A visual field test maps the sensitivity of your vision at dozens of individual points across your field of view, then compares those measurements against age-matched normal values. The printout you receive is dense with numbers, probability plots, and summary statistics, but its core logic is straightforward: dark or flagged spots mean reduced sensitivity, and the pattern those spots form tells your doctor where along the visual pathway a problem might exist. Understanding a few key sections of the report turns what looks like an inscrutable grid into a readable picture of how well you see.
What the Test Actually Measures
The most common version, standard automated perimetry (usually the Humphrey Visual Field Analyzer), tests one eye at a time. You stare at a central fixation point inside a bowl-shaped instrument while small lights of varying brightness flash at different locations. Every time you see a flash, you press a button. The machine adjusts the brightness up and down at each location to find the dimmest light you can detect there, a value measured in decibels (dB). Higher decibel values mean you can see fainter stimuli, which means better sensitivity. The entire process takes roughly four to eight minutes per eye, depending on the test pattern used.
The two most common test patterns are the 24-2 and the 30-2. The numbers refer to how many degrees of the visual field are tested. A 24-2 covers the central 24 degrees in each direction, which captures most of the visual field relevant to daily activities and is the standard for glaucoma monitoring. A 30-2 extends slightly further and is often used in neuro-ophthalmology. A 10-2 pattern zooms in on just the central 10 degrees and is useful when damage is concentrated near fixation.
Reliability Indices Come First
Before interpreting what the test found, you need to know whether to trust the results. Three reliability metrics appear near the top of most printouts: fixation losses, false positives, and false negatives. Each one captures a different way the test can go wrong.
Fixation losses measure how often your eye wandered from the central target during the test. The machine periodically projects a stimulus onto your physiological blind spot, where you should not be able to see anything. If you press the button anyway, it suggests you were not looking at the fixation point. A high fixation loss rate means the sensitivity measurements may have been recorded at the wrong locations in your visual field.
False positives occur when you press the button even though no stimulus was presented. This is essentially trigger-happy responding. Research shows false positives have a bigger effect on test accuracy than the other reliability measures, and their impact grows sharply once they exceed about 20% of catch trials. Beyond that threshold, each additional 10% of false positives inflated the mean deviation by roughly 1.5 to over 3.5 dB depending on disease severity, making the field look better than it actually is.1PubMed Central. Evidence-based Criteria for Assessment of Visual Field Reliability That distortion is particularly dangerous in glaucoma monitoring because it can mask real progression.
False negatives happen when you fail to respond to a bright stimulus in a location where you previously responded to a dimmer one. A high false-negative rate can signal inattention or fatigue, but it can also simply reflect genuine visual field damage, since damaged areas produce more variable responses. One study found that as the false-negative rate climbed from 0% to 16%, the probability of a false-positive visual field classification (incorrectly labeling someone as having a normal field) rose from about 9% to 82%.2JAMA Ophthalmology. Role of Visual Field Reliability Indices in Ruling Out Glaucoma In other words, unreliable tests do not just add noise; they can completely obscure real disease.
Fixation losses, by contrast, had relatively little impact on the overall accuracy of the mean deviation, shifting it by less than 0.2 dB per 10% increase regardless of disease severity.1PubMed Central. Evidence-based Criteria for Assessment of Visual Field Reliability That does not mean fixation losses are irrelevant, but false positives deserve the most scrutiny when you are evaluating whether to trust a given test.
The Grayscale Plot
The grayscale image is the most visually intuitive part of the printout. It converts sensitivity values into shades of gray: lighter areas represent normal or near-normal vision, while darker areas represent reduced sensitivity. Completely black regions indicate no measurable response to the brightest stimulus the machine can produce. Most people look at the grayscale first, and it does give a quick impression, but clinicians treat it cautiously because small differences in decibel values can look similar in grayscale shading. Two locations that differ meaningfully in sensitivity might appear the same shade of gray. The grayscale is useful for getting the overall shape of a defect but unreliable for fine distinctions.
