Check out this excellent blog post and test scenario by Kevin Willmorth, owner of Wisconsin-based Lumenique.
I agree with most of his conclusions in his post, but with two comments:
First, I hope to see other optics laboratories replicate the results, to confirm the science. Always good practice!
Second, I will also bring up a counter-point about office environments that I believe is missing entirely from modern lighting design: What is the comparative eye strain relationship of modern LCD screens to various levels of ambient or task lighting? As I wrote two years ago, I personally have suffered from significant eye strain in office environments illuminated to such a high level, they “blow out” the light levels of LCD screens — to the point I actually went to an eye doctor to check my vision, which turned out not to be the problem.
No modern discussion of office lighting can be taken seriously without first considering the output ranges of modern LCD screens (older fluorescent LCD, modern LED LCD, and both glossy and matte screen types) and the relative impact of “glare” produced by the contrast ratios to the ambient environment (NOT just “veiling glare” on old CRT monitors).
I would love to see a professional optics lab conduct some experiments on modern displays especially in open office configurations! If anybody knows of any such contemporary research, let me know.
For posterity I captured a PDF of Kevin’s piece here: