These are good points but miss an essential part of the story: reflected light almost by definition matches the brightness of the surrounding environment, meaning that the periphery is roughly the same brightness and white balance as the reading medium.
If I look at my laptop on full brightness in a nearly pitch black room, it is far less comfortable to read than if I turned on a lamp.
So there are two solutions to this. Turn on a lamp if you're using a screen at night time, and lower your brightness to have roughly the same level as the ambient light.
Or, use night mode if you don't want to turn a lamp on.
If I'm doing something like programming, I prefer the former. If I'm just in bed and feeling lazy and want to read twitter, I go for the latter so I don't have to turn a lamp on.
Yes, it probably correlates with ambient light. I like light mode but I also like to open the blinds, turn on the lamps etc, as opposed to the stereotypical dark basement hacking with a hoodie on.
Dark mode is also very often buggy, clumsy, the color schemes for syntax highlight are subjectively less pleasing, it feels gloomier, less cheerful, I see more distracting dust on the screen, smudges etc. Yeah I could also clean it more etc. The point is there can be multiple equilibra, one develops habits and preferences. It's not an absolute thing.
I use gruvbox, light when there is a lot of light around (e.g. im outside) and dark mode when im indoors.
There is more to this than all the biological talk too.
I use it as a context switch and prefer to review code in light mode, while i prefer to write code in dark mode.
Either it helps with the context switching or reading is just easier in light mode, while writing is easier in dark mode because its easier focus in on what im writing.
I think given a lot of people prefer dark mode for various things, people should assume they lack all the information for making bold claims based on biological studies alone.
Also I'm becoming more and more convinced that we humans are extremely adaptive. With regards to major and minor things as well. People can adapt to eating 6 small meals a day and then swear they would not be able to concentrate after skipping one of them. Others eat one meal a day, exercise fasted and skip entire days of food and report very good mental clarity and love it. Some only eat vegan and with a few supplements live very well. Other eat purely carnivore (exclusively salt, meat and water) and report good results after an initial phase of toilet hell.
Some people never exercise and live to 90 years. Some people are night owls and do their best work at 3 AM and have no concept of a schedule but get stuff done brilliantly. Others have to dogmatically stick to a schedule and a rigid morning and evening routine and swear by waking up at 4:30.
Partially it's their genetics but a huge part is habits and adaptation. Humans can function in the polar circle, in the jungle, on the savanna, in urban hell, in the suburbs, on farms etc.
Any study that wants to test the difference in benefits of different habits would need to allow for the adaptations to take place, the adjacent habits to adjust to the new style etc. You cannot just take one thing in isolation. Maybe I like light color schemes because I do most of my work in the daylight and sleep at night. There can be tons of confounders and doing months long randomized studies is rare.
As a photographer, as well as a Technical PM who spends all their time on a computer, and a non-zero amount of time optimizing my setup, and someone who has taken photos of computers and done exactly this to get the exposure correct: I don't know why I didn't think of doing this on a daily basis!
Every time I bought a new computer monitor in the last 10+ years, I had to turn the brightness way down in order to use it comfortably. I've gone as far down as 8% with some of the brightest monitors, and even with more reasonable ones I rarely go above 40%. Maybe these things are designed to be readable outdoors in direct sunlight. They're absolutely not suitable for indoors use, especially at night.
If I look at my laptop on full brightness in a nearly pitch black room, it is far less comfortable to read than if I turned on a lamp.
So there are two solutions to this. Turn on a lamp if you're using a screen at night time, and lower your brightness to have roughly the same level as the ambient light.
Or, use night mode if you don't want to turn a lamp on.
If I'm doing something like programming, I prefer the former. If I'm just in bed and feeling lazy and want to read twitter, I go for the latter so I don't have to turn a lamp on.