Cortical Blindness

Abstract

Cortical blindness is the loss of vision caused by damage to the primary visual cortex of the occipital lobe, not the eyes. The eyes and optic nerves are intact and the pupils still react to light, yet the brain can no longer construct a conscious visual image. Its most extraordinary feature is blindsight: some patients with damage to primary visual cortex can, when forced to guess, respond above chance to stimuli they report not seeing — localising a light, detecting motion, or judging a fearful face without any conscious experience of it. This dissociation between visual performance and visual awareness has made cortical blindness a central case in the study of consciousness, because it separates seeing from knowing that one sees.

Keywords: cortical blindness, blindsight, primary visual cortex, visual awareness

Cortical blindness matters to cognitive psychology because it forces apart two things ordinary vision fuses seamlessly: the processing of visual information and the experience of seeing. When the primary visual cortex is destroyed, the retina still captures light and subcortical structures still route it into the brain, but the machinery that yields a conscious visual scene is gone. The result is a person who is, by their own honest report, blind — and who may nonetheless act on visual information they insist they cannot see (Weiskrantz, Warrington, Sanders, & Marshall, 1974).

The account below moves from what cortical blindness is, through its causes and the clinical picture, to blindsight and the residual vision that survives the loss of the primary visual cortex, then to the two anatomical pathways that explain how sight can outlast awareness, and finally to how the condition is assessed. The recurring theme is that what is spared and what is lost — the image gone, the guess preserved — is a map of how the intact brain turns light into conscious sight.

Figure 1

Two Routes From Eye to Cortex After Destruction of Primary Visual Cortex

What Cortical Blindness Is

Cortical blindness is the loss of vision produced by damage to the visual cortex of the brain, in a person whose eyes are structurally normal. The name locates the fault precisely: the blindness is cortical, arising after the point where the optic nerves and the eyes have done their work. Light still enters the eye, the retina still transduces it, and the pupils still constrict to a bright light — a reflex that runs through the midbrain and does not depend on the cortex — yet the person cannot see, because the region that builds a conscious visual image has been destroyed (Aldrich et al., 1987). This is the diagnostic signature that separates cortical from ocular blindness: intact pupillary responses and a normal eye examination in a patient who reports no vision.

The damage is to the primary visual cortex (also called V1, the striate cortex, or Brodmann area 17), which sits at the back of each occipital lobe and is the first cortical stage of vision. Because each hemisphere's visual cortex serves the opposite half of the visual field, destroying the visual cortex of one hemisphere produces blindness in one half-field — a homonymous hemianopia — whereas complete cortical blindness requires damage to both occipital lobes (Celesia, Bushnell, Toleikis, & Brigell, 1991). The condition is therefore a family of related deficits, from a restricted blind region to total loss, defined by the location and extent of the cortical injury rather than by any fault in the eye.

What makes cortical blindness more than a lesion label is that the visual system does not fall silent when its principal cortical target is removed. Subcortical structures continue to receive retinal input and to route it to other cortical areas, so visual information can still enter the brain and influence behaviour even when it never reaches consciousness. That surviving traffic is the basis of blindsight and of the residual capacities discussed below, and it is why cortical blindness is studied not only as a clinical problem but as a natural experiment on the boundary between vision and awareness (Weiskrantz, 1996).

Causes and Clinical Features

The commonest cause of cortical blindness is stroke in the territory of the posterior cerebral arteries, which supply the occipital lobes; bilateral posterior-circulation infarction can destroy both primary visual cortices and produce sudden, complete blindness in a person with entirely normal eyes (Aldrich et al., 1987). Other causes converge on the same region by different routes: head trauma, cardiac arrest or other episodes of cerebral hypoxia, eclampsia in pregnancy, hypoglycaemia, certain infections, and the reversible posterior encephalopathy that can follow severe hypertension. In children, cortical visual impairment from perinatal hypoxia or malformation is a leading cause of low vision. Prognosis varies widely with cause and extent, from near-complete recovery after a transient insult to permanent loss after extensive infarction (Aldrich et al., 1987).

Clinically, the hallmark is a mismatch between a normal eye examination and a patient who cannot see. Pupillary light reflexes are preserved, the retina and optic disc look healthy, and eye movements are typically intact, yet visual acuity in the affected field is absent (Celesia et al., 1991). In a minority of cases the mismatch extends to the patient's own report: Anton syndrome, or visual anosognosia, is the striking condition in which a cortically blind patient denies being blind, confabulating detailed descriptions of a scene they cannot in fact see — a failure not only of vision but of the awareness of its loss.

