical ablations of the superior colliculus while preserving the primary visual cortex and the geniculostriate pathway intact, the behavioral profile was inverted. These tectally ablated hamsters demonstrated complete spatial orientation blindness. When sunflower seeds were dangled directly in front of them, slightly to the left, or into their peripheral fields, the animals exhibited no orienting response whatsoever. They failed to turn their heads, walked past the food targets as if they did not exist, and demonstrated a profound inability to direct their gaze or spatial actions toward localized visual events.
Yet, when these tectally lesioned animals were placed into the Y-maze pattern discrimination apparatus, their performance was remarkable. They could readily learn to discriminate between horizontal and vertical stripe patterns, pushing through the correct door to retrieve their reward with high accuracy. They could successfully identify “what” the visual pattern was, despite being fundamentally unable to turn their heads toward a localized target in space to determine “where” it was. This pristine double dissociation established beyond dispute that “identifying” and “locating” were mediated by anatomically distinct, parallel neural systems: the geniculostriate pathway for pattern identification and the retinotectal pathway for spatial localization.
3.3 Theoretical Implications for Vertebrate Encephalization
Schneider’s 1969 publication sent shockwaves through evolutionary neurobiology and physiological psychology. At the time, prevailing evolutionary models assumed that as the vertebrate brain evolved from fish through amphibians, reptiles, and mammals, higher cognitive and sensory functions underwent complete “encephalization.” Under this hierarchical view, functions once performed by primitive midbrain structures were supposedly transferred to and subsumed by the newly evolved cerebral cortex, leaving ancestral structures as vestigial or degenerate. Schneider’s empirical double dissociation proved that this hierarchical encephalization model was fundamentally flawed.
The evolutionary expansion of the telencephalon had not suppressed or replaced ancestral midbrain processing; rather, evolution had conserved the optic tectum as an independent, highly specialized, and exceptionally fast visuomotor engine while developing the geniculostriate system in parallel to perform complex, computationally intensive pattern and feature analyses. Schneider demonstrated that mammalian visual architecture was an integrated composite of phylogenetic strata, operating not through an absolute master-servant cortical hierarchy, but as a collaborative network of parallel visual circuits.
Furthermore, Schneider’s conceptual division between “what” and “where” radically restructured how neuroscientists conceptualized the relationship between perception and action. By showing that spatial localization was intrinsically linked to the subcortical motor systems mediating head- and eye-turning, Schneider laid the neuroethological groundwork for understanding that sensory systems did not evolve merely to generate abstract internal representations of the world, but to provide immediate, actionable spatial commands for behavioral execution. The tectal pathway was recognized as an action-oriented visual system, while the cortical pathway functioned as an identification system—a conceptual framework that would profoundly influence visual neuroscience for the next half-century.
4. David Ingle’s Neuroethological Investigations in Lower Vertebrates
4.1 Behavioral Dissociations in Amphibians (Anura)
While Schneider was demonstrating the tectal-cortical dissociation in rodents, David Ingle was conducting foundational neuroethological investigations at McLean Hospital into the visual organization of lower vertebrates, specifically anuran amphibians such as the northern leopard frog (Rana pipiens) and the marine toad (Bufo marinus). Amphibians were historically dismissed as rigid, reflex-bound automata governed by primitive, homogeneous sensory mechanisms. Pioneering neurophysiological work by Jerome Lettvin and colleagues at MIT in their famous 1959 paper, “What the Frog’s Eye Tells the Frog’s Brain,” had demonstrated that the anuran retina possessed dedicated ganglion cell classes acting as specialized feature detectors—such as “bug detectors” responding to small, moving convex boundaries.
Ingle, however, recognized that visual behavior in amphibians was far more nuanced and computationally sophisticated than a single set of retinal feature detectors wired to a universal motor program. He designed naturalistic behavioral assays to observe how toads and frogs interacted with multiple, conflicting visual stimuli in rich environments. Ingle observed that when a prey object (such as a mealworm or a small moving fly) was presented within a frog’s visual field, the animal executed a highly stereotypical, ballistic hunting sequence: an immediate orienting turn of the head and body toward the prey, followed by an accurate forward lunge and tongue snap. This prey-catching behavior was extraordinarily sensitive to small, moving targets exhibiting high velocity contrast against the background.
Conversely, when Ingle placed large, stationary vertical obstacles (such as white palisades, barriers, or sheets of glass) between the frog and the prey target, the animal demonstrated an entirely different class of visuomotor behavior. Instead of snapping at the barrier or repeatedly colliding with it, the frog accurately steered its body around the edges of the barrier, using visually guided sidestepping and obstacle-avoidance maneuvers before orienting toward the prey behind it. If a large, dark object loomed rapidly toward the frog from above or behind, the animal executed an explosive, omnidirectional escape jump away from the threat trajectory. Ingle realized that the frog was not operating through a single, general-purpose visual space, but was deploying distinct, mutually exclusive motor programs driven by categorically different visual cues.
4.2 Tectal and Pretectal Neural Architecture
To identify the neural substrates governing these segregated behavioral repertoires, Ingle conducted meticulous microlesion and neural tract-tracing studies within the anuran diencephalon and mesencephalon. His investigations revealed an extraordinary anatomical and functional segregation between two adjacent subcortical structures: the optic tectum and the pretectal nuclear complex (comprising the caudal thalamic and pretectal neuropil).
When Ingle made selective bilateral surgical lesions of the optic tectum while sparing the pretectum, he observed an absolute, permanent abolition of all prey-catching behavior. The tectally ablated frogs and toads showed no behavioral registration of flies, mealworms, or small moving targets anywhere within their visual fields. They never oriented toward, pursued, or snapped at prey. However, to Ingle’s astonishment, these tectally “blind” amphibians retained virtually intact obstacle-avoidance and spatial navigation capabilities. When placed in an open arena obstructed by multiple vertical barriers, the animals deftly circumnavigated the obstacles without bumping into them, displaying accurate stereoscopic depth perception and distance estimation. Furthermore, when presented with looming shadows, their escape jump mechanisms remained intact. The optic tectum was specifically dedicated to orienting toward small, mobile prey targets, while stationary spatial geometries were processed elsewhere.
When Ingle produced selective lesions within the pretectal and caudal thalamic regions while leaving the optic tectum structurally intact, he observed the precise mirror-image dissociation—and an astonishing behavioral disinhibition. Pretectally lesioned toads lost all ability to perceive barriers, obstacles, and stationary edges. When placed in front of a white palisade or glass wall, they repeatedly walked or jumped headfirst into the obstruction, completely blind to the presence of physical barriers. Simultaneously, these pretectally lesioned animals exhibited an uninhibited, hyperactive prey-catching reflex. They would snap compulsively at small moving targets even when positioned directly behind impassable barriers, and would even snap at their own limbs or visual stimuli that normal toads ignored. Ingle deduced that the pretectum not only mediated stationary barrier avoidance and spatial navigation, but also exerted tonic, inhibitory control over the orienting machinery of the optic tectum.
4.3 The Concept of Multiple Functional Action-Vision Subsystems
David Ingle’s neuroethological discoveries carried profound theoretical ramifications that transcended amphibian biology. Ingle recognized that the brain did not construct a single, unified “picture” of the visual world from which various motor commands were subsequently extracted. Instead, vision had evolved as an assemblage of multiple, semi-autonomous, task-specific visuomotor modules, each with its own sensory tuning parameters, neural pathway, and dedicated motor output circuitry.
