Desk Lighting and Screen Glare: The Posture Problem Nobody Measures

Everything else on a desk gets measured. Desk height gets a tape measure, monitor height gets a published angle range, chair seat depth has a number and a way to check it. Light is the one variable people adjust by feel, and it is the one OSHA explicitly ties to posture as well as to eyesight.

The wording in the Computer Workstations eTool is worth reading slowly. If lighting is excessive or causes glare on the monitor screen, you may develop eyestrain or headaches, "and may have to work in awkward postures to view the screen." A federal ergonomics resource is saying that a reflection on your display is a musculoskeletal problem, because the head moves to get around it.

Why light becomes posture

A reflection sits in a fixed place relative to the panel and the light source. Your head does not. So when a window or a ceiling fixture puts a bright patch across the top third of your display, you rarely get up and close the blind. You tilt, you slide down in the chair, you lean to one side, or you move your face closer to read through the washed-out region. Each of those is a posture held for as long as the sun is in that part of the sky.

The opposite failure does the same thing. A screen that is legible but under-lit relative to the paper beside it pulls you forward, because the cheapest way to raise apparent contrast is to shorten the viewing distance. That is behind a lot of neck ache that gets blamed on the monitor arm. If your screen already sits at the right height and you still finish the day with a stiff upper back, check the geometry against the five published monitor height targets before buying anything else.

The two numbers that fight each other

This is why one ceiling light is never right for a desk with both a screen and paper on it.

The Canadian Centre for Occupational Health and Safety publishes an illuminance table drawn from the IESNA Lighting Handbook, CSA Z412:24 and the British Columbia OHS Regulations. Computer tasks sit at 75 to 300 lux, with the range explained by screen type and polarity. Reading medium-contrast or small print sits at 500 to 1000 lux. General office lighting is usually 500 to 1000 lux measured 76 cm, or 30 inches, above the floor.

Those two requirements do not overlap at all. A room lit correctly for reading a printed page is at minimum 1.7 times brighter than the top of the screen band, and can be more than 13 times the bottom of it.

OSHA states the same tension in foot-candles. For paper tasks and offices with CRT displays, office lighting should range between 20 and 50 foot-candles, with a minimum of 30 foot-candles in the construction industry under 1926.56. Where LCD monitors are in use, higher levels are usually needed for the same viewing tasks, up to 73 foot-candles. One foot-candle is 10.76 lux, so 20 to 50 foot-candles is roughly 215 to 538 lux, and 73 foot-candles is about 785 lux.

Task	Published target	Source
Computer screen work	75 to 300 lux.	CCOHS illuminance table.
Reading medium contrast or small print	500 to 1000 lux.	CCOHS illuminance table.
General office lighting	500 to 1000 lux at 76 cm above the floor.	CCOHS illuminance table.
Paper tasks, offices with CRT displays	20 to 50 foot-candles, about 215 to 538 lux.	OSHA Computer Workstations eTool.
Same viewing tasks with LCD monitors	Up to 73 foot-candles, about 785 lux.	OSHA Computer Workstations eTool.

An overhead fixture set bright enough to read a delivery note by is too bright for the display, and one set correctly for the display leaves the paperwork in shadow. OSHA's answer is to split them: task lighting for writing and reading, with brightness around the monitor kept down.

One study actually measured the posture

Most lighting research measures eye fatigue and stops. A 2015 field study in the journal Work scored posture as well.

Joines and colleagues recruited 100 participants and randomly assigned them to an intervention group, which received an adjustable LED task light, and a control group. Alongside self-reported eye fatigue and musculoskeletal discomfort, the researchers recorded workspace illumination and assessed posture with RULA, a standard observational scoring method, during standardised tasks. Between baseline and short-term follow-up, the intervention group improved significantly on discomfort, eye fatigue, perception of job content and posture. The authors reported no negative results from task light use.

That is a small field study with a short follow-up, and it tells you what it tells you. A task lamp is still one of the very few desk interventions with a randomised comparison that scored posture as an outcome at all.