Electrophysiology sharpens the picture. Because the visual evoked potential recorded over the occipital scalp depends on the integrity of the primary visual cortex, it is characteristically abnormal or absent in cortical blindness, distinguishing a genuine cortical lesion from other causes of unexplained vision loss and confirming that the fault lies at or beyond V1 rather than in the eye or optic nerve (Celesia et al., 1991). The clinical task is thus to establish two things at once: that the eyes work, and that the cortex does not.

Blindsight: Vision Without Awareness

The most remarkable feature of cortical blindness is blindsight: the capacity of some patients to respond to visual stimuli in their blind field at rates well above chance, while sincerely reporting that they see nothing. The phenomenon was established in the modern era by Weiskrantz and colleagues, who studied a patient, known as D.B., with a surgical lesion of the primary visual cortex. Asked to point to or guess the location of a spot of light in his blind field, D.B. performed far better than chance, and could discriminate the orientation of lines and simple shapes — yet he insisted throughout that he was merely guessing and had no visual experience of the stimuli at all (Weiskrantz et al., 1974). Comparable residual function had been reported in the wounded soldiers Pöppel and colleagues studied, whose eye movements were drawn toward targets in their blind fields (Pöppel, Held, & Frost, 1973).

The term captures the paradox exactly: sight without seeing. Because the patient's honest report is of blindness, the preserved ability can be revealed only by forced-choice testing, in which the patient must respond even when they feel they are answering at random; the above-chance accuracy of those “random” answers is the evidence that visual information is being processed and used (Weiskrantz, 1996). This methodological point is essential: blindsight is not a weak or partial form of ordinary vision but a dissociation, in which performance and awareness come apart, and it can be measured only by a design that does not rely on the patient's conscious report.

Blindsight has since been demonstrated for several visual attributes — the location and movement of stimuli, simple form and orientation (Trevethan, Sahraie, & Weiskrantz, 2007), wavelength, and even the emotional expression of faces, which cortically blind patients can guess above chance and which drives activity in the amygdala without any conscious sight (Tamietto & de Gelder, 2010). A parallel phenomenon in monkeys, whose behaviour after striate-cortex removal closely mirrors human blindsight, showed that the animals treated a stimulus they could clearly detect as though it were nothing there to be seen — strong evidence that blindsight reflects genuinely unconscious processing rather than a residual sliver of normal vision (Cowey & Stoerig, 1995).

The Two Visual Pathways

How can vision survive the destruction of the primary visual cortex? The answer is that the retina does not send its signals to V1 alone. The dominant geniculostriate pathway carries the great majority of visual information from the retina through the lateral geniculate nucleus of the thalamus to V1, and it is this route, and the conscious image it builds, that cortical blindness destroys. But an older, parallel route runs from the retina to the superior colliculus in the midbrain and onward through the pulvinar to extrastriate visual areas, bypassing V1 entirely; and a direct projection from the lateral geniculate nucleus to extrastriate cortex, notably the motion area V5/MT, can survive a V1 lesion (Stoerig & Cowey, 1997).

Modern imaging has tested these candidate routes directly. Diffusion tractography in blindsight patients shows that the direct pathway from the lateral geniculate nucleus to area V5/MT is preserved in those who retain residual vision and degraded in those who do not, implicating this geniculo-extrastriate projection as a principal substrate of blindsight (Ajina et al., 2015). The collicular-pulvinar route, meanwhile, appears especially important for the non-conscious processing of biologically salient stimuli such as fearful faces, providing a subcortical channel to the amygdala (Tamietto & de Gelder, 2010). These pathways are not mutually exclusive; the residual vision of any given patient probably reflects a combination of whichever routes their lesion has spared (Ajina & Bridge, 2017).

The pathway account also makes sense of the Riddoch phenomenon, in which a patient with an occipital lesion can consciously perceive a moving object in an otherwise blind field while being unable to see the same object when it is stationary. First described by George Riddoch in soldiers with occipital gunshot wounds during the First World War (Riddoch, 1917), this dissociation of movement from static form was later linked to preserved activity in the motion area V5/MT, reachable from the retina without passing through V1 (Zeki & ffytche, 1998). The Riddoch phenomenon shows that residual vision is not always wholly unconscious: a spared extrastriate route can, in some cases, support a rudimentary awareness of motion.

Assessment and Diagnosis

Assessing cortical blindness has two aims: to confirm that the blindness is cortical rather than ocular, and to characterise what residual vision, if any, survives. The first rests on the mismatch already described — a normal eye examination, preserved pupillary reflexes, and a patient who cannot see — supported by neuroimaging that reveals the occipital lesion and by a visual evoked potential that is abnormal or absent because its generator in V1 is damaged (Celesia et al., 1991). Together these establish that the eyes and optic nerves are working and that the fault lies in the cortex.