Ingle articulated a multi-system visual processing architecture wherein the sensory properties of a stimulus—its size, velocity, trajectory, and stationary geometry—were routed directly to the specific brain structure evolved to handle that ecological problem. The optic tectum was the specialized engine for appetitive orienting (prey tracking and targeting); the pretectal nucleus was the specialized engine for defensive navigation and obstacle negotiation (barrier avoidance and edge following); and the accessory optic system was the specialized engine for stabilizing the visual world on the retina during whole-body movement (optokinetic reflexes).
By demonstrating that even anamniote vertebrates possessed multiple, anatomically segregated visual action systems operating in parallel, Ingle provided the evolutionary blueprint for understanding mammalian visual duality. The focal and ambient systems identified by Trevarthen and the “what” and “where” pathways mapped by Schneider were not evolutionary novelties invented de novo by primates or mammals; they were the direct evolutionary elaborations of ancient, modular visuomotor circuits that had originated in the earliest vertebrates to coordinate survival within an unpredictable physical world.
5. Neuroanatomical Substrates of the Focal-Ambient Dichotomy
5.1 The Retinotectal and Tectofugal Projections (Ambient Pathway)
The neuroanatomical architecture mediating the ambient visual system is rooted in the phylogenetically ancestral retinofugal projections that bypass the primary thalamic relay of the dorsal lateral geniculate nucleus. In mammals, this ambient pathway originates largely from a specialized population of retinal ganglion cells characterized by large somas, widely branching dendritic fields, and thick, heavily myelinated axons supporting exceptionally fast conduction velocities—known historically as Y-cells in carnivores and rodents, and parasol (alpha/M-like) ganglion cells in primates.
These specialized ganglion cell axons exit the optic chiasm and travel via the optic tract to terminate directly within the superficial laminae of the superior colliculus: the stratum zonale, stratum griseum superficiale, and stratum opticum. The superficial collicular layers maintain an orderly retinotopic map of the contralateral visual hemifield, heavily weighting the broad peripheral visual field. From these superficial layers, ambient visual signals are routed along two distinct architectural trajectories. One trajectory descends into the intermediate and deep collicular laminae (stratum griseum intermediale and stratum griseum profundum), where visual inputs are dynamically integrated with auditory and somatosensory maps of space. The deep colliculus acts as a multisensory transformation hub, projecting directly via the tectospinal tract (descending into the cervical spinal cord) and the tectobulbar/tectoreticular tracts to coordinate immediate, ballistic orienting movements of the eyes, pinnae, head, and torso toward sudden peripheral events.
The second ambient trajectory is ascending: efferent axons from the superior colliculus project rostrally to the pulvinar nucleus of the thalamus (specifically the inferior and lateral pulvinar) and the lateral posterior nucleus (LP in non-primates). This tecto-pulvinar-extrastriate pathway projects directly into associative cortical regions, including the middle temporal visual area (MT/V5) and the posterior parietal cortex, completely bypassing the primary visual cortex (V1). This ascending ambient conduit provides the parietal cortex with rapid, coarse spatial and motion coordinates, sustaining spatial localization, optical flow analysis, and visuomotor guidance even in the total structural absence of the geniculostriate focal system.
5.2 The Retinogeniculostriate Architecture (Focal Pathway)
In contrast to the panoramic, subcortically dominated ambient pathway, the focal visual system is sustained by the sophisticated, highly specialized retinogeniculostriate architecture, which reached its evolutionary zenith in haplorhine primates. The focal pathway originates predominantly within the macula and fovea centralis of the retina, driven by midget ganglion cells (X-cells or P-cells in primates). These cells possess compact, restricted dendritic arborizations and minute receptive fields, often receiving input from a single cone photoreceptor via a single midget bipolar cell within the central fovea—establishing a pristine, un-converged 1:1 functional transmission line optimized for maximum spatial acuity.
The axons of these midget ganglion cells project via the optic nerve and tract to terminate within the four parvocellular (P-cell) layers (layers 3, 4, 5, and 6) of the dorsal lateral geniculate nucleus (dLGN) of the thalamus. The dLGN exhibits a rigid, six-layered laminar architecture wherein inputs from the contralateral and ipsilateral eyes are strictly segregated into alternating eye-specific laminae. The parvocellular layers are structurally tuned for high spatial frequency transmission, sustained tonic firing rates, chromatic opponency (red-green spectral contrast mediated by L- and M-cone comparisons), and fine spatial contrast.
From the dLGN, the focal visual channel ascends via the optic radiations (geniculocalcarine tract), sweeping around the lateral ventricles (partially via Meyer’s loop) to terminate in the primary visual cortex (striate cortex, Area 17, or V1), concentrated densely within layer 4Cbeta. Striate cortex is organized into exquisite vertical functional columns: orientation columns, ocular dominance columns, and cytochrome oxidase blobs. Within V1, the focal pathway extracts fundamental local features of the visual scene: precise spatial orientation of edges, fine spatial phase, boundary discontinuities, and chromatic signatures. From V1, this focal information is relayed sequentially into extrastriate visual areas V2, V4, and the inferior temporal cortex (IT), establishing the feedforward hierarchical cascade dedicated to invariant object recognition, facial identification, and semantic scene comprehension.
5.3 Subcortical-Thalamic-Cortical Loops and Integration Hubs
The focal and ambient pathways do not operate as hermetically sealed parallel pipes; rather, they are continuously woven together through complex subcortical-thalamic-cortical loops and integration hubs that coordinate sensory registration with motor behavior. The primary integration engine of this distributed network is the pulvinar nucleus of the thalamus. As the largest thalamic nucleus in the primate brain, the pulvinar maintains extensive, reciprocal connections with both subcortical midbrain engines (the superior colliculus) and wide swaths of the cerebral cortex, spanning V1, V2, MT, the posterior parietal cortex, and the frontal eye fields (FEF).
These reciprocal loops allow the pulvinar to function as an active routing switch and synchronizing bridge. While the superior colliculus can project rapidly through the tecto-pulvinar pathway to deliver raw spatial and motion signals to the parietal cortex, cortical areas simultaneously send massive, descending corticocollicular feedback projections back down to the colliculus. Layer 5 pyramidal neurons in V1, as well as specialized projection neurons in the frontal eye fields and posterior parietal cortex, project directly down the internal capsule to terminate within the intermediate and deep layers of the superior colliculus. This massive descending feedback grants the advanced focal cortical system top-down inhibitory and modulatory control over primitive ambient orienting reflexes, preventing the eyes from being compulsively captured by every irrelevant flicker of motion in the visual periphery.
Simultaneously, the ambient system is integrated with the accessory optic system (AOS)—a specialized collection of terminal midbrain nuclei comprising the medial, lateral, and dorsal terminal nuclei (MTN, LTN, DTN)—and the nucleus of the optic tract (NOT). The AOS receives direct retinal projections specialized for detecting slow, large-scale, whole-field retinal slip. The NOT and AOS project directly into the vestibular nuclei, the inferior olive, and the flocculonodular lobe of the cerebellum. This subcortical loop drives the optokinetic reflex (OKR) and coordinates the vestibulo-ocular reflex (VOR), dynamically stabilizing the visual image upon the retina during head rotations and whole-body perturbations, maintaining a stable visual platform upon which focal inspection can operate.
