Severe Sensitivity to Light Emitted from LED-Backlit Displays

[MagnuM]

As an update, I'm back to the drawing board, and back to my Dell U2410 (CCFL). My last update where I mentioned how my brain/body "put up" with a problematic display over time did a hairpin turn one day where I suddenly felt very hot and unwell and felt like I would pass out or puke or something. I had bought the Dell P2425 LED-backlit IPS on June 20th, invoiced June 22nd, received it in shipping June 26th. I tried to return it on July 23rd after that bad neurological episode, but Dell refused the return saying my 30-day return policy had elapsed. They claimed that the clock starts on your invoice date, not when you receive the monitor. Talk about false advertising. Won't be bothering with that company again. Here I tried to be patient and try to find a solution when I should have just reboxed this thing and sent it back to them. Now I have to try to sell it for a loss.

You're absolutely right that our species has not looked a light source other than those produced by black body radiation sources such as the sun, fire, lanterns, candles, incandescent lightbulbs, etc, up until only a few decades ago when fluorescent lighting was introduced (and gave some poor susceptible people migraines). People have correctly pointed out that there is a lot of blue light emitted from the sun. However, I don't have any problems from the sun or being outside. I only have problems on LED-backlit computer monitors. Perhaps it is because there is such a sharp spike in the blue range which isn't balanced out by all the other colors? I heard that our eyes/brain will essentially take the average of a color spectrum when looking at the color white (which might explain why whites look so "blue" from LED light like harsh car headlights).

I guess I should remind myself not to get stuck on a story or theory to keep bias in check when experimenting. I'm so sure that blue light is causing my neurological symptoms, even though the more recent studies suggest that blue light does not cause eye strain. I do counter back with "does it cause 'brain strain'?" though.

Take the curious case of the user Jason38 here. They are the only user that described my issue basically perfectly. They suffered these symptoms for over a decade like I did, but they seemed to have it even worse than I did. They had to wear these dark orange SCT glasses just to use any monitor. I bought a pair. They are so aggressive that a pure yellow square on a white background will be completely camouflaged because the white background is as yellow as the square is. Does it cut down my symptoms on an LED? Sure does, but at a cost.

The user above hasn't posted for years because they solved their issue and likely went on with their life. They purchased a ViewSonic XG270 monitor, and they mentioned that they can use the monitor 16 hours a day if they wanted to without issues, and without the dark orange SCT glasses above. Looking at the specs of this monitor, it makes 0 sense why this monitor is working so amazingly well for them. Perhaps the dark orange SCT glasses were a red herring all along for all we know, and blue light was only part of the answer. All I can tell you from what I have read is that the ones that had the most success did so by pure trial-and-error, trying the most amount of monitors they could, until one worked. When they found the one that worked, they did not know why it worked, and frankly did not care.

The more recent scientific studies did agree that blue light does seem to affect sleep though. Oddly enough, I don't really seem to notice any affects on my sleep. I can use a computer right up to when I hop into bed and fall asleep without much difficulty. Perhaps it is affecting the quality of my sleep though for all I know.

I also have no issues with sunlight or looking at a pure blue sky. I don't really need shades to be outside. I want to stress that I really only have issues with computer monitors backlit by an LED. Similar less intense symptoms can be reproduced on this Dell U2410 CCFL if I increase the brightness from 0% to 50%. I will do this purposefully for color-sensitive work, put up with the pain and brain-strain symptoms, and revert back to 0% when I'm done. The problem with trying to put an LED monitor to 0% that I have found is that it is so dark and dim that I can barely see anything, while 0% on this Dell U2410 is still surprisingly bright (especially at night).

You are right that perhaps the reason I don't seem to have much issues with phones and TVs is FOV (field-of-view). The binocular vision therapy office that I'm currently attending pointed this out as well. The phone is small, so there are much other non-phone things my eyes and brain see while using it. Same with a TV in the distance. A computer monitor though is like... right there taking up most of what you see, so that's a lot of directional light coming at ya from pretty close. A good test for me honestly would be using a TV at a distance, because I haven't tested that for 8+ hours a day before.

Basically I'm trying to find the root cause of my sensitivity to be able to plan for the future properly. I know my Dell U2410 won't last forever, and neither will all other CCFLs.

