Gemsand
I'm sorry I don't have direct experience, but I think I still have some useful points to make:
CRTs used to have multiple refresh rate settings, many people found that the lower refresh rates caused them eyestrain but the higher ones were acceptable. These were around 50/60hz on the lower end, and 75-80hz on the higher end.
I've read papers (though I can't find them at the moment) that demonstrate that most peoples upper limit for responding to strobing light is between 1,000 and 10,000hz - these papers looked at the brain's response to visual stimulus, rather than participants being consciously aware of the strobing light. This is for strobing light however, so would apply more to backlight PWM than dithering.
This paper here is worth a read https://www.nature.com/articles/srep07861 but, its looking at perception of dithering, rather than eye strain.
Something else to bare in mind is that temporal dithering doesn't occur every frame. If you have a monitor that can display a light blue, and a dark blue, but it needs to display a medium light blue it might do something like this:
Dark blue -> (next frame) -> light blue -> (next frame) -> light blue -> (next frame) -> light blue -> (next frame) -> Dark blue
That's 3 frames of light blue for every dark blue frame, so a 480hz monitor would be flickering at 120hz. And that's just one example, what if it was 9 frames of light blue for every dark blue, then it'll be flickering at 48hz, and so on. This means there is a range of flickering frequencies due to temporal dithering between 0hz and half the max refresh rate of the monitor. A higher refresh rate monitor will raise all those frequencies, but some lower perceptible and problematic frequencies will always be present.
So in short, I have no idea if a higher refresh rate will help. If they weren't so expensive it'd make an interesting experiment.