6. Physiological and Functional Characteristics of the Ambient System
6.1 Spatial and Temporal Tuning Properties
The electrophysiological properties of the ambient visual system are exquisitely adapted for rapid surveillance, spatial orientation, and change detection. At the neural level, neurons within the ambient pathway—spanning the magnocellular layers of the dLGN, the superficial and intermediate laminae of the superior colliculus, and motion-selective extrastriate cortex (area MT)—are tuned to exceptionally low spatial frequencies. They possess large receptive fields that can encompass several degrees to tens of degrees of visual angle in the periphery. Rather than parsing the sharp edges, fine textures, or high-acuity details of an object, ambient neurons extract the global geometric envelope of the environment—the coarse distributions of light, shadow, horizons, and broad physical surfaces.
In the temporal domain, the ambient system operates at the opposite extreme: it is tuned to exceptionally high temporal frequencies. Ambient neurons display transient, rapidly adapting firing profiles. When a visual stimulus appears or moves within their receptive fields, these cells fire a high-frequency, short-latency burst of action potentials that decays almost immediately. This transient response makes the ambient system hyper-sensitive to instantaneous changes in luminance, sudden movement vectors, flicker, and environmental transients. The conduction velocities along ambient axonal pathways—such as the heavily myelinated fibers of the retinotectal and tectospinal tracts—are among the fastest in the visual system, with signal transmission latencies as brief as 30 to 50 milliseconds from retinal stimulation to subcortical motor activation.
These spatial and temporal tuning profiles allow the ambient system to operate effectively across the entirety of the visual field, with special emphasis on the far periphery, where rod photoreceptors dominate and cone density falls toward zero. While the human fovea represents less than one percent of the total retinal surface, the ambient system processes the expansive remaining ninety-nine percent, maintaining continuous, spherical surveillance of the three-dimensional surround.
6.2 Optical Flow Analysis and Visuomotor Equilibrium
A foundational operational domain of the ambient visual system is the real-time computational extraction of the optical flow field. As an animal moves through its environment, its physical displacement generates a continuous, structured vector field of apparent motion across the retina. Elements in the direction of locomotion appear to expand radially outward from a central focus of expansion (heading direction), while elements behind contract toward a focus of contraction, and stationary lateral surfaces slide past with velocity gradients determined by their relative distance (motion parallax).
The ambient system possesses specialized neural populations—particularly within the pretectal nucleus of the optic tract, the accessory optic system, the superior colliculus, and the medial superior temporal area (MST)—that decode these complex, large-field velocity distributions. By computing the global divergence, curl, and deformation of the optical flow field, the ambient system provides instantaneous feedback regarding self-motion (ego-motion), heading trajectory, and spatial orientation. This visual flow computation is integrated directly with mechanical inputs arriving from the vestibular labyrinth (semicircular canals and otolith organs) and proprioceptive stretch receptors embedded within cervical spinal musculature.
This multisensory convergence within ambient integration hubs maintains visuomotor equilibrium and postural stability. Because ambient vision relies on low spatial frequencies and large-field optical motion, it exhibits remarkable physiological resilience against sensory degradation. While focal vision is immediately crippled by optical blur, atmospheric fog, refractive error, or low scotopic (nighttime) illumination, ambient spatial tracking and postural stabilization continue to operate effectively. A human or animal can walk through dense fog or a darkened landscape without falling, effortlessly maintaining balance and heading, long after the focal visual system has lost all ability to read, identify objects, or recognize faces.
6.3 Non-Conscious Motor Guidance and Collision Avoidance
The ambient visual system operates predominantly beneath the threshold of conscious perceptual experience, executing complex sensorimotor transformations in an automatic, non-volitional mode. A quintessential example of this non-conscious computation is the detection and evasion of looming visual targets—objects on a direct collision course with the organism. The ambient pathway, specifically through dedicated collicular and pretectal circuits, continuously calculates the optical expansion rate of approaching boundaries, computing the mathematical parameter known in ecological psychology as tau—the ratio of an object’s instantaneous retinal image size to its rate of optical expansion.
This computation yields an accurate, invariant prediction of the time-to-contact (TTC) directly from raw retinal kinematics, without requiring the brain to know the object’s actual physical size, true linear velocity, or absolute metric distance. When the tau threshold reaches a critical survival boundary, the deep laminae of the superior colliculus trigger an immediate, ballistic evasive reaction: ducking, flinching, shielding the eyes with the arms, or executing an explosive jump. These defensive motor outputs occur with extraordinarily short latencies—often within 60 to 80 milliseconds—substantially faster than the 150 to 250 milliseconds required for the focal-geniculostriate-frontal cortex axis to consciously recognize the identity of the threatening object and deliberate on a voluntary motor response.
Similarly, during ordinary human locomotion over uneven terrain, the ambient system continuously guides foot placement, stride length, and body posture without requiring deliberate conscious attention. A hiker navigating a rocky trail can engage in deep, conscious philosophical conversation—relying entirely on focal attention directed toward thoughts, social interaction, or distant scenery—while their ambient visual system automatically scans ground contours in the lower peripheral visual field, adjusting foot trajectories and compensatory ankle stiffness to maintain forward momentum without tripping.
7. Physiological and Functional Characteristics of the Focal System
7.1 High-Acuity Representation and Chromatic Specialization
The physiological architecture of the focal visual system is engineered to solve an entirely different computational problem: extracting the invariant structural, spatial, and material properties of isolated objects at maximum spatial resolution. To achieve this, the focal system relies on an exquisite, high-density sensory substrate centered within the fovea centralis. Within this central pit, which spans a visual angle of only 1 to 2 degrees, photoreceptor packing reaches its theoretical physical limit, dominated exclusively by tightly packed, elongated L- and M-cone photoreceptors, completely devoid of rod intrusion and free from overlying retinal capillary beds and inner nuclear layers.
This foveal sensory surface is coupled with an exceptionally low neural convergence ratio. In the foveal midget pathway, each individual cone photoreceptor synapses onto a single on- or off-center midget bipolar cell, which in turn contacts a single midget ganglion cell projecting to the parvocellular layers of the dLGN. This 1:1:1 private line wiring preserves pristine, unblurred spatial contrast, enabling the visual system to resolve fine spatial frequencies exceeding 30 cycles per degree of visual angle. Within the striate cortex, this high-acuity focal representation is afforded massive neural real estate through cortical magnification: the central fovea, representing less than 0.1% of the total retinal surface area, is allocated nearly 25% of the primary visual cortex (Area 17).
In tandem with high spatial acuity, the focal system is defined by chromatic specialization. By contrasting inputs from long-wavelength (L), middle-wavelength (M), and short-wavelength (S) cones through spectrally opponent receptive fields (such as Red-versus-Green and Blue-versus-Yellow channels), parvocellular focal networks compute pure chromatic contrast independent of luminance. This chromatic processing, further refined within the specialized cytochrome oxidase blobs of V1, the thin stripes of V2, and the modular globs of visual area V4, serves as an indispensable tool for figure-ground segregation—allowing primates to detect camouflaged fruit against dense foliage, discern subtle variations in skin tone during social interactions, and extract high-order surface textures from the surrounding visual environment.
7.2 Attentional Modulation and Cognitive Interfacing
A definitive physiological hallmark of the focal visual system is its intense susceptibility to top-down attentional modulation. Because the high-resolution processing machinery of the parvocellular-striate-extrastriate pathway is computationally expensive and capacity-limited, it cannot process multiple complex objects simultaneously without severe structural interference. As a result, focal vision is dynamically governed by the frontoparietal attention network, comprising the frontal eye fields (FEF), the superior parietal lobule, and the ventral prefrontal cortex.