Mapping the blind field itself is done by perimetry, in which the patient fixates a central point while small targets are presented at many locations and their detection is recorded, yielding a map of where vision is present and absent. Standard perimetry, however, relies on the patient's conscious report and so, by design, registers a blind field as simply blind; it cannot reveal blindsight, which requires forced-choice testing in which the patient must respond to stimuli they deny seeing, and the tell-tale signal is above-chance accuracy on trials the patient believes are guesses (Weiskrantz et al., 1974). The distinction is fundamental: one method measures what the patient can consciously see, the other what their visual system can do without awareness.

Careful psychophysics can go further, establishing which attributes a given patient's residual vision can process — location, motion, orientation, wavelength — and even that residual capacities can be trained. Repeated stimulation of a fixed location in the blind field can raise sensitivity there over time, a finding with direct implications for rehabilitation (Sahraie et al., 2006). Documenting the profile of spared and lost function in this way is both the clinical basis for prognosis and the empirical core of what cortical blindness has taught cognitive science.

Worked Example

Consider how a clinician establishes that blindsight is genuinely present rather than a statistical fluke. A patient with a dense left homonymous hemianopia is given a forced-choice localisation task: on each trial a spot of light appears at one of two positions in the blind field, and the patient must say “upper” or “lower” even though they report seeing nothing. Because there are two equally likely alternatives, pure guessing would yield 50% correct in the long run.

chance level = 1 ÷ (number of alternatives) = 1 ÷ 2 = 50%

Suppose the patient completes 100 trials and is correct on 72. Is 72% meaningfully above the 50% expected from guessing? The standard error of a proportion under the chance hypothesis is the square root of p(1 − p)/n, where p is the chance rate and n the number of trials.

standard error = √[(0.5 × 0.5) ÷ 100] = √0.0025 = 0.05

The observed proportion, 0.72, lies (0.72 − 0.50) / 0.05 = 4.4 standard errors above chance. A result more than four standard errors from the chance expectation would occur by luck far less than once in a thousand replications, so the patient's “guesses” are carrying real visual information even though they deny any experience of the stimuli. The example makes the logic of blindsight explicit: the diagnosis does not turn on what the patient reports seeing — by their account, nothing — but on whether their forced responses beat chance by a margin too large to attribute to luck. It is exactly why a forced-choice design, rather than the patient's report, is the indispensable tool (Weiskrantz, 1996).

Discussion

Cortical blindness has mattered to cognitive psychology far out of proportion to how often it is seen, because blindsight turns an abstract question — what is the relation between visual processing and visual consciousness? — into an observable dissociation. When a patient can point to a light they say they cannot see, the natural conclusion is that some visual processing proceeds without giving rise to conscious experience, and that the two can be pulled apart by the right lesion. Blindsight thus became a central case in the science of consciousness, offering a rare handle on what the primary visual cortex contributes to awareness that the surviving pathways do not (Cowey, 2010).

The interpretation has been vigorously contested, and that debate is part of the phenomenon's value. A persistent alternative holds that blindsight is not truly unconscious vision but degraded conscious vision — that a small amount of spared or nearby cortex supports a faint, hard-to-report experience rather than none at all. Ruling this out is what the animal work was designed to do: showing that a monkey with a striate lesion classifies a stimulus it can clearly detect as equivalent to a blank field is strong evidence that the residual capacity is unaccompanied by seeing (Cowey & Stoerig, 1995). The question of whether any given patient's residual vision is wholly unconscious remains one that must be settled case by case, and it bears directly on what blindsight can and cannot tell us about consciousness (Cowey, 2010).

For cognitive theory the enduring lesson is that the visual brain is not a single stream from eye to awareness but a set of parallel pathways with different destinations, only some of which yield conscious sight. That architecture explains how sight can outlast the loss of its principal cortical target, why residual vision is patchy and attribute-specific, and why the primary visual cortex occupies so privileged a place in theories of visual consciousness. It also reframes the clinical loss: a cortically blind field is not visually empty but visually silent to awareness, and the practical question becomes whether that silent processing can be recruited for recovery (Ajina & Bridge, 2017).