Total Deviation and Pattern Deviation
Below the grayscale, most printouts show two grids of numbers and two probability maps: total deviation and pattern deviation. These are the most informative parts of the report for understanding what is wrong and where.
Total deviation compares the measured sensitivity at each test point against the expected sensitivity for someone your age. If your sensitivity at a given point is 5 dB below normal, the total deviation value at that location will be roughly –5. The accompanying probability plot flags points that fall outside normal ranges, using symbols that correspond to different significance levels (typically marking the worst 5%, 2%, 1%, and 0.5% of the normal distribution). Total deviation catches everything that reduces sensitivity, whether it is a localized disease process, a cataract clouding the whole field, or an uncorrected refractive error.
Pattern deviation takes the total deviation data and mathematically adjusts for any overall depression of the field. It shifts all the values so that the general height of the visual field is closer to normal, then flags the points that remain abnormal. The effect is to strip away diffuse causes of reduced sensitivity (like cataract) and highlight localized defects. If you have early glaucoma and a cataract in the same eye, the pattern deviation map is more likely to reveal the glaucomatous damage that the cataract would otherwise mask on the total deviation map.
This distinction matters for tracking changes over time. Research comparing total and pattern deviation analyses in glaucoma found that pattern deviation classified about 15% fewer eyes as progressing, and when both methods agreed progression was happening, total deviation tended to detect it earlier.3PubMed. Visual field progression in glaucoma: total versus pattern deviation analyses The reason is that glaucomatous damage almost always includes a diffuse component alongside the focal loss, and pattern deviation, by design, adjusts away that diffuse component. Neither map is inherently better; they answer different questions. Total deviation asks “how does this eye compare to normal?” Pattern deviation asks “is anything abnormal beyond any overall dimming?”
Global Indices That Summarize the Whole Field
Three summary numbers distill the entire visual field into single values. Mean deviation (MD) is the average of all the total deviation values across the field, weighted to give more importance to central points. A normal MD is close to 0 dB. A negative MD means overall sensitivity is below average; the more negative the number, the worse the field. An MD of –2 dB is mild, while –12 dB indicates substantial loss.
Pattern standard deviation (PSD) measures how unevenly the field losses are distributed. A normal field has a low PSD because sensitivity is relatively uniform. Early focal damage, like a small cluster of depressed points, raises the PSD. Paradoxically, in very advanced disease where everything is depressed, PSD can actually drop back down because the field becomes uniformly poor again.4PubMed Central. Suitability of the Visual Field Index according to Glaucoma Severity
The Visual Field Index (VFI) expresses the field as a percentage, where 100% is a fully intact field and 0% is a completely extinguished one. VFI is weighted toward the center of the field and is designed to be less affected by cataract than MD. It correlates strongly with MD, with studies showing correlation coefficients above 0.95.5PubMed Central. Correlation between Visual Field Index and Other Functional and Structural Measures in Glaucoma Patients and Suspects VFI is also the main index used in trend-based progression analysis, where its value is plotted over time and a slope is calculated to estimate how fast vision is declining.
Recognizing Glaucomatous Patterns
Glaucoma does not damage the visual field randomly. It follows the anatomy of the nerve fiber layer in the retina, which arcs over and under the macula on its way to the optic nerve. Because of this architecture, early glaucomatous defects tend to appear in characteristic locations. Research categorizing initial visual field defects in open-angle glaucoma found four main starting patterns: superior paracentral defects, inferior paracentral defects, superior nasal defects, and inferior nasal defects.6PubMed Central. Location of Initial Visual Field Defects in Glaucoma and Their Modes of Deterioration
The classic early glaucoma defect is an arcuate (arc-shaped) scotoma that curves from the blind spot around fixation, following the path of the nerve fiber bundles. A nasal step, where sensitivity drops abruptly across the horizontal midline on the nose side of the field, is another hallmark. These defects respect the horizontal midline because the nerve fibers above and below the retinal horizontal raphe drain separately to the optic nerve. A defect that crosses the horizontal midline freely is less likely to be glaucoma and more likely to have a neurological origin.