Attentional modulation acts as a dynamic neural gain-control mechanism. When an individual directs covert spatial attention or feature-based attention toward a specific target, firing rates of neurons within visual areas V4, IT, and MT whose receptive fields encompass that target are markedly elevated, while spontaneous and evoked activity of neurons responding to unattended distractor stimuli is actively suppressed. This attentional gating selectively filters the sensory flow, allowing only behaviorally relevant information to breach the threshold into working memory and executive planning circuits within the prefrontal cortex.
Consequently, the focal visual channel exhibits significantly prolonged processing latencies relative to the subcortical ambient system. While an ambient orienting response can execute through the colliculus in less than 50 milliseconds, full focal object recognition—sweeping through the extensive synaptic hierarchy from retina to LGN, to V1, V2, V4, and terminating in the anterior inferior temporal cortex (AIT)—requires 150 to 200 milliseconds or longer. This temporal latency represents the biological cost of depth: the time required to perform successive non-linear filtering, cross-talk with semantic memory stores, and formulate a conscious perceptual interpretation of the visual scene.
7.3 Perceptual Constancy and Invariant Object Identification
The ultimate computational triumph of the focal visual system is the achievement of perceptual constancy—the ability to recognize the invariant identity of an object despite radical, continuous fluctuations in the sensory image landing upon the retina. In the real world, as an observer moves or as lighting conditions fluctuate, the retinal image of an object undergoes dramatic changes: its size expands or contracts with distance, its geometric shape distorts due to perspective transformations, and its spectral composition shifts drastically between noon daylight and incandescent twilight.
The focal pathway solves this ill-posed inverse problem through hierarchical, non-linear coordinate transformations unfolding along the ventral processing stream. Within visual area V4 and the lateral occipital complex (LOC), neurons compute color constancy by discounting the illuminant and extract shape boundaries using relative curvature and spatial phase relationships. As information arrives within the anterior inferior temporal (AIT) cortex, receptive fields become extraordinarily large (often encompassing 20 to 40 degrees) and invariably include the central fovea. AIT neurons demonstrate remarkable invariance: a single neuron will fire robustly to the presentation of a specific face, complex tool, or natural object regardless of its retinal size, position in the visual field, planar orientation, or illumination angle.
Through this progressive abstraction, the focal system transforms fragile, retinotopic sensory signals into robust, object-centered (allocentric) representations linked directly to meaning, language, and memory. Crucially, it is this focal stream that directly sustains the qualitative, phenomenal fabric of conscious visual awareness (qualia). The rich, vivid, subjective experience of seeing a recognizable world—of knowing that a red, curved, glossy shape is an apple ready for consumption—is the unique product of the focal visual system.
8. Comparative Neurobiology: Evolutionary Conservation of Dual Vision
8.1 Visual Specialization Across Anamniotes and Sauropsids
The functional division of vision into focal and ambient channels is not an evolutionary novelty invented by mammalian brains; it is an ancient organizational principle deeply conserved across hundreds of millions of years of vertebrate phylogenesis. In anamniote vertebrates (agnathans, chondrichthyes, teleosts, and amphibians), the optic tectum reigned as the supreme visual integration center. In teleost fishes, which navigate complex three-dimensional aquatic environments without the benefit of a laminated cerebral cortex, the optic tectum exhibits an exquisite, highly differentiated multi-laminar architecture comprising up to fifteen distinct histological strata, serving as both a spatial map and an analytical sensorimotor computer.
In sauropsids (reptiles and birds), this ancient midbrain supremacy was preserved and amplified, while visual processing also expanded into specialized forebrain structures. Birds, which possess some of the most sophisticated visual systems in the animal kingdom, exhibit a dual visual architecture that closely parallels the mammalian focal-ambient organization. In the avian brain, the massive optic tectum and its ascending projection to the nucleus rotundus and the entopallium (historically termed the ectostriatum) constitute the tectofugal pathway. This pathway is specialized for rapid motion detection, visual Looming computations, stereoscopic orientation, and ambient flight control.
In parallel, birds possess a separate ascending pathway termed the thalamofugal pathway: retinal fibers project to the dorsal thalamus, which in turn projects to the visual Wulst within the telencephalon. The avian visual Wulst operates as a functional analog to the mammalian primary visual cortex, exhibiting binocular disparity tuning, fine spatial frequency selectivity, and high-acuity pattern processing. Whether analyzing the precision hunting of raptors or the obstacle-avoidance maneuvers of hummingbirds, the avian nervous system demonstrates that the dual requirements of vision—panoramic spatial guidance during high-speed locomotion versus high-resolution target interrogation—inevitably drive the evolution of segregated, parallel neural circuits.
8.2 Mammalian Radiations and the Reconfiguration of Tectal Primacy
With the adaptive radiation of mammals following the Cretaceous-Paleogene transition, the structural relationship between the ancestral midbrain tectum and the forebrain underwent profound reconfiguration. Early mammalian ancestors were small, nocturnal, burrowing creatures that experienced what paleontologists term the “nocturnal bottleneck.” In these early mammals, high-acuity diurnal vision was de-emphasized in favor of olfaction, audition, and tactile whiskers (vibrissae). In nocturnal rodents, such as mice and hamsters, the superior colliculus retained a major share of visual processing, remaining the primary engine driving visually guided orienting and defensive freezing or fleeing behaviors.
However, with the emergence of primates and their transition to an arboreal, diurnal ecological niche, visual evolution exploded in a new direction. Navigating the complex three-dimensional canopy of tropical forests, calculating delicate branch jumps, and foraging for small fruits and insects required unprecedented enhancements in depth perception, chromatic discrimination, and high-resolution spatial vision. This ecological pressure drove the massive expansion of the primate neocortex, specifically the primary visual cortex (V1) and a constellation of extrastriate visual areas, coupled with the evolution of a specialized, cone-rich central fovea and forward-facing binocular eyes.
Yet, critically, this massive mammalian neocortical expansion did not eradicate or replace the ancient collicular ambient pathway. Instead, the superior colliculus was retained and re-wired into a sophisticated sensorimotor satellite. While the primate V1 expanded to manage the computational burdens of stereopsis and high-acuity focal analysis, the colliculus was retained as the rapid-response targeting engine—the ultimate controller of the ocular motor plant, responsible for generating the high-speed saccadic eye movements necessary to point the high-resolution foveal cortex at relevant features in the environment. The mammalian brain achieved its perceptual dominance not by discarding its ancestral midbrain, but by layering an analytical focal cortex on top of an agile ambient base.
8.3 Adaptive Trade-Offs Between Rapid Survival Actions and Detailed Scene Parsing
The evolutionary persistence of two distinct visual systems represents an exquisite biological resolution to a fundamental physical and computational trade-off: speed versus resolution. A visual system optimized exclusively for high spatial resolution requires enormous computational infrastructure. Processing high spatial frequencies requires millions of dedicated photoreceptors, minimal neural convergence, massive parallel axonal channels, and extensive multi-layered cortical networks to compute edge orientations, textures, and three-dimensional surface geometries. This complex processing pipeline inevitably introduces substantial synaptic transmission delays, resulting in behavioral reaction latencies on the order of several hundred milliseconds.