[MPRT|GTFO]

I only have problems on LED-backlit computer monitors. Perhaps it is because there is such a sharp spike in the blue range which isn't balanced out by all the other colors? I heard that our eyes/brain will essentially take the average of a color spectrum when looking at the color white (which might explain why whites look so "blue" from LED light like harsh car headlights).

It's not simply one average. A more detailed understanding about how the spectrum is sensed and processed may be relevant for your case. You can check the Wikipedia article section Color_vision#Physiology_of_color_perception, or at least the cone cell response spectra chart there:

Normalized response spectra of human cones to monochromatic spectral stimuli, with wavelength given in nanometers

Cone-fundamentals-with-srgb-spectrum.svg.png (28.87 KiB) Viewed 11951 times

Basically, each cone type (Short, a.k.a. "blue", Medium, a.k.a. "green" and Long, a.k.a. "red") actually responds to a large portion of the visible spectrum that overlaps with the others (e.g. green light is detected to some degree by all photoreceptors, not just the M-cones). So, each cone type will produce its own different signal value which is like a weighed average (more like integral math, actually) of the spectrum of the light received. And then to determine the color at a given point, nerves process signals coming from the nearest S, M and L cones: it is the ratios of the signal values coming from the different cone types that matters. But there are different ways to achieve a given combination of signals.

So display and lighting engineers, as usual, want to have more relaxed requirements when they can, and so they assume they don't need to produce a broad, flat spectrum to produce a white color (like what incandescent bulbs do, which approximate black body radiators). Apparently they assume that as long as the ratio of the S:M:L signals produced is mostly right for most individuals, the spectrum is good enough. One weakness in making this assumption is due to the fact that the response spectra of human cones (see the chart above again) are actually not exactly the same for every individual. There are separate genes that code the S, M and L response spectra, respectively. For each of these genes there are at least a couple of different variants that produce a curve that is slightly offset to the left or right of the other variants. Black body radiation of any temperature is going to be perceived pretty much the same regardless of these genes. However, once you start playing with spiky light spectra you can end up with ones that trigger very different combinations of S:M:L signals between individuals with different cone pigment genes. That's why colors on a monitor can look correct or nice to one individual and incorrect or ugly to another. But I suspect that it also means that, for some individuals some monitors can look unnatural enough to become an irritant, especially when their light comprises most of the light that is coming in. There could be many other things about the physiology that we don't fully understand, and which could be complicating the reaction to unnatural spectra.

So, I'd be researching the light spectra of monitors, as per your leading theory. I'd be comparing both the overall shape and the wavelength of the peak(s) and trying to find something much different from the known bad ones. I don't know which monitor and panel/backlight manufacturers even provide such data, other than cases mentioned earlier in the thread. I guess some products targeted at professionals are more likely to be engineered for a flatter spectrum, and to advertise such a feature, because that must be a way to achieve good color accuracy as perceived by more of the population. Or I'd be looking directly for professional experience or scientific theories linking light spectra to neurological symptoms. I hope you figure out something.

[Tisik]

Did you try a TN screen? I have a Zowie BenQ ZOWIE XL2540K - 240Hz | Full HD | 24,5'' | TN which helped me get rid of any motion sickness or headaches.

[MagnuM]

Thank you for your very detailed reply, MPRT|GTFO! Definitely learned a lot from ya!

My apologies for taking so long to reply myself. Work was starting to get busier, and I was doing some more experimentation and testing, and wanted to wait until I had some results to report back on.

What's interesting about the chart you shared is how there seems to be a triple-low-point around the cyan part of the spectrum around 480 - 490 nm. I notice how most spectrometer graphs of monitors are always very low in this area after the big blue spike finishes. I wonder if there's something to that? I remember reading on a Reddit post once that the color cyan seems to trigger the most amount of migraines in the susceptible (if my memory serves me correctly).

I've always found colors on most LED monitors to be not good-looking at all (especially reds). The icy cold color temperature never helped much when it came to this perception. Maybe it's just the fact that my old Dell U2410 is a wide gamut display that a lot of graphics artists used, and came out late in the CCFL technology cycle.

It would really help to have a color spectrometer to measure the displays I'm using so that I can know exactly what I'm dealing with. I heard they are prohibitively expensive though for non-professionals.