What the muscle studies add, and where they stop

Two studies from the Centre for Musculoskeletal Research at the University of Gävle explain the mechanism behind the discomfort half of that finding.

In a 2013 paper in the Journal of Electromyography and Kinesiology, 66 participants, 33 with neck pain and 33 without, performed visually demanding near work through four trial-lens conditions. Eye-lens accommodation on its own did not significantly affect trapezius muscle activity. When accommodation and convergence were made incongruent, a significant positive relationship appeared between the two, with no difference between the pain and control groups.

A 2015 follow-up in the European Journal of Applied Physiology used the same 66 participants, median age 38, fixating a contrast-varying grating from 65 cm through four counterbalanced seven-minute periods. Trapezius activity increased significantly over time in every lens condition, which the authors attributed to the mental effort needed to hold performance steady as the task wore on.

Be precise about what that shows. Those experiments manipulated the eye with lenses. No lamp was moved. What they establish is that visual effort loads the trapezius, and that the load builds over minutes instead of appearing at once. Glare and insufficient light are two ordinary ways of raising visual effort at a desk, which makes the link plausible and cheap to act on, though it is not proven end to end.

The scale of the problem is not in doubt. A 2025 literature review in Future Science OA put reported computer vision syndrome prevalence at 69.0%, listed neck pain among the extraocular symptoms, and named glare and reduced screen contrast among the factors that worsen it.

The window rule

One fix on OSHA's list is free, takes thirty seconds, and is wrong in most home offices.

Place the face of the display at right angles to windows and light sources. Not facing the window, which puts a bright field behind the screen and forces your pupils to adapt to the window instead of the text. Not backing onto it either, which lays the window's reflection across the panel. Perpendicular.

Blind type matters too, and the eTool is specific: vertical blinds work best for east and west facing windows, horizontal blinds for north and south. That follows from where the sun sits when it is low enough to be a problem.

Two more from the same page that cost nothing. Tilt the monitor down slightly so it stops reflecting the ceiling fixture back at you. Set the display to dark characters on a light background, which OSHA notes are less affected by reflections than light characters on a dark one.

Surfaces, not just fixtures

Glare does not only arrive directly. CCOHS publishes suggested reflectance percentages for a typical office, which make a useful audit of the room itself.

Surface	Suggested reflectance	What a miss looks like
Ceiling	70 to 80%.	A dark ceiling forces brighter, harsher fixtures.
Walls	50% maximum.	Gloss paint beside the desk becomes a second light source.
Window blinds	40 to 50%.	Bright blinds re-radiate the window you just covered.
Furniture	25 to 45%.	A gloss white desktop throws the ceiling light up at your face.
Business machines	50% maximum.	A glossy keyboard produces reflected glare of its own.
Floor	20 to 40%.	Occasionally the cause with a polished floor and a low sun.

CCOHS adds a point a desk layout guide will not: workers sitting with their backs to windows, or with fixtures overhead or behind them, cast shadows on their own work surface. OSHA adds that many low-powered lamps produce less glare than a few high-powered ones, and that walls and surfaces near the screen should be a medium colour in non-reflective paint.

The order to fix it in

The first five of these cost nothing.

1. Turn the desk so the screen face is at right angles to the window. If the desk cannot move, the monitor often can.
2. Adjust the blind in your field of view, vertical for east and west, horizontal for north and south.
3. Tilt the monitor down a few degrees until the ceiling fixture is out of the panel.
4. Switch to dark text on a light background in the application you spend the most hours in.
5. Clean the screen. Dust scatters light across the whole panel instead of reflecting it in one spot, which is why a dusty screen looks washed out instead of mirrored.
6. Add an aimable task light for paper, positioned so none of its output lands on the display. This is the intervention the Joines study tested.
7. Reduce the overhead level if the screen still looks grey. OSHA suggests diffusers, louvers or egg crates in the fixture, and removing the middle bulbs from a four-bulb fluorescent unit.