In life-or-death survival scenarios—such as an ambush predator lunging from the periphery or a sudden rockfall during mountain navigation—a reaction delay of 200 milliseconds is often fatal. Natural selection cannot tolerate an organism that insists on meticulously identifying the species, sex, and coloration of a predator before initiating an evasive maneuver. Survival demands an instantaneous, “good-enough” motor reflex driven by coarse, fast, low-latency visual data. The ambient visual system provides precisely this evolutionary insurance policy. It sacrifices fine acuity, color, and conscious identification in exchange for panoramic coverage, rapid axonal conduction, and direct, low-latency access to motor output circuits in the brainstem and spinal cord.
This evolutionary balance is clearly reflected in the ecological specializations of different animal species. Prey species (such as ungulates, lagomorphs, and rodents) typically possess laterally placed eyes with broad, panoramic visual fields, expansive horizontal visual streaks, and highly developed ambient tectal pathways optimized for the instant detection of looming threats from any azimuth. Conversely, predatory species (such as felids, raptors, and primates) feature frontal eyes with overlapping binocular fields, centralized foveas, and massive cortical focal expansions designed to meticulously parse camouflage, estimate metric depth, and guide predatory limbs during high-precision strikes. The focal-ambient duality thus represents a universally conserved evolutionary dialectic between immediate physical survival and analytical environmental mastery.
9. Cross-Talk, Synergy, and Integration Between Focal and Ambient Channels
9.1 Dynamic Interplay During Natural Visual Search
In ecological reality, the focal and ambient visual systems do not function as isolated, competing silos; rather, they engage in a continuous, dynamic, and seamless collaborative cycle during natural visual exploration. This synergy is best observed during natural visual search. As an organism navigates a novel landscape, the ambient visual system operates as a wide-angle surveillance radar. It continuously monitors the global optical flow field to maintain postural balance and heading, while its transient peripheral detectors constantly scan for anomalous visual events—a sudden movement vector, an unexpected luminance transient, or a coarse geometric asymmetry in the visual periphery.
The moment an anomalous or behaviorally relevant event is detected in the peripheral field, the ambient system—specifically the intermediate and deep layers of the superior colliculus—computes a precise motor error vector: the difference between the target’s current retinal locus and the center of the fovea. This triggers a rapid, ballistic conjugate saccadic eye movement, accompanied by compensatory head rotations, swinging the eyes toward the peripheral coordinate in a fraction of a second. The ambient system thus acts as the spatial targeting mechanism that serves the focal system.
Upon completion of the saccade, the target of interest lands directly within the central fovea centralis, bringing the high-acuity machinery of the focal system online. The parvocellular-striate-temporal network immediately engages in fine-grained feature extraction, chromatic analysis, and invariant object identification, determining whether the visual anomaly is a threat, a resource, or an irrelevant distractor. During the execution of the saccade itself, a specialized neural mechanism known as saccadic suppression actively dampens the gain of ambient visual motion channels (specifically within the magnocellular pathway and area MT), preventing the brain from experiencing a disorienting, smeared visual blur as the world sweeps across the retina during the eye shift. Once the eye comes to rest, suppression is instantly lifted, and the focal-ambient collaboration resumes.
9.2 Spatial Frame Transformations and Reference Systems
A fundamental computational distinction between the focal and ambient visual systems lies in the coordinate reference frames they utilize to organize spatial information. The ambient visual system is fundamentally anchored in egocentric reference frames—coordinates that define spatial locations relative to the physical body of the acting organism. Depending on the specific motor plant being engaged, ambient circuits dynamically compute target locations in retinotopic (eye-centered), craniotopic (head-centered), or scapulotopic (torso-centered) coordinates.
When a looming predator approaches from the left, the ambient superior colliculus must compute the trajectory strictly relative to the animal’s current head and body position in order to generate a biomechanically appropriate evasive jump away from the threat vector. Similarly, during optical flow analysis, heading direction is computed as an egocentric vector defining the self’s trajectory relative to the physical surfaces of the immediate environment. The ambient system answers the direct, practical question: “Where is that surface or event located relative to my acting body?”
The focal visual system, by contrast, must transcend purely egocentric coordinates to achieve robust object recognition, migrating toward allocentric reference frames—coordinates that define spatial relationships relative to the object itself or to the broader environmental background, independent of the observer’s instantaneous perspective. When inspecting a tool, a written character, or a conspecific’s face, the focal system must represent the spatial relationships between the object’s constituent parts (e.g., the nose relative to the eyes, or the blade relative to the handle) in an object-centered frame. If the focal system relied solely on egocentric coordinates, an object would be registered as fundamentally completely different every time the head tilted or the eyes moved. The posterior parietal cortex and parahippocampal areas act as crucial coordinate transformation engines, converting raw egocentric ambient coordinates into allocentric focal representations and vice versa, enabling seamless spatial transitions between action and perception.
9.3 Perturbations and Conflicting Visual Inputs
The operational boundaries and independent neurobiological reality of the focal and ambient systems are dramatically exposed when experimental perturbations create sensory conflicts between them. A classic experimental demonstration is the phenomenon of vection—the visually induced illusion of self-motion. When an observer is seated inside a large, hollow, stationary drum whose inner walls are painted with vertical stripes, and the drum begins to rotate while the observer remains motionless, the large-field visual motion stimulates the ambient visual system. Because the ambient system interprets panoramic optical motion as proof of ego-motion, the observer rapidly experiences the compelling, visceral sensation that their own physical body is rotating in the opposite direction, despite their vestibular otoliths and semicircular canals correctly signaling that they are stationary.
Crucially, if the observer is instructed to focus their high-acuity focal attention on a small, stationary cross held within their hands inside the rotating drum, a profound focal-ambient decoupling occurs. The focal system clearly perceives and recognizes the cross as stationary relative to the eyes; yet the ambient system continues to drive the powerful, illusory vection of whole-body physical rotation, often accompanied by autonomic symptoms of motion sickness, postural instability, and compensatory muscle adjustments. The ambient system completely overrides conscious focal awareness and vestibular reality in its governance of postural equilibrium.
This functional segregation is further demonstrated through the use of peripheral optical prisms and displacement lenses. When experimental subjects are fitted with lenses that distort or laterally displace only the far peripheral visual field while leaving the central focal foveal field clear, they experience immediate, severe postural destabilization—staggering, swaying, and reaching erratically for supports. Conversely, if the central fovea is optically distorted while the peripheral ambient field is left clear and unobstructed, subjects maintain flawless, steady upright balance and smooth, coordinated locomotion, despite experiencing extreme difficulty reading text or identifying objects in their hands. Posture and spatial navigation belong entirely to the ambient domain.
10. Theoretical Evolution: From Ingle, Schneider, and Trevarthen to Modern Dual-Stream Models
10.1 Ungerleider and Mishkin’s Cortical Dichotomy (1982)
The foundational insights established by Ingle, Schneider, and Trevarthen in the late 1960s served as the direct intellectual catalyst for the next major milestone in visual neuroscience: the formulation of the cortical dual-stream model by Mortimer Mishkin and Leslie Ungerleider at the National Institute of Mental Health (NIMH) in 1982. In their landmark chapter, “Two Cortical Visual Systems,” Ungerleider and Mishkin took Schneider’s original midbrain-cortical dichotomy of “what” versus “where” and mapped it directly into the structural architecture of the primate cerebral neocortex.
Investigating macaque monkeys with selective cortical ablations and using radioisotope tract-tracing techniques, Ungerleider and Mishkin demonstrated that visual processing leaving the primary visual cortex (V1) bifurcates into two anatomically and functionally segregated neocortical processing streams. The first, which they termed the ventral stream (or the “what” pathway), projects from V1, through secondary visual areas V2 and V4, and terminates within the inferior temporal cortex (areas TEO and TE). Lesions along this ventral pathway produced catastrophic impairments in visual shape discrimination, pattern learning, and object recognition, while leaving spatial localization intact—representing the neocortical heir to Trevarthen’s focal system and Schneider’s cortical pattern mechanism.