I was lucky enough to be full work-from-home for the grand majority of this decade, but I heard that was likely coming to an end a couple of months ago. I guess I'm still lucky in a way, as they are only going to have us come in once a week. However, this was still enough to cause panic in me, as all they have at work are LED monitors. I was kicking myself for not solving this issue during the 5.5 years I had to control all the variables consistently at home. At least it was enough to galvanize me into trying a new setup that I was hesitating against due to the inconvenience.

A friend helped me lug my 40" CCFL-backlit LCD TV from the basement that I bought in 2008 and am still using today. It's similar to my Dell U2410 computer monitor in that it's a "known good" daily driver that I've been using for a very long time. It was almost 2x the size as my 24" Dell U2410, and I put it on a table at 2x the viewing distance. The increased viewing distance was an interesting test - and one that the "Chief Blur Buster" has on his list of "left field" experiments to try. Looking at a large display 4 feet away instead of a smaller display only 2 feet away I thought would really help my eye converging muscles to get to go on "light duties" for a while.

The TV's resolution was 1920 x 1080, and at 40", it was essentially only half the pixel density as my 24" 1920 x 1200 Dell U2410. I tried it out for work and personal for about 3 consecutive weeks. I was amazed to find that it was essentially about the same as my Dell U2410 when it came to comfort. Working on spreadsheets with white backgrounds didn't seem to be the most comfortable as always. My Dell U2410 is by no means perfect as my daily driver - I still get a lot of eye achiness after long usage or looking at brighter colors on it for a while - but at least I don't get those sharp headaches and neurological symptoms that I do from most LED displays.

There was another odd thing I was subconsciously doing with the TV setup. Since the display was now twice as far away, I felt compelled to sit on the edge of my computer chair, which meant that I had to sit up straight as a board to try to maintain a proper posture. At first my back muscles shook and trembled a lot, as they didn't feel strong enough to maintain this posture. Ever since a teenager, I don't think I was ever able to sit up straight like all of our mothers told us to! It was much too tempting to just round out the back into the computer chair's backrest like a lazy slouch! Being tall at 6' 4", some bad forward head posture developed due to these bad habits. I wonder if that's why I've had chronic tension headaches and a bit of wooziness for nearly 20 years? I know some posters on LEDStrain.org were linking some of their eye strain symptoms to bad neck posture, but I don't know if there is a correlation or not.

After trying the TV setup for about 3 weeks or so, I did fully revert back to my old setup on the Dell U2410. An odd thing followed me though - the compulsion to sit up straight. So perhaps there's a positive change that came out of the TV test. Weirdly enough, I've been pretty good over the past month on the eye symptom front, but that can be due to luck. I've gone long periods of time over the summer without updating my eye strain log in year's past, as the increased daylight and time spent outside I think really makes a difference.

During the one on-site day at work I've had so far, I tried the Dell P2425 they had there with 75% brightness. I was fine in the morning after a few hours, but after lunch, I remember the neurological symptoms suddenly surfacing. It's this bizarre queasiness or sense of dread almost that starts setting in. It's like some sort of core communication channel being setup between your neurological system and "you", and it's saying "whatever you're doing right now buddy, you gotta stop!" After a little walk-around break, I activated Windows Night Light on the O/S, and was weirdly able to finish the day without much of a worsening of symptoms, or the triggering of a migraine.

This is just a crazy thing to try to figure out. I'm coming up to 14 years of first noticing the issue after buying one of the first LED gaming monitors that hit the market in December 2011.

[MagnuM]

Did you try a TN screen? I have a Zowie BenQ ZOWIE XL2540K - 240Hz | Full HD | 24,5'' | TN which helped me get rid of any motion sickness or headaches.

Hi Tisik, thank you for your reply!

No I haven't tried a TN panel paired with an LED-backlit display. My CCFL-backlit LCD monitors from 2005 --> 2013 though were all TN panels if I recall correctly. My issue seems to be so specific to the backlight type rather than the panel type.

I have heard of some people posting on Reddit suggesting eye strain sufferers to try using a TN panel. I don't really know the reason why though. Faster response times? Dimmer display? I heard a lot complain about the colors though when they are used to fancier panels!

[MPRT|GTFO]

I only provided some information and ideas I thought were relevant and missing.

Of course, in this whole case it is for you to do what you need to do when you need to do it, so you don't need to apologize for anything.

