When most people, including most clinicians, think about the visual cortex, they picture the back of the brain, the occipital lobe. This is the area that lights up in brain scans when someone looks at a picture. It's a tidy mental model, and like most tidy mental models of the nervous system, it only tells part of the story; the part of the story that gets left out is often what makes everything make sense when viewed as a cohesive whole.
The truth is that vision is not a single pathway from eye to occipital lobe, regardless of how many beautifully printed diagrams you will see online and in textbooks. It is a sprawling, bidirectional, highly distributed network that reaches deep into the parietal and temporal lobes, loops back on itself through dozens of feedback connections, and sends a significant portion of its fibers to destinations that never touch the cortex at all.
Understanding this complexity doesn't just satisfy intellectual curiosity. It reframes how clinicians need to think about visual dysfunction, rehabilitation, and the remarkable resilience of sight. And it helps explain why some of the most clinically significant visual problems a patient can have will never show up on a standard eye exam.
The Traffic That Never Reaches V1
Let's start where most neuroscience education stops: the optic nerve. Light hits the retina, signals travel down the optic nerve, cross at the chiasm, and arrive at the primary visual cortex (V1) in the occipital lobe. This is a clear and linear picture, but it is incomplete.
A meaningful fraction of retinal ganglion cell axons peel off from this main highway and never reach V1. Their destinations reveal just how many separate jobs vision is actually doing simultaneously.
Some fibers project to the superior colliculus, a midbrain structure that coordinates orienting movements of the eyes and head. These are not conscious vision fibers, they are reactive, reflexive, and operate faster than conscious awareness. Others branch into the pretectal nuclei that govern the pupillary light reflex. When someone shines a light in your eye the reflex they're testing depends entirely on fibers that bypass the cortex altogether.
A smaller but critical set of axons project directly to the suprachiasmatic nucleus of the hypothalamus, the brain's master circadian clock. This retinohypothalamic tract is how your body knows what time of day it is. It is why light exposure at night disrupts your sleep even if, consciously, you barely registered the light. Your circadian system is running a parallel visual process entirely outside your awareness and entirely outside your occipital lobe.
Still other fibers feed the accessory optic system, a collection of nuclei that stabilize your visual world as your head and eyes move; this is the reason the room doesn't appear to lurch every time you turn your head. None of these targets are feeding into V1. None of them are generating the images you consciously see. And yet dysfunction in any of them produces distinct, measurable impairment.
The Myth of the One-Way Visual Pathway
Even within the cortical visual system, the popular image of a one-way feedforward pathway from the retina → LGN → V1 → higher areas misrepresents how the system actually works.
The lateral geniculate nucleus (LGN), the thalamic relay that sits between the eye and V1, is a perfect example. Most people learn it as a simple relay station between the retina and the occipital lobe. But the LGN receives far more input from the cortex via feedback projections than it does from the retina itself. The dominant flow of information into the LGN is top-down, not bottom-up. The LGN is not a passive relay; instead it is an active gate, shaped by attention, arousal, and expectation, constantly filtering and modulating what gets sent to cortex.
This pattern of bidirectional connectivity repeats at every level. V1 sends signals forward to V2, V3, V4, and MT/V5. But all of those areas also send projections back to V1. Higher association areas in the temporal and parietal lobes project back to early visual cortex, influencing how low-level features are processed in light of higher-level context. Your brain is not just reading the visual world, it is constantly predicting it and using incoming signals to update those predictions.
The Visual Cortex Extends Far Beyond the Occipital Lobe
Once signals leave V1 and the early extrastriate cortex, they diverge into two broad streams that reach deep into the brain, and this is where the neuroscience connects powerfully to something behavioral optometry figured out decades ago. In fact, behavioral optometry's understanding of the neuroscience of vision predates what was made famous by scientists like Colwyn Trevarthen in the 1960s.
The ventral stream sweeps forward through the temporal lobe, supporting object recognition, face identification, color processing, and word reading. The dorsal stream projects upward into the parietal lobe, supporting motion perception, spatial processing, and visually guided action, skills like reaching, grasping, navigating. Both of these streams are involved in the concept of the "visual cortex", and neither are the occipital lobe.
Damage Versus Dysfunction: A Crucial Distinction
Here is why the clinical picture is essential to understand, as this is what often gets missed in classroom or whiteboard instruction videos.
When we speak of damage to these streams, whether a stroke, a tumor, a traumatic lesion, the deficits are dramatic and unmistakable. Damage to the dorsal stream can produce optic ataxia, where a patient cannot accurately reach for an object they can clearly see. Damage to the ventral stream produces prosopagnosia, the inability to recognize faces, or pure alexia, the inability to read despite intact language. These are rare, striking, and neurologically obvious.