The second pathway, which they designated the dorsal stream (or the “where” pathway), projects from V1, via areas V2 and V3, into the middle temporal area (MT/V5) and terminates within the posterior parietal cortex (specifically the inferior and superior parietal lobules). Monkeys with bilateral ablations of the posterior parietal cortex demonstrated profound impairments in a landmark spatial reversal task—such as choosing a food well based on its proximity to a visual cylinder (landmark task)—while retaining flawless visual pattern and object discrimination. Ungerleider and Mishkin successfully translated Schneider’s “where” system into a neocortical network, demonstrating that the cerebral mantle itself had adopted the ancient dualistic division of labor.
10.2 Goodale and Milner’s Perception vs. Action Model (1992)
A decade after Ungerleider and Mishkin’s formulation, cognitive neuroscientists Melvyn Goodale and A. David Milner revolutionized the field once again by fundamentally revising the functional interpretation of the dorsal stream. In a seminal 1992 publication in Trends in Neurosciences, and their subsequent monograph The Visual Brain in Action, Goodale and Milner argued that the critical distinction between the ventral and dorsal streams was not the attribute of the visual stimulus being processed (object qualities versus spatial locations: “what” vs. “where”), but rather the purpose for which the visual information was being deployed: conscious visual perception versus the real-time visual control of skilled motor action.
Goodale and Milner relabeled the ventral stream as the Perception Stream (mediating the conscious, allocentric identification of objects, faces, and scenes) and the dorsal stream as the Action Stream (mediating the automatic, egocentric visuomotor guidance of limb, hand, and eye movements: “how” vision). This theoretical reinterpretation resonated profoundly with Colwyn Trevarthen’s original conceptualization of ambient vision and David Ingle’s modular action systems. The dorsal/action stream was recognized as a specialized computational engine operating in real time to calculate metric grip apertures, reaching vectors, and postural adjustments—computations that do not enter conscious awareness, but are directly coupled to skeletal motor execution.
The empirical cornerstone of Goodale and Milner’s model was their meticulous investigation of patient D.F., a young woman who suffered profound, bilateral damage to the ventrolateral occipital cortex (sparing V1 and the dorsal parietal cortex) resulting from accidental carbon monoxide poisoning. Patient D.F. presented with severe visual form agnosia: she could not consciously identify, draw, or describe the shape, size, or orientation of common objects. When shown a card and asked to describe whether a rectangular slot was oriented vertically, horizontally, or at an angle, she performed at pure chance levels. However, when instructed to take the card in her hand and “post” it through the slot, she executed the movement flawlessly—her hand and wrist rotated fluidly during the reaching trajectory to match the precise orientation of the slot before insertion. Her dorsal action stream possessed the visual information required to guide motor behavior, despite her ventral perception stream being completely destroyed.
10.3 Re-Evaluating Subcortical Pathways in the Era of Modern Connectomics
While the cortical dual-stream models of Ungerleider-Mishkin and Goodale-Milner dominated late-twentieth-century visual neuroscience, the twenty-first century has witnessed a dramatic, connectomics-driven resurgence of interest in the original subcortical visual pathways discovered by Ingle and Schneider. For decades, the dominant cortical paradigms treated the primary visual cortex (V1) as an obligatory bottleneck through which all visual information must pass before reaching either the ventral or dorsal cortical streams. Recent neuroanatomical, optogenetic, and functional neuroimaging breakthroughs have radically dismantled this corticocentric assumption.
Modern viral tract-tracing and diffusion magnetic resonance tractography have definitively proven the existence of prominent, direct subcortical-extrastriate conduits in both human and non-human primates. Most notable among these is the colliculo-pulvinar-MT pathway: a robust axonal projection originating in the superficial and intermediate laminae of the superior colliculus, synapsing within the inferior pulvinar, and terminating directly within extrastriate motion area MT (V5) and the posterior parietal cortex—entirely bypassing V1. Optogenetic manipulations in rodents and non-human primates have shown that when V1 is functionally silenced, visual motion information still arrives in area MT via this tecto-pulvinar bridge with minimal latency degradation, sustaining spatial localization, optical flow extraction, and rapid orienting behaviors.
Contemporary neuroscience now conceptualizes the visual system not as a simple cortical binary, but as a multi-tiered, hierarchical dual architecture. The subcortical ambient/tectal pathway and the cortical dorsal/action stream form an integrated, evolutionary continuum dedicated to the spatial guidance of behavior, while the retinogeniculostriate pathway and the ventral cortical stream form a parallel continuum dedicated to detailed analysis, categorization, and conscious perceptual experience. The early neuroethological intuitions of David Ingle, Gerald Schneider, and Colwyn Trevarthen have thus been profoundly vindicated by modern connectomics.
11. Clinical and Neuropsychological Manifestations of Pathway-Specific Deficits
11.1 Cortical Blindness and the Phenomenon of Blindsight
The clinical vindication of the Two-Visual-Systems Hypothesis is nowhere more striking than in the phenomenon of blindsight, first rigorously documented and named by Lawrence Weiskrantz, Michael Sanders, and colleagues in the 1970s. When human patients sustain damage to the primary visual cortex (Area 17/V1)—whether through ischemic stroke, penetrating traumatic brain injury, or surgical resection—they develop a dense, homonymous scotoma or complete hemianopia. Within this affected visual field, patients report complete, profound phenomenological blindness: when asked what they see within the blind zone, they report experiencing absolute darkness, a complete absence of visual sensations, and an inability to consciously detect any visual stimuli.
However, when Weiskrantz and subsequent investigators utilized forced-choice experimental paradigms—instructing these cortically blind patients to guess the location of a target, point their finger toward a flash of light, or direct their eyes toward an event within their scotoma—patients demonstrated astonishing, statistically robust accuracy. Despite adamantly asserting that they were “merely guessing” and could “see nothing at all,” patients localized targets across their blind fields with high precision, distinguished between moving and stationary stimuli (the Riddoch phenomenon), and even mimicked facial expressions presented within their blind hemifields.
Blindsight represents the pure, uninhibited functional expression of the preserved subcortical ambient visual system operating in the total absence of the focal-geniculostriate apparatus. Deprived of the high-resolution, consciously accessible machinery of V1 and the ventral stream, the intact superior colliculus, pulvinar, and ascending tectofugal projections continue to extract spatial coordinates, detect motion vectors, and drive motor reaching mechanisms. The patient localizes the target because their ancient midbrain ambient engine successfully computes “where,” while they remain consciously blind because their focal cortical engine is no longer present to compute “what.”
11.2 Visual Agnosia and Optic Ataxia
The classical double dissociation between the focal and ambient/dorsal systems is vividly illustrated through two classic, contrasting neurological syndromes: visual agnosia and optic ataxia. Visual agnosia represents a profound disruption of the focal recognition pipeline. In apperceptive visual agnosia, typically resulting from bilateral damage to the lateral occipital complex, patients cannot synthesize visual features into coherent perceptual forms; they cannot copy simple drawings, recognize geometric shapes, or match identical objects. In associative visual agnosia, resulting from damage to more anterior inferior temporal structures, patients can perceive and meticulously copy complex drawings, but are completely unable to assign semantic meaning to them—looking at a drawing of an anchor, they can replicate every curve perfectly, yet have no idea what the object is or what function it serves.