As glaucoma advances, isolated scotomas deepen and enlarge, eventually merging into broader arcuate defects. Late-stage disease can leave only a small island of central vision and sometimes a temporal crescent. Different glaucoma subtypes can show different starting patterns; a systematic review noted that open-angle glaucoma, normal-tension glaucoma, angle-closure glaucoma, and juvenile open-angle glaucoma each show somewhat distinct visual field patterns, though the most progression data exists for open-angle and normal-tension subtypes.7PubMed Central. Visual field patterns in glaucoma: A systematic review
Neurological Patterns and What They Mean
When visual field loss is caused by a problem behind the eye, somewhere along the optic nerve, chiasm, or brain, the pattern on the printout changes dramatically. The most telling distinction is which midline the defect respects. Glaucomatous defects respect the horizontal midline. Neurological defects from the optic chiasm or beyond respect the vertical midline, producing loss that affects the same side (right or left) in both eyes.
A homonymous hemianopia, where both eyes lose the same half of the visual field, is the most common pattern from post-chiasmal lesions like strokes or tumors. A large study of 904 cases of homonymous hemianopia found that roughly 38% were complete (the entire half-field gone) while 62% were incomplete. Among incomplete types, quadrantanopia, where just one quarter of the field is lost, was the most common at 29%.8PubMed. Homonymous hemianopias: clinical-anatomic correlations in 904 cases
Bitemporal hemianopia, where both eyes lose the outer (temporal) half of the field, points to compression at the optic chiasm, often from a pituitary tumor pressing on the crossing nerve fibers. This pattern is distinctive because the damage affects the nasal retinal fibers from both eyes as they cross at the chiasm.
An interesting feature of some hemianopias is macular sparing, where central vision remains intact despite the surrounding half-field being blind. This occurs because the occipital pole, the area of the brain that processes central vision, often receives a dual blood supply from both the posterior and middle cerebral arteries. When a posterior cerebral artery stroke knocks out the visual cortex responsible for peripheral vision, this collateral blood flow can keep the macular representation alive.9PubMed Central. The Mechanism of Macular Sparing
Artifacts That Mimic Real Disease
Not every abnormality on a visual field printout reflects actual vision loss. Artifacts can produce convincing-looking defects that lead to misdiagnosis if not caught. Lens rim artifact is one of the most common. During the test, a corrective lens is placed in front of your eye. If that lens or its holder is not positioned correctly, its rim can block peripheral stimuli and create a pattern of apparent field loss, typically a ring or partial ring of depressed points around the edges. One retrospective review found lens rim artifact in about 10% of fields performed with corrective lenses, and it led to both overdiagnosis and underdiagnosis of real conditions.10Ophthalmology. Lens Rim Artifact in Automated Threshold Perimetry Risk factors include older age and high hyperopic (farsighted) corrections.
Droopy eyelids are another culprit. If the upper lid sags enough to block the superior visual field, the printout will show a band of depressed points across the top, mimicking a superior arcuate defect. In the large Ocular Hypertension Treatment Study, testing artifacts from lens rim and droopy lids accounted for about 1% of all visual fields, with lid artifacts typically causing superior depression and lens rim artifacts causing peripheral or inferior abnormalities.11Journal of Glaucoma. Visual Field Quality Control in the Ocular Hypertension Treatment Study (OHTS)
Even technically “reliable” fields (those passing all the standard reliability criteria) can harbor artifacts. A review from a neuro-ophthalmology practice documented cases where uncorrected refractive error, particularly significant astigmatism, produced focal abnormalities rather than the expected uniform dimming.12PubMed Central. Artifactual Visual Field Defects Identified on Technically “Reliable” Visual Field Studies in a Neuro-Ophthalmology Practice The takeaway is that a single abnormal visual field should always be confirmed with repeat testing before making clinical decisions.