Current Directions

The most active current front is visual rehabilitation — the effort to turn residual capacity into usable recovered vision. Building on the finding that repeated stimulation can raise sensitivity in a blind field, controlled training studies have shown that intensive visual discrimination practice at fixed blind-field locations can recover aspects of conscious vision in chronic cortically blind patients, and that such training measurably improves performance on standard clinical perimetry (Melnick, Tadin, & Huxlin, 2016; Cavanaugh & Huxlin, 2017). Recovery in the chronic phase is typically restricted to the trained locations and demands sustained practice, which has focused attention on how to make gains generalise and endure (Das, Tadin, & Huxlin, 2014).

A second, related direction concerns timing and mechanism. Training begun in the subacute period soon after a stroke appears to preserve more visual function and to permit greater restoration than training begun once the deficit is chronic, suggesting a window of heightened plasticity that rehabilitation should exploit (Saionz, Tadin, Melnick, & Huxlin, 2020). High-resolution imaging has begun to identify the neural substrate of these gains, showing that recovered luminance detection depends on spared islands of activity in perilesional V1 — evidence that the residual tissue immediately around the lesion, not only the alternative subcortical routes, is a key engine of recovery (Barbot et al., 2021). Together these lines are moving the field from documenting what cortically blind patients can still do toward deliberately restoring what they have lost.

Common Misconceptions

“Cortical blindness means the eyes are damaged.”

The opposite: the eyes, retina, and optic nerves are intact, and the pupils still react to light. The fault is in the visual cortex of the brain, which is why the eye examination is normal in a patient who cannot see (Aldrich et al., 1987).

“Blindsight is just a little bit of remaining normal vision.”

In its clearest cases it is not. The patient sincerely reports no visual experience yet responds above chance, and animal studies show the residual capacity can accompany behaviour that treats the stimulus as unseen — a dissociation of performance from awareness, not a faint image (Cowey & Stoerig, 1995).

“A cortically blind field can never recover.”

Recovery is limited but real. Intensive visual training can restore aspects of conscious vision at trained locations and improve clinical perimetry, especially when begun soon after the injury (Cavanaugh & Huxlin, 2017).

“Every cortically blind patient knows they are blind.”

Not always. In Anton syndrome the patient denies the blindness altogether, confabulating descriptions of scenes they cannot see — a loss of awareness of the loss itself (Celesia et al., 1991).

Commonly Confused With

Cortical blindness vs. Charles Bonnet syndrome

Both involve vision loss and the brain, but they are opposites in what they add. Cortical blindness removes the conscious image while sometimes preserving unconscious processing; Charles Bonnet syndrome, by contrast, is the occurrence of vivid visual hallucinations in people losing their sight, typically from eye disease rather than cortical damage. One subtracts seeing, the other adds unbidden images.

Cortical blindness vs. visual agnosia

In cortical blindness the person cannot see the stimulus at all; in visual agnosia they see it perfectly well but cannot recognise what it is. Agnosia is a failure of recognition on intact perception, whereas cortical blindness is a failure of perception itself — a distinction of stage, not degree.

Glossary

Amygdala.

A subcortical structure involved in processing emotional salience; it can be driven by fearful faces presented to a blind field, without conscious sight, via a subcortical visual route.

Anton syndrome.

Visual anosognosia: a cortically blind patient denies being blind and confabulates descriptions of what they cannot see; a loss of awareness of the loss of vision.

Blindsight.

The ability of some patients with damage to the primary visual cortex to respond above chance to stimuli in their blind field while reporting no conscious visual experience of them.

Cortical blindness.

Loss of vision caused by damage to the visual cortex of the brain, with intact eyes, optic nerves, and pupillary reflexes.

Extrastriate cortex.

Visual cortical areas beyond the primary (striate) cortex, such as the motion area V5/MT; some can be reached from the retina without passing through V1 and support residual vision.

Forced-choice testing.

A method in which the patient must respond on every trial even when they feel they are guessing; above-chance accuracy reveals visual processing that conscious report would miss.

Geniculostriate pathway.

The dominant visual route from the retina through the lateral geniculate nucleus to the primary visual cortex; the substrate of the conscious visual image, destroyed in cortical blindness.

Homonymous hemianopia.

Blindness in the same half of the visual field of both eyes, produced by damage to the visual pathway or cortex of one hemisphere.

Lateral geniculate nucleus (LGN).

The thalamic relay that carries retinal signals to the primary visual cortex; a spared direct projection from it to extrastriate cortex is implicated in blindsight.

Perimetry.

The clinical mapping of the visual field by presenting small targets at many locations while the patient fixates centrally and reports what they detect.

Primary visual cortex (V1).

The first cortical stage of vision, in the occipital lobe (striate cortex, area 17); its destruction produces cortical blindness.

Pupillary light reflex.

The constriction of the pupil to light, mediated through the midbrain and independent of the cortex; preserved in cortical blindness, a key diagnostic sign.