Critically, patients with severe visual agnosia typically retain intact ambient and dorsal stream capabilities. As demonstrated by Goodale and Milner’s patient D.F., an agnosic patient who cannot verbally describe or consciously recognize a coffee cup can reach out and grasp it with flawless biomechanical precision, automatically adjusting their hand orientation, finger opening, and deceleration profile to match the physical dimensions of the object they cannot consciously identify.
The precise inverse of this clinical presentation is optic ataxia, a cardinal component of Bálint’s syndrome resulting from bilateral or unilateral lesions of the posterior parietal cortex. Patients suffering from optic ataxia exhibit no impairment in focal object recognition: they can look at an object, immediately identify it, describe its color, texture, and name, and detect the subtlest geometric flaws. However, when instructed to reach out and touch or grasp that object with their hand, they exhibit profound visuomotor decompensation. Their reaching trajectories are wildly inaccurate, missing the target by large margins (misreaching), and their hand fails to form the correct grip aperture to match the target’s physical size. Optic ataxia represents a specific, structural breakdown of the visual guidance of action—a failure of the ambient/dorsal spatial conduit—occurring against a background of completely preserved focal object recognition.
11.3 Vestibulopathy, Visual Vertigo, and Spatial Disorientation
In the domain of neuro-otology and vestibular rehabilitation, the focal-ambient distinction has emerged as an essential diagnostic and therapeutic framework for understanding chronic balance disorders, particularly Persistent Postural-Perceptual Dizziness (PPPD), visual vertigo, and spatial disorientation following traumatic brain injury (TBI). The ambient visual system normally works in seamless, complementary synergy with the peripheral vestibular apparatus to maintain verticality, balance, and spatial orientation.
When an individual suffers acute bilateral or unilateral vestibular loss (such as through vestibular neuritis or ototoxicity), the central nervous system rapidly upregulates its reliance on the ambient visual system to maintain postural stability. Under normal conditions, this visual compensation is highly adaptive. However, in a significant subset of clinical patients, the ambient system becomes pathologically hyperactive, leading to a condition termed visual dependence. When these visually dependent patients enter visually complex, dynamic environments—such as busy grocery stores with high-contrast fluorescent aisles, crowded pedestrian malls, or train platforms—the massive, disorganized ambient optical flow fields overwhelm the visual system, triggering severe visual vertigo, postural instability, nausea, and intense spatial disorientation.
Furthermore, following concussions, blast injuries, and moderate-to-severe traumatic brain injuries, patients frequently develop what neuro-optometrists diagnose as Post-Trauma Vision Syndrome (PTVS) or Ambient Visual Dysfunction. In PTVS, traumatic axonal shearing disrupts the delicate balance and temporal synchronization between the focal and ambient systems. The ambient peripheral visual system fails to properly stabilize the spatial manifold, forcing the high-acuity focal visual system to compensate by attempting to perform spatial orientation and balance control—tasks for which it is computationally unsuited. This creates intense asthenopia, cognitive fatigue, reading difficulty, spatial panic, and movement intolerance. Specialized neuro-optometric rehabilitation utilizing yoked prisms, peripheral visual stimulation, and ambient retraining protocols is specifically designed to re-establish proper subcortical-cortical ambient calibration, relieving the overtaxed focal system of its compensatory burden.
12. Contemporary Applications, Methodological Legacies, and Future Horizons
12.1 Ergonomics, Human Factors, and Display Engineering
The architectural principles of the Two-Visual-Systems Hypothesis have become indispensable design cornerstones in modern ergonomics, aviation safety, and industrial display engineering. In high-performance military and commercial aviation, pilots operate within extreme, highly dynamic environments where visual spatial disorientation represents a leading cause of catastrophic loss-of-control accidents. Historically, cockpit instruments were designed exclusively for focal visual interrogation: pilots had to scan small, centralized dials and screens to read altimeters, attitude indicators, and heading displays, mentally synthesizing these symbolic focal data points into an abstract cognitive model of aircraft orientation.
Under high stress, spatial acceleration, or degraded visual conditions (such as flying into thick clouds at night), the focal system easily becomes cognitively saturated or falls prey to powerful vestibular illusions, such as the “graveyard spiral” or the somatogravic illusion. To solve this mortal problem, human factors engineers designed display systems that deliberately leverage the ambient visual system. A revolutionary example is the Malcolm Horizon or peripheral attitude display: a laser projector that paints an artificial, full-field horizon bar across the entire cockpit instrument panel and peripheral visual field. Because this display stimulates the pilot’s peripheral, low-spatial-frequency ambient system, the pilot subconsciously and automatically perceives the aircraft’s true attitude relative to the Earth’s horizon without having to shift their conscious focal gaze away from target tracking or critical flight navigation instruments.
Similarly, in automotive engineering and Advanced Driver Assistance Systems (ADAS), understanding the focal-ambient divide is vital for human-machine interface (HMI) design. Modern vehicles increasingly saturate the driver’s focal visual attention with complex in-cabin infotainment touchscreens, GPS navigators, and digital cluster readouts. When a pedestrian suddenly steps into the roadway, a driver whose focal attention is captured by an interior screen experiences severe cognitive brake reaction delays. Consequently, safety engineers are shifting emergency collision warnings from focal screen icons to peripheral ambient warnings: dynamic ambient LED strips embedded along the base of the windshield that illuminate with rapid, looming expansion profiles. These peripheral stimuli instantly trigger the driver’s subcortical ambient orienting reflex, forcing an immediate, ballistic eye and head shift back to the forward roadway long before a focal dashboard warning could be cognitively read and interpreted.
12.2 Virtual Reality, Augmented Reality, and Simulator Sickness
The explosive development of immersive virtual reality (VR) and augmented reality (AR) technologies has brought the functional dynamics of the focal and ambient visual systems to the absolute forefront of computational design. A persistent barrier to widespread VR adoption is cybersickness (or simulator sickness), a form of motion sickness characterized by nausea, oculomotor strain, and profound disorientation. Extensive research has demonstrated that cybersickness is primarily an ambient-vestibular conflict disorder: the user’s ambient visual system registers large-field optical flow indicating high-speed forward movement through a virtual world, while their physical vestibular otoliths and semicircular canals signal that their physical body is sitting motionless in an office chair.
To mitigate this ambient mismatch, VR software architects implement dynamic visual vignetting: the moment the virtual avatar begins to rotate or accelerate, the software artificially constricts the field of view, blacking out the peripheral ambient visual field and leaving only the central focal aperture visible. By temporarily eliminating the large-field peripheral motion that stimulates the ambient vection engine, the visual-vestibular conflict is neutralized, drastically reducing symptoms of motion sickness. Once the virtual acceleration ceases, the peripheral vignette recedes, restoring the full panoramic view.
Simultaneously, the focal-ambient distinction has enabled a computational breakthrough in graphics rendering known as foveated rendering. Because the focal visual system provides high acuity only within the central 1 to 2 degrees of gaze, rendering an entire 110-degree VR display at ultra-high 4K or 8K resolution across every pixel is computationally wasteful. Using high-speed infrared eye-tracking cameras, modern VR headsets continuously track the user’s instantaneous point of foveal gaze. The graphics pipeline renders only that narrow focal region at maximum geometric complexity, ultra-high texture resolution, and full anti-aliasing. Concurrently, the expansive peripheral visual field is rendered at a fraction of the computational resolution, displaying coarse geometry and blurred textures. Because the ambient visual system is physiologically insensitive to high spatial frequencies and optical blur, the human user perceives the entire virtual world as pristinely sharp, saving up to 75% of graphics processing unit (GPU) computational overhead without degrading perceptual quality.