The Learning Effect
If you have never taken a visual field test before, your first result is likely to underestimate your true visual function. The test is unusual and unintuitive: sitting in the dark, staring at a fixed point, pressing a button for flashes you are not sure you saw. People get better at it simply through practice, a phenomenon called the learning effect. One study of both healthy subjects and glaucoma patients found significant improvement in reliability parameters and global indices with each successive test visit. In glaucoma patients, the researchers recommended performing at least three tests before accepting a baseline, while normal subjects could achieve a reliable baseline by the second test.13PubMed Central. Impact of learning effect on reliability factors and global indices in visual field testing
This matters because if your clinician is going to track your visual field over time, the starting point needs to be accurate. A first-ever test that looks worse than reality will make it seem like your field improved on the next visit, even if nothing changed. The learning effect is especially pronounced with certain test types. With frequency-doubling technology perimetry, one study of healthy subjects showed that false-positive classification rates ranged from 50% to 64% on the first session but dropped to 18% to 32% by the third session.14Eye. Learning effect in visual field testing of healthy subjects using Humphrey Matrix frequency doubling technology perimetry That means more than half of normal people initially looked abnormal.
Tracking Progression Over Time
A single visual field test is a snapshot. The real clinical value comes from comparing multiple tests over months and years to determine whether vision is stable or worsening. Two main analytical approaches exist for this.
Event-based analysis compares each new test to your established baseline and flags any point or cluster that has deteriorated beyond what would be expected from normal variability. The Guided Progression Analysis (GPA) is a widely used version of this approach on Humphrey machines. It classifies change at each test location as “possible progression” or “likely progression” based on whether the deterioration repeats on consecutive tests.
Trend-based analysis plots VFI over time and fits a regression line to the data. The slope of that line tells you how fast the field is declining in percentage points per year. A slope of –1% per year, for instance, means the field is losing roughly 1% of its function annually. This lets your doctor project where you might be in five or ten years if the current trajectory continues. More sophisticated methods combine both approaches, using event-based findings to inform the trend analysis and improve sensitivity to real change.15PubMed Central. Integrating Event- and Trend-based Analyses to Improve Detection of Glaucomatous Visual Field Progression
Either way, the quality of progression analysis depends on having enough data points. A handful of tests spread over many years gives a weak trend line. Most clinicians aim for at least five or six tests in the first two years after diagnosis to establish a reliable trajectory, then continue with regular testing to confirm any apparent changes are real and not just test variability.
How Visual Fields Connect to Imaging
Visual field testing measures function: what you can see. Optical coherence tomography (OCT) measures structure: the physical thickness of the nerve fiber layer and retinal ganglion cells. Clinicians interpret the two together because structure and function do not always move in lockstep. In early disease, structural thinning on OCT may appear before any measurable visual field loss, because the retina has enough redundant nerve fibers to compensate. In advanced disease, the structural changes plateau while the field continues to worsen, because there is not much more nerve fiber left to lose.
Lesions affecting the front of the visual pathway (the retina and optic nerve) produce field defects that respect the horizontal midline, reflecting the arcuate path of ganglion cell axons. OCT in these cases typically shows thinning in the peripapillary nerve fiber layer and ganglion cell complex that corresponds topographically to the field defect.16PubMed. Visual fields and optical coherence tomography in neuro-ophthalmology: Structure-function correlation When the two tell the same story, the diagnosis is more secure. When they disagree, clinicians look more carefully at whether the field test was reliable, whether the OCT has its own artifacts, or whether something else is going on.
In the macular region specifically, refined mapping techniques that carefully align OCT sectors with the corresponding visual field test points produce stronger correlations between structural and functional measures.17Scientific Reports. Elucidating macular structure–function correlations in glaucoma This matters because central visual field testing (the 10-2 pattern) is increasingly used alongside macular OCT scans to catch early damage near fixation that a standard 24-2 might miss.