What's interesting about the chart you shared is how there seems to be a triple-low-point around the cyan part of the spectrum around 480 - 490 nm. I notice how most spectrometer graphs of monitors are always very low in this area after the big blue spike finishes. I wonder if there's something to that? I remember reading on a Reddit post once that the color cyan seems to trigger the most amount of migraines in the susceptible (if my memory serves me correctly).

Display engineers must be viewing light in that spectrum in between as just a sort of waste, as it doesn't help get higher saturation of the primary colors, while being less energy efficient (I guess adding less perceived brightness than other wavelengths). Interestingly though, that's exactly where the sensitivity peaks of other types of photoreceptor cells are (but those cells do not seem to directly play an important role for color vision).

Overview_of_the_retina_photoreceptors_(b).png (216.18 KiB) Viewed 162 times

If you're curious, look up "Intrinsically photosensitive retinal ganglion cell" on Wikipedia. Their job is to influence the sleep-wake cycle, the pupillary light reflex, and who knows what else. I wonder if the unusual situation of all the color receptors (especially the "blue"/S cones) being stimulated strongly while the ipRGCs being stimulated much less (due to the dip in the spectrum) may have some subtle, or in some cases not so subtle effect on the human body. Like, the pupils may not be constricting sufficiently in response to mostly LED monitor light, because they are not getting the signal for that, leading to too much light from the LED hitting the retina. I honestly wouldn't put it past display manufacturers to intentionally try to trick human vision into seeing their displays brighter, just for marketing purposes, and ignoring health consequences.

If this idea about the difference between the stimulation of the cone cells and the other two types of cells has any merit, then I think it should be possible to compensate for the dip in the spectrum of a LED monitor by adding some more light of that wavelength in your environment. My guess is that it shouldn't matter where on the retina it hits - it should be about the balance between the total quantity of light around 480nm hitting the retina vs. the total quantity at other wavelengths. The idea here would be to try to bring the overall light spectrum closer to natural daylight.

On the other hand, it may turn out that an even deeper and wider dip around 480nm may be what would be better for you instead. There's this sentence at the end of the article: "Intrinsically photosensitive RGCs have also been implicated in the exacerbation of headache by light during migraine attacks." along with a link to some research paper about it. This does seem to confirm what you remember about a link between cyan and migraines. This hypothesis that you might need less cyan is not supported by the fact that you have a normal response to daylight, but during the night it may well turn out to be even more important than the color temperature per se (though reducing the color temperature will still reduce ipRGC response even if your monitor had a deep and wide dip around 480nm, because the ipRGC spectral sensitivity inevitably overlaps with the spectrum that is needed for reproducing green, at least).

Ever since a teenager, I don't think I was ever able to sit up straight like all of our mothers told us to! It was much too tempting to just round out the back into the computer chair's backrest like a lazy slouch! Being tall at 6' 4", some bad forward head posture developed due to these bad habits. I wonder if that's why I've had chronic tension headaches and a bit of wooziness for nearly 20 years? I know some posters on [...] were linking some of their eye strain symptoms to bad neck posture, but I don't know if there is a correlation or not.

To me it doesn't sound far fetched at all that there could be links between posture or backbone issues and headaches or other neurological symptoms, as I've heard of similar things too.

Weirdly enough, I've been pretty good over the past month on the eye symptom front, but that can be due to luck. I've gone long periods of time over the summer without updating my eye strain log in year's past, as the increased daylight and time spent outside I think really makes a difference.

This (as well as my earlier idea that the 480nm dip may have a negative side) makes me think you should consider room lighting again. Maybe augmenting your daylight to be more like a summer day outdoors could be beneficial? I noticed that Chief Blur Buster has given some good and thorough advice on the subject of getting natural-like light, which I suppose you've read, but if it was long ago maybe you want to revisit it here and in his follow-up comments: viewtopic.php?p=40981#p40981 I think it may be a good idea to have some high CRI (so, plenty of 480nm) dimmable light source around and test to see if there's a good brightness balance to find between a screen and the light, or if it's really better to minimize 480nm everywhere and all the time (again, hard to believe that 480nm that is also in daylight is bad per se).
Or maybe your vitamin D levels go low after summer (extremely common in most populations, apparently) and maybe that could play some unexpected role. These are the possible seasonal links I can think of.

I heard a lot complain about the colors though when they are used to fancier panels!

Just to mention that some of the newer TN panels I've read about use quantum dots, which has supposedly improved their colors a lot compared to older TN panels.