A glare filter that attaches to the monitor surface is on OSHA's list of options, with the caveat that it should not significantly decrease screen visibility, but it belongs after the six free fixes and not in place of them.

Where lighting sits among the other fixes

Light is worth auditing early because it is free to change and quick to check, but it is no substitute for the measurements. A screen at the wrong height stays wrong in perfect light. If the geometry has not been checked, start with the full sequence of desk measurements in the order they need doing, then come back to the room.

Three cases where light and geometry interact. A laptop on its own keyboard puts the screen low and close, raising neck flexion and visual demand together, and the fix order is in the guide to why laptop neck pain is a geometry problem first. Two screens multiply the angles a window can reflect from, so the rotation advice in positioning two screens without twisting your neck has to be reconciled with the right-angle window rule. And pushing a monitor back to escape a reflection needs somewhere to push it to, which is how deep a desk actually needs to be.

Questions people actually ask

How bright should a desk be for computer work?

CCOHS puts computer tasks at 75 to 300 lux and general office lighting at 500 to 1000 lux measured 30 inches above the floor. OSHA gives 20 to 50 foot-candles for paper tasks and CRT offices, rising to 73 foot-candles where LCD monitors are used. The two bands do not overlap, which is why a separate task light beats one brighter ceiling fixture.

Which way should a desk face relative to a window?

Sideways. OSHA's guidance is to place the face of the display screen at right angles to windows and light sources. Facing the window puts a bright field behind the screen, and backing onto it lays a reflection across the panel.

Does a desk lamp improve posture, or just comfort?

A 2015 field study in Work randomly assigned 100 office workers to an adjustable LED task light or a control condition and scored posture with RULA. The intervention group improved significantly on discomfort, eye fatigue and posture between baseline and short-term follow-up.

Are blue light glasses a fix for this?

The 2025 Future Science OA review of computer vision syndrome reports that blue light blocking spectacles show limited efficacy against its symptoms and that the role of blue light remains contested. Glare position and illuminance are the variables with published targets attached.

Is dark mode better for glare?

For reflections, no. OSHA recommends dark characters on a light background, because dark-on-light is less affected by reflections. Dark mode is still preferable in a genuinely dim room, where the screen would otherwise be the brightest surface present.

Sources

• OSHA Computer Workstations eTool, Workstation Environment, for the awkward-posture wording, the foot-candle figures, the right-angle window rule, blind orientation, monitor tilt, dark-on-light characters, diffusers, louvers and screen cleaning.
• 29 CFR 1926.56, Illumination, for the 30 foot-candle construction industry office minimum.
• CCOHS: Lighting Ergonomics, Survey and Solutions, for the illuminance table, the 76 cm measurement height, the reflectance percentages and the shadow-casting note. Its figures come from the IESNA Lighting Handbook 9th edition, CSA Z412:24 and the British Columbia OHS Regulations Section 4.65.
• Joines S, James T, Liu S, Wang W, Dunn R, Cohen S. Adjustable task lighting: field study assesses the benefits in an office environment. Work, 2015, 51(3), 471-481, for the 100-participant design, the RULA posture assessment and the outcomes.
• Richter HO, Zetterberg C, Forsman M. Trapezius muscle activity increases during near work activity regardless of accommodation/vergence demand level. European Journal of Applied Physiology, 2015, 115(7), 1501-1512, for the 65 cm fixation distance, the four seven-minute periods and the increase over time.
• Zetterberg C, Forsman M, Richter HO. Effects of visually demanding near work on trapezius muscle activity. Journal of Electromyography and Kinesiology, 2013, 23(5), 1190-1198, for the accommodation and convergence incongruence finding.
• Kahal F, Al Darra A, Torbey A. Computer vision syndrome: a comprehensive literature review. Future Science OA, 2025, 11(1), 2476923, for the 69.0% reported prevalence, neck pain among the extraocular symptoms, and the blue light spectacle efficacy note.