If you don't feel well you likely aren't going to perform well. If your eyes hurt, you are going to want to stop working. If I had to stare at a bright white screen all day, I wouldn't get the task done because it hurts my eyes. Common sense.
Yet another study where it's trying to trump opinion with a study. You can't be wrong if you prefer dark mode/light mode, it's subjective. A study is going to have a real hard time proving one is "better" than the other and we certainly shouldn't take the results on faith.
I found where blue light causes eye strain, and it most certainly does, I immediately notice the difference when using a blue light filter on my monitor, but you can have both dark mode and a blue light filter enabled.
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6020759/
Anecdotally, dark mode causes my eyes and my partner's eyes less strain.
You can notice by how dry and red they get, as well as the headaches, it's not rocket science.
EDIT (reached my post limit)
@shawnz: No, the light physically hurts. It's immediate. It's not because of the aesthetics.
I don't think this is "rocket science" but that doesn't mean it is handwavingly obvious either
Having said that I think my preferences are mainly, if not entirely, aesthetic. Dark-on-light terminals feel weird as do light-on-dark documents/spreadsheets - that can't be anything other than just what I'm used to.
Better?
If you want to truly understand how you specifically react to something, then do science yourself. You don't need to be very rigorous with this, as the goal isn't to prove some huge effect size. The goal is to just get an objective answer to the question of whether you are more productive, rather than a subjective (and therefore very likely wrong) one. Find an objective measure of your own personal performance (anything you care about, doesn't have to be something a boss would care about); maybe find a way to blind yourself to which days you're using the treatment vs. a placebo; write down numbers; and then stick 'em in Excel and see what you get.
I would point at Gwern as a good example of someone who runs self-experiments to determine the objective n=1 effects of different "life hacks" on things they care about. (For Gwern, this is usually brain performance—measured by, among other things, the https://en.wikipedia.org/wiki/N-back task, data for which can be collected in the form of "playing" a quick round of a computer game.) For example, here's their look into magnesium supplementation: https://www.gwern.net/nootropics/Magnesium
The placebo effect is fairly strong. And how productive someone feels vs how productive they actually are when it's measured often varies, of course.
I value my feelings in a pleasant work environment a hell of a lot more than productivity metrics.
Of course. All you are measuring is how comfortable you feel, which is irrelevant. What they are measuring is how well people actually perform on tasks while using dark mode. Just because dark mode feels better to you doesn't imply anything about the latter question.