Why the Test Feels So Unpleasant
Patients consistently describe visual field testing as one of the more stressful ophthalmic procedures. The combination of darkness, head restraint, prolonged concentration, and the need to keep one eye covered creates what some people experience as claustrophobia. One qualitative study captured a participant’s description: “It is so claustrophobic. Your head is in that thing. You have to keep concentrating on that cross and you just feel wrecked by the end of it, and then you realize that you have to do the other eye too.”18Journal of Glaucoma. Patient Experience and Barriers to the Visual Field Test for Glaucoma
This is not just a comfort issue. Anxiety directly affects test performance. Studies have found that explaining the test procedure beforehand or showing a short training video significantly improved reliability scores, likely by reducing the stress and confusion that lead to inconsistent button presses.19PubMed Central. Study of anxiety in patients with glaucoma undergoing standard automated perimetry and optical coherence tomography If you are anxious about an upcoming test, asking the technician to walk you through what will happen before it starts is a practical way to improve both the experience and the accuracy of your results.
When Visual Field Loss Affects Everyday Life
The numbers on a visual field printout translate into real-world consequences, though the relationship is not always straightforward. Peripheral field loss affects mobility, navigation, and driving. Superior field loss makes it harder to see overhead obstacles like signs and branches; inferior field loss affects tasks like reading and walking down stairs. Central field loss impairs reading and face recognition.
The speed of decline matters as much as the absolute level. A study of glaucoma patients found that those who reported falling in the previous year had faster rates of binocular visual field loss than non-fallers, at roughly twice the rate. Rapid visual field loss was significantly associated with fall risk even after accounting for other factors like age.20JAMA Ophthalmology. Association of Fast Visual Field Loss With Risk of Falling in Patients With Glaucoma This finding underscores why progression analysis is not just an academic exercise; identifying fast decliners early gives clinicians a chance to intervene with more aggressive treatment or refer patients for fall-prevention strategies.
Specialized Test Types and Their Roles
Standard automated perimetry using white-on-white stimuli is the gold standard, but specialized variants exist. Short-wavelength automated perimetry (SWAP) uses a blue stimulus on a yellow background to isolate a subset of retinal ganglion cells thought to be damaged early in glaucoma. Frequency-doubling technology (FDT) perimetry uses low-spatial-frequency gratings that flicker at high temporal frequency, targeting another vulnerable cell population. Both were developed with the hope of catching glaucoma damage before standard perimetry could.
Research comparing these approaches has produced a mixed verdict. At least 20% of patients with structural glaucoma damage but normal standard fields showed defects on FDT or SWAP, suggesting these tests can sometimes detect functional loss earlier.21PubMed. Can frequency-doubling technology and short-wavelength automated perimetries detect visual field defects before standard automated perimetry in patients with preperimetric glaucoma? However, when used to monitor established glaucoma over time, neither SWAP nor FDT showed a clear advantage over standard perimetry in detecting progression. Intriguingly, each test type sometimes detected progression in different individual eyes, suggesting the methods may be capturing partially different aspects of retinal damage.22PubMed Central. Comparison of Standard Automated Perimetry, Short-Wavelength Automated Perimetry, and Frequency-Doubling Technology Perimetry to Monitor Glaucoma Progression
Artificial Intelligence and Automated Interpretation
Reading visual fields is a skill that takes years to develop, and even experienced clinicians sometimes disagree on whether a field is normal or abnormal. AI models are being developed to standardize interpretation. One recent model trained to classify visual fields as normal or altered achieved 80% sensitivity and about 95% specificity, with an overall diagnostic accuracy (area under the curve) of 0.93.23PubMed Central. Artificial intelligence-based model for the interpretation and reporting of standard automated perimetry Those numbers mean it catches most abnormal fields while rarely misclassifying normal ones. AI tools like this are not yet replacing clinical judgment, but they are moving toward a role as a second reader, flagging fields that deserve closer attention and potentially reducing the workload in high-volume clinics where hundreds of visual fields are generated each week.
The broader trajectory is toward integrating visual field data with OCT scans, patient demographics, and longitudinal trends into unified predictive models. For patients, the practical implication is that future visual field reports may come with AI-generated summaries that highlight concerning patterns and estimate progression risk, making the dense printout less dependent on a single clinician’s expertise to decode.