Riddoch phenomenon.

The conscious perception of movement in an otherwise blind field while stationary objects there remain unseen; linked to preserved activity in the motion area V5/MT.

Superior colliculus.

A midbrain structure receiving direct retinal input; part of the retinotectal route that reaches extrastriate cortex via the pulvinar, bypassing V1.

Visual evoked potential (VEP).

An electrical response recorded over the occipital scalp in response to visual stimulation; abnormal or absent in cortical blindness because its generator in V1 is damaged.

Key Researchers

Alan Cowey

(deceased, 1935–2012). Visual neuroscientist at the University of Oxford whose comparative studies of blindsight in monkeys and humans, many with Petra Stoerig, provided some of the strongest evidence that residual vision after striate damage is genuinely unconscious (Cowey & Stoerig, 1995; Cowey, 2010). Wikipedia

Beatrice de Gelder

(living). Cognitive neuroscientist at Maastricht University whose work on affective blindsight showed that cortically blind patients can respond to the emotional expression of faces they cannot consciously see, driving the amygdala through a subcortical route (Tamietto & de Gelder, 2010). ORCID

Krystel R. Huxlin

(living). Vision scientist at the University of Rochester whose research programme established that intensive visual training can recover conscious vision in the blind fields of cortically blind patients and improve clinical perimetry, founding the modern approach to visual rehabilitation (Das, Tadin, & Huxlin, 2014; Cavanaugh & Huxlin, 2017). ORCID

George Riddoch

(deceased, 1888–1947). Scottish neurologist who, studying soldiers with occipital gunshot wounds in the First World War, first described the preserved perception of movement in an otherwise blind field — the dissociation later named the Riddoch phenomenon (Zeki & ffytche, 1998). Wikipedia

Arash Sahraie

(living). Vision scientist at the University of Aberdeen whose psychophysical studies of blindsight demonstrated that repeated stimulation of a fixed location in the blind field can increase residual sensitivity, laying groundwork for rehabilitation (Sahraie et al., 2006). ORCID

Lawrence Weiskrantz

(deceased, 1926–2018). Neuropsychologist at the University of Oxford who coined the term blindsight and, through his studies of the patient D.B., established the modern experimental study of vision without awareness after damage to the primary visual cortex (Weiskrantz et al., 1974; Weiskrantz, 1996). Wikipedia

Frequently Asked Questions

What is cortical blindness?

Cortical blindness is loss of vision caused by damage to the visual cortex of the brain, rather than to the eyes. The eyes, optic nerves, and pupillary reflexes are intact, but the brain can no longer build a conscious visual image, so the person cannot see despite a healthy visual apparatus.

How is cortical blindness different from blindness caused by the eyes?

In ocular blindness the fault is in the eye, retina, or optic nerve, and signs such as the pupillary light reflex are usually abnormal. In cortical blindness the eye examination is normal and the pupils still react to light; the damage is in the occipital cortex, and the visual evoked potential is typically abnormal or absent.

What is blindsight?

Blindsight is the ability of some patients with damage to the primary visual cortex to respond above chance to visual stimuli in their blind field — localising a light, detecting motion, or guessing a facial expression — while sincerely reporting that they see nothing. It reveals visual processing that proceeds without conscious awareness.

How can someone see without their visual cortex?

The retina sends signals along more than one route. Besides the main pathway to the primary visual cortex, older routes reach other visual areas through the superior colliculus and pulvinar, and a direct projection to the motion area V5/MT can survive the lesion. These spared pathways carry visual information that can guide behaviour without producing a conscious image.

What causes cortical blindness?

The commonest cause is stroke affecting both occipital lobes, but it can also follow head trauma, cardiac arrest or other oxygen deprivation, eclampsia, severe hypertension, or infection. In children it often results from perinatal hypoxia or brain malformation.

What is the Riddoch phenomenon?

It is the ability to consciously perceive a moving object in a blind field while being unable to see the same object when it is still. First described by George Riddoch a century ago, it is linked to preserved activity in the brain's motion area, reachable without the primary visual cortex.

Can cortical blindness be treated or improved?

Recovery is limited but possible. Intensive, repeated visual training at specific blind-field locations can restore aspects of conscious vision and improve performance on clinical field tests, and training appears more effective when begun soon after the injury, during a window of heightened plasticity.

What does blindsight tell us about consciousness?

It shows that visual processing and visual awareness can be separated: information can be registered and acted upon without any conscious experience of seeing. This dissociation makes cortical blindness a central case in the science of consciousness and highlights the special role of the primary visual cortex in conscious sight.

References