12.3 Artificial Intelligence, Computer Vision, and Neuromorphic Robotics
The architectural insights established by Ingle, Schneider, and Trevarthen are increasingly inspiring paradigm shifts within artificial intelligence, computer vision, and autonomous robotics. For decades, classical computer vision systems operated under an engineering philosophy analogous to the obsolete monolithic view of biological perception. Autonomous robots were equipped with frame-based, high-resolution cameras that captured the entire visual scene at a fixed temporal rate (e.g., 30 or 60 frames per second). Each massive pixel matrix was then passed through deep convolutional neural networks (CNNs) attempting to simultaneously perform object detection, spatial localization, and motion estimation within a single computational pipeline.
This monolithic approach suffers from crippling physical limitations: it is computationally cumbersome, generates massive data bottlenecks, consumes unsustainable electrical wattage, and suffers from severe temporal latency—often requiring tens to hundreds of milliseconds to process a single video frame. In dynamic, real-world robotic control—such as an autonomous drone navigating a dense forest at high speed or a robotic manipulator catching a falling tool—these latencies result in catastrophic physical collisions.
To overcome these limitations, neuromorphic roboticists are designing bio-inspired, dual-stream artificial architectures that directly mirror the biological focal-ambient division. The ambient channel is implemented using event-based neuromorphic vision sensors (such as Dynamic Vision Sensors or DVS). Unlike conventional cameras, neuromorphic event sensors do not capture synchronous frames of static pixels; instead, each independent pixel asynchronously reports changes in local logarithmic luminance with microsecond temporal resolution. This sparse, high-temporal, low-latency event stream mimics the transient, magnocellular ambient pathway, allowing autonomous robots to compute optical flow, avoid looming obstacles, and execute high-speed motor balance adjustments with sub-millisecond latencies while consuming milliwatts of power.
In parallel, the robot’s focal channel is implemented using high-resolution, deep neural networks that are selectively triggered only when the ambient event stream detects a salient spatial target. The high-resolution focal net is selectively pointed at the target coordinate, executing deep feature extraction, semantic segmentation, and invariant object identification. By combining a fast, sparse, low-resolution neuromorphic ambient engine for sensorimotor control with a slow, deep, high-resolution focal engine for object recognition, modern autonomous systems are finally matching the extraordinary energetic efficiency, speed, and behavioral agility of biological organisms.
Conclusion
The Two-Visual-Systems Hypothesis, forged through the pioneering investigations of David Ingle, Gerald Schneider, and Colwyn Trevarthen in the late 1960s, dismantled more than a century of corticocentric dogma and permanently transformed our understanding of sensory processing. By demonstrating that vision is fundamentally fractured into an ancient, subcortical, action-oriented ambient system and an advanced, neocortical, identification-oriented focal system, these three visionary researchers revealed the evolutionary architecture that underpins all visual behavior.
Their work established that the brain does not construct a single, static photographic representation of the external world. Instead, the visual brain is a parallel processing network of specialized sensorimotor and cognitive modules. The ambient system—anchored in the optic tectum, pretectum, and ascending tectofugal pathways—functions as an unceasing, unconscious spatial surveillance engine, continuously registering the optical flow field, maintaining postural balance, evading looming threats, and guiding locomotion through three-dimensional space. The focal system—anchored in the fovea, lateral geniculate nucleus, striate cortex, and the ventral cortical hierarchy—acts as an analytical aperture, deploying attention and saccades to extract fine spatial details, resolve chromatic nuances, and construct invariant object representations that populate conscious experience.
From the midbrain neuroethology of amphibians and the classic double dissociations of golden hamsters to the split-brain baboons of California, the legacy of Ingle, Schneider, and Trevarthen lives on. It formed the indispensable theoretical bedrock for modern dual-stream cortical neuroscience, provided the definitive clinical explanation for blindsight and visual agnosias, revolutionized aviation ergonomics and virtual reality engineering, and continues to guide the frontiers of neuromorphic robotics and artificial intelligence. More than half a century after its inception, the Two-Visual-Systems Hypothesis remains an enduring testament to the power of comparative, evolutionary neurobiology in unraveling the deepest mysteries of the seeing mind.
References
- Goodale, M. A., & Milner, A. D. (1992). Separate visual pathways for perception and action. Trends in Neurosciences, 15(1), 20–25. https://doi.org/10.1016/0166-2236(92)90344-8
- Gallego, G., Delbruck, T., Orchard, G., Bartolozzi, C., Taba, B., Censi, A., Leutenegger, S., Catalano, T., Scaramuzza, D., & Mueggler, E. (2020). Event-based vision: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(1), 154–180. https://doi.org/10.1109/JPROC.2020.2974843
- Holmes, G. (1918). Disturbances of visual orientation and loss of the continuous impression of movement in the visual fields. Brain, 41(3–4), 246–306. https://doi.org/10.1093/brain/39.1-2.34
- Ingle, D. (1967). Two visual mechanisms underlying the behaviour of fish. Psychologische Forschung, 31(1), 44–74. https://doi.org/10.1007/BF00422409
- Ingle, D. (1973). Two visual systems in the frog. Science, 181(4104), 1053–1055. https://doi.org/10.1126/science.181.4104.1053
- Lee, D. N. (1976). A theory of visual control of braking based on information about time-to-collision. Perception, 5(4), 437–459. https://doi.org/10.1016/0010-0285(76)90007-1
- Lettvin, J. Y., Maturana, H. R., McCulloch, W. S., & Pitts, W. H. (1959). What the frog’s eye tells the frog’s brain. Proceedings of the IRE, 47(11), 1940–1951. https://doi.org/10.1109/JRPROC.1959.287207
- Milner, A. D., & Goodale, M. A. (2006). The visual brain in action (2nd ed.). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198524724.001.0001
- Schneider, G. E. (1967). Contrasting visuomotor functions of tectum and cortex in the golden hamster. Psychologische Forschung, 31(1), 52–62. https://doi.org/10.1007/BF00422410
- Schneider, G. E. (1969). Two visual systems: Brain mechanisms for localization and discrimination are dissociated by tectal and cortical lesions. Science, 163(3870), 895–902. https://doi.org/10.1126/science.163.3870.895
- Staab, J. P., Eckhardt-Henn, A., Horii, A., Jacob, R., Strupp, M., Brandt, T., & Bronstein, A. (2017). Diagnostic criteria for persistent postural-perceptual dizziness (PPPD): Consensus document of the committee for the Classification of Vestibular Disorders of the Bárány Society. Journal of Vestibular Research, 27(4), 191–208. https://doi.org/10.3233/VES-170622
- Trevarthen, C. B. (1968). Two mechanisms of vision in primates. Psychologische Forschung, 31(4), 299–337. https://doi.org/10.1007/BF00422717
- Ungerleider, L. G., & Mishkin, M. (1982). Two cortical visual systems. In D. J. Ingle, M. A. Goodale, & R. J. W. Mansfield (Eds.), Analysis of visual behavior (pp. 549–586). MIT Press. https://doi.org/10.1016/B978-0-12-386660-8.50022-3
- Weiskrantz, L. (1986). Blindsight: A case study and implications. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780198521921.001.0001