But dysfunction, on the other hand, is a subtler disruption of these same systems without frank structural lesion, and it presents very differently than physical damage. It is quieter, more variable, and far more common. And visual dysfunction maps with striking precision onto a framework developed by the behavioral optometrist A.M. Skeffington decades before modern neuroimaging confirmed the two-stream architecture.
Dorsal Stream Dysfunction and Skeffington's "Centering"
Skeffington's concept of "centering" addresses the question "where is it?". This is our awareness of where objects reside in space relative to ourselves, supported through the eye movements and binocular vision. This is where the idea of depth perception comes into play so as to understand how far away things are from us. In neuroanatomical terms, this is precisely the province of the dorsal stream: spatial localization, motion tracking, binocular vergence, and the coordinate transformations needed to guide the body toward objects in the environment.
When the dorsal stream is damaged, the result is frank optic ataxia or simultanagnosia. But when it is dysfunctional, as in cases of developmental dorsal stream vulnerability, acquired brain injury, or visual stress, the presentation is subtler and is often misunderstood because the symptoms have nothing to do with visual acuity.
These patients have difficulty judging distances and spatial relationships. They struggle in visually crowded environments, places like busy grocery stores, cluttered classrooms, open-plan offices, because the system tasked with organizing where things are relative to the body is operating inefficiently. They may have convergence insufficiency, difficulty sustaining binocular alignment in near space, or poor smooth pursuit, not because their eyes are broken but because the cortical machinery that drives and monitors those movements is not functioning cleanly. They may bump into things, misjudge steps, feel disoriented in unfamiliar spaces, or fatigue rapidly when driving. Their visual fields may be technically full, their acuity may be 20/20, and yet they cannot reliably navigate and interact with the spatial world around them; this is exactly what Skeffington's centering circle describes.
Ventral Stream Dysfunction and Skeffington's "Identification"
Skeffington's "identification" circle addresses the question "what is it?". This is the ability to adjust accommodation, achieve 20/20 acuity, and derive meaning from what is being observed. This maps directly onto the ventral stream's role in resolving fine form, color, surface detail, and ultimately the identity of objects, faces, and words.
When the ventral stream is damaged, the results are dramatic, conditions like prosopagnosia, achromatopsia, visual agnosia. But ventral stream dysfunction presents as something many professionals will immediately recognize: difficulty with sustained focus, slow or effortful processing, poor rapid naming, labored reading despite adequate decoding skills, and difficulty with visual perceptual tasks that require quick discrimination between similar forms.
These patients can often eventually identify what they're looking at, but it takes longer, costs more effort, and breaks down under time pressure or fatigue. They may struggle to rapidly distinguish visually similar letters (b/d, p/q), not because of a phonological deficit but because the system that rapidly and automatically resolves fine visual form is not running efficiently. They may find reading exhausting because each word requires more cortical work to identify than it should. They may struggle with face recognition in social settings, not dramatically enough to call it prosopagnosia, but enough to cause real social difficulty.
Accommodation helps a person identify the image they are focused on and allows them to visually answer "what is it?" and when identification and centering are concurrently active, visual perceptual skills develop through intersensory integration. When either system is running below capacity, the integrated act of vision, the fact that it should be effortless, automatic, and fast, begins to fail in ways that look like attention problems, learning difficulties, or simply "not trying hard enough."
The Framework That Unifies It
What makes the Skeffingtonian model remarkable in retrospect is that it described, in functional clinical terms, a two-stream architecture that neuroscience would spend the next half-century working out in anatomical detail. Skeffington's Four Circle Model of Vision anticipated interdisciplinary research in visual cognitive neuroscience, with centering mapping onto the dorsal "where/how" stream and identification mapping onto the ventral "what" stream.
The distinction between damage and dysfunction matters enormously for clinical practice. A patient with frank dorsal or ventral stream damage will present to a neurologist. A patient with dorsal or ventral stream dysfunction, the far larger population, will walk into an optometrist's, pediatrician's, or neuro-rehabilitation clinic's office, describing symptoms that sound vague, inconsistent, and hard to pin down on standard testing.
They are not imagining it. Their visual cortex, all of it, parietal, temporal, occipital and even frontal (more on that in another document) together, is simply not working as a fully integrated system. And recognizing which stream is underperforming, using Skeffington's deceptively simple questions as a guide, is often the first step toward understanding why.