45

I was discussing with a colleague about using dark-mode vs. light mode and remembered an article arguing that humans vision is more adapted to light-mode rather than dark-mode:

I know that the trend “du jour” is to have a dark mode for pretty much everything, but look around the world is not dark with a bit of light, it’s actually the contrary. And as the human has evolved its vision to adapt to this reality, it’s asking extra efforts on many people.

Unfortunately, no reference is provided to support this claim, so I am wondering if this is just an opinion or there are some studies to support this.

Wikipedia seems to confirm this somewhat since we are adapting much faster to "light mode" transition than to dark mode one:

This adaptation period is different between rod and cone cells and results from the regeneration of photopigments to increase retinal sensitivity. Light adaptation, in contrast, works very quickly, within seconds.

Also, some studies confirm that working using light mode is on average more efficient than using dark mode:

light mode won across all dimensions: irrespective of age, the positive contrast polarity was better for both visual-acuity tasks and for proofreading tasks.

I am looking for arguments coming from evolutionary biology to confirm (or not) the assumption that human evolution favors light mode.

AliceD
  • 52,402
  • 19
  • 174
  • 238
Alexei
  • 1,665
  • 3
  • 11
  • 24
  • 6
    Meh, this evo thing is probably the wrong angle here. If you're looking at something dark in an otherwise fairly bright (room) environment, it is actually tiresome. – the gods from engineering Jan 04 '21 at 12:03
  • 4
    As an aside, you're less sensitive to flicker in a dark environment. https://www.sciencedirect.com/science/article/pii/B9780123742032002359 (see full text) So if that bothers or tires you (which is plausible with screens), then dark mode in a dark environment does help in that regard. – the gods from engineering Jan 04 '21 at 12:19
  • 5
    @Fizz: Yes, the world does not have a raster scan running through it :-( Not so bad with LCD/LED displays as it was with CRTs (which would give me splitting headaches if I worked for a prolonged period in "light mode". But the evolutionary biology arguments are beside the point: it is observed fact that many people do prefer dark mode. – jamesqf Jan 04 '21 at 17:48
  • Challenge: "Adapted to" is insufficiently specified. Visual resolution will be higher with a brighter display due to pupil constriction, but night vision will be impaired. – chrylis -cautiouslyoptimistic- Jan 04 '21 at 21:55
  • 3
    The "some studies" are wrong. Astronomical citings routinely have greater precision than the lens rules calculated in bright environments say they should have. Optical resolution simply is better with withe on black. – Joshua Jan 04 '21 at 22:49
  • 1
    dark mode looks better to me at night, and light mode during the day, so... – user253751 Jan 05 '21 at 16:33
  • 2
    Even if the article’s conjectures were to hold, they would do so for quite other reasons than that “the world is not dark with a bit of light”. Yes; and the world is also experienced predominantly as indirect reflected light rather than direct projected light. Conditions rightly described as “optimal” in one environment (the great outdoors) may legitimatey differ in another (my home study). Even if my study is objectively sub-optimal, it may be beyond my means/power to make it so; thus, in the dark vs. light debate, one does not take all — providing both is important for accessibilty. – Mark G. Jan 05 '21 at 17:46
  • 1
    Most of us programmers who practically spend our life behind monitors, we prefer dark mode right from the start. It's better for the eyes, it doesn't take a lot of research, a white background is simply annoying. You can't focus, you turn on every light source in the room to help your eyes but still, white screen is disturbing and your eyes get tired a lot sooner. Dark mode is there for a reason. We've been using it since the age of Unix and MS-DOS, it's nothing new, it's only reinvented, as in, finally everyone else realized it too. It was a long lasting fight to convince people that no matte – dkellner Jan 07 '21 at 10:04
  • "we are adapting much faster to "light mode" transition than to dark mode one" Of course: It's better to not see anything for 10 seconds than to have pain in the eyes for 10 seconds. – Fabian Röling Feb 03 '21 at 19:20
  • 1
    We should not forget that not all humas are equal. People from the autism spectrum (about 1% of all humans) tend to have a sensory processing disorder which can mean that bright light appears unpleasant. So even if light mode would be the better choice for most, it is not necessarily the best choice for all. – xystum Feb 05 '21 at 10:26
  • 1
    Found myself revisiting this in the wake of 2 developments: 1. I learned those w/ astigmatism can find dark-modes unusable due to the ‘starburst effect’, which light-modes don’t provoke as much due to diffs in pupil dilation. 2. My own eyes now hate light-colored backgrounds (especially on IPS displays). Bloodshot within minutes. No idea why; it’s never been an issue before! Since I’m now experiencing physical discomfort w/ lighter colors, I’m using dark-modes for respite… I stand by my earlier claim: it’s an accessibility issue, not mere style preference. One size does not fit all. – Mark G. Oct 12 '23 at 01:39

