AGI I believe there is an intuitive link between illuminance and color temperature (lower CCT is preferred by most people when it is dimmer, higher CCT when it is brighter), but this link may solely be preference as someone who grew up with incandescent and halogen lighting that "warm dims." Basically no other light source does this natively.
I'd be curious if those fluorescent lights weren't flickering a lot when dimmed, especially if there was a mismatch between the ballast and dimmer electronics. Fluorescent and HID sources that were color shifted were typically malfunctioning, so that may have something to do with it. I was in an H&M store the other day that had a bunch of HID lights that had shifted to green, and I see it on car headlights fairly regularly as well. Fluorescent tends to shift to pink when malfunctioning. I'd also be curious if you ever had the opportunity to test a low-CCT fluorescent (like 3500K) at full output.
There may be something to noticing the problematic aspects of lighting at lower illuminances. I tend to notice flicker when my home lighting (mostly Philips Hue) and computer displays are at a lower illuminances, but this is bearing in mind that the flicker tends to be worse to achieve the dimming in the first place. With that said, I can see really bad flicker in some very bright light sources, such as many "vintage" LED bulbs.
The benefit of having this discussion today is that we have very powerful tools for understanding color, such as TM-30-18. But the challenge is we have LEDs, which can vary widely in spectral output, even among sources with the same CCT. Because both the tools and the light source are so new, we are really just at the beginning of understanding the relationship between illuminance and color, if any, beyond preference. I think we'll have a better understanding in the coming years as more research is done with these new tools.