3 Answers3

30

A question that requires quite a lot of guts to ask on this site :) Nonetheless, and risking sparking a debate, there are a few arguments that spring to (my!) mind that can support the notion that we thrive better in 'day mode' (i.e., photopic conditions).

Literature
- Bompas et al., Iperception (2013); 4(2): 84–94
- Kolb, Webvision - The Organization of the Retina and Visual System (2012), Moran Eye Center

Further reading
- Why does a light object appear lighter in your peripheral vision when it's dark?

AliceD
  • 52,402
  • 19
  • 174
  • 238
  • 5
    The di/trichromat approach really isn't all that relevant to computer displays, unless you're really into syntax colorization. (Something I detest, but tastes differ.) What matters is contrast between light and dark, and you could probably have readable displays at intensity levels where only the rods are activated. The example I like to use is stars on a dark sky. Tt's easy to see individual stars, or lighted pixels, on a black background, not so easy to pick out a few dark pixels on a white display. – jamesqf Jan 04 '21 at 22:28
  • 7
    This answer gives some interesting thought directions, but they don't really have anything to do with dark mode vs light mode. Dark mode isn't dark in the sense that the range of brightness is lower, but in that the bright features are smaller. In practice, it is perfectly possible to use dark mode in photopic vision. Vice versa it is even possible to use light mode in scotopic vision, though that requires turning the total luminosity way down (which I'm pretty sure is worse for the eyes than scotopic dark mode). – leftaroundabout Jan 04 '21 at 22:41
  • @jamesqf, I agree, but the fact we can operate well in the dark doesn't take away the fact we do better in daylight overall. In the end, evolution will not have favored spotting stars in the sky, or pixels on a dark screen. I chose to approach the question from a scientifically more likely angle. – AliceD Jan 04 '21 at 23:47
  • 1
    @leftaroundabout I personally want my screen to be generally similar to ambient brightness, easier to ignore surroundings (= light mode during day and dark mode during night with lights off). From this answer we should preferably be in a well-lit environment. Then the question becomes - should the features be even brighter, or darker? Tech to make them darker is trivial (just not turn those pixels on) while not many screens let you make them even brighter - though it would be interesting to see what is better in that case. – Zizy Archer Jan 05 '21 at 08:26
  • 1
    @jamesqf the brightest stars are seen with cones. This is easy to understand if you remember that you can see e.g. Sirius looking at it directly, with your fovea. OTOH, it takes significantly more effort to view very dim stars (that can stimulate only rods): one needs to use the averted vision technique, which not only gives smaller resolution, but also is a bit uncomfortable: one has to train to avoid the temptation to look at the object directly, since in that case it will become invisible. – Ruslan Jan 05 '21 at 12:52
  • @AliceD I'm sure spotting night predators in the dark would have been quite important to early apes and humans. Cats and wolves are very easy to spot for humans in the dark. It's also handy for things like finding water, and in general, specular reflections are a typical example of light-on-dark objects of interest. And of course, there's considerable variation in what different people find comfortable or efficient; I feel people overly focus on finding the "one best thing". If dark-on-light works for you, great; if it doesn't, also great. It's not as if there isn't choice. – Luaan Jan 05 '21 at 14:32
  • 1
    @AliceD: As usual, the problem with saying "we do better" is that we differ. The demonstrated fact that a significant fraction of computer users prefer dark mode demonstrates that, I think. And I would bet that a good many of those who say they prefer light mode have never given dark mode a real chance :-) – jamesqf Jan 05 '21 at 17:08
  • @jamesqf I personally highly prefer dark mode as it is more relaxing to the eyes and mind. But the visual system hasn't evolved for 10-hours-per-day-screenwork. I guess that's the point of my post. I've never attempted to defend 'use light mode when behind the screen'. I've just attempted to write an answer that highlights the evolutionary backgrounds of the visual system, without stepping into an opinion that cannot be backed up by literature. – AliceD Jan 05 '21 at 20:27
  • 7
    Humans are thought to have evolved from apes... replace this with Humans share a common ancestor with apes, and.... Our evolutionary history is more fleshed out than just a "thought" – Isaac Jan 06 '21 at 00:00
  • @jamesqf I don't think it's clear-cut. I've always found that light-on-dark themes cause me more eyestrain when dealing with text. – ssokolow Jan 06 '21 at 16:03
  • @ssokolow: As I said, people differ, and one size, or one color scheme, obviously doesn't fit everyone. – jamesqf Jan 07 '21 at 17:49
  • @Isaac and more recently, "Apes" is used as a monophyletic group, including humans and other hominids. So Humans are apes. – Colombo Feb 04 '21 at 23:18
  • Scotopic periphery vs. photopic fovea seems like it may correspond to my preference for "dark lite" themes, where only the UI elements are dark, while content areas remain light. – SubGothius Feb 05 '21 at 00:36
10

From experiments it seems you should better use dark mode if you want to prevent myopia:

Using optical coherence tomography (OCT) in young human subjects, we found that the choroid, the heavily perfused layer behind the retina in the eye, becomes about 16 µm thinner in only one hour when subjects read black text on white background but about 10 µm thicker when they read white text from black background. ... Therefore, reading white text from a black screen or tablet may be a way to inhibit myopia, while conventional black text on white background may stimulate myopia.

So although humans are commonly awake during the day and sleep at night it seems that this fact is actually irrelevant when it comes to the effects of dark/light mode for the health of your eyes.

asmaier
  • 201
  • 1
  • 2
  • This makes sense. I've been using white text on dark backgroun since the early nineties - not because it was something the cool kids did (none of them did it back then :) ) but because when most pixels on your screen are at RGB 255,255,255 - i.e. full intensity white - it's like staring into a floodlight all day - something I didn't think was a healthy option. A few people asked why my screen looks weird and I explained my eyes never get tired this way and some switched to dark backgrounds on their PCs and were amazed at the difference to their eye fatigue. – Volksman Feb 05 '21 at 07:07
7

Both.

More specifically, we're a kitbashed light-mode version of a primarily dark-mode group of animals. Mammals famously have terrible color vision, most mammals are dichromats (red color-blind) whereas most other animals like fishes, amphibians, reptiles, and birds are trichromats or even tetrachromats (seeing into the ultraviolet spectrum). In their evolution mammals got rid of a lot of their cones and replaced them with rods, giving them much better dark vision at the expense of not being able to see good colors. But this was back in the Mesozoic, when most mammals were nocturnal, crespuscular, or burrowing, and color vision wasn't a huge deal.

Primates then re-evolved color vision by duplicating and repurposing a cone for green some time in the Eocene. It is thought that they did this because it helps in identifying ripe fruits, which are often some shade of red, orange, or yellow, from green unripe fruit. We also lost our tapetum lucidum, which is the reflective layer of the eye that you see when shining a light in most animal's eyes that greatly enhances night-time vision. No one knows why, it may be genetic drift. Unfortunately our color vision isn't great, so we're kind of stuck in this awkward intermediate between other mammals, which can see great in the dark but have no color vision, and birds and reptiles, which have amazing color vision but are borderline blind in the dark (owls being the obvious exception).

user2352714
  • 410
  • 2
  • 5