Shifts my sleep forward by half an hour. (I’m a heavy computer user.)
Sounds like Redshift is more of a subtle reminder to go to bed, than something that affects sleep more directly?
How about comparing nights on which you went to bed at comparable times, after staring at the screen for similar amounts of time, with and without Redshift?
At least if one is already using melatonin like I do. Probably the effects would be more dramatic if one wasn’t taking any other pro-melatonin interventions.
I’m not sure that makes sense? If one is going to bed at similar times on a Redshift-less night and a Redshift-ful night, then clearly something odd is happening on the Redshift-ful night to equalize it: I know Redshift causally affects my sleep, I just did the randomized experiment showing it. If you have a specific analysis in mind, I provide my sleep data so you can run it yourself.
How do you decide when to go to bed? Only when you reach a certain level of tiredness?
I decide when to go to bed like most people: I feel tired, it seems like a good time to stop, or it strikes me that it’s late and I should go to bed. There’s no real system to it although I do try to go to bed before midnight unless I’m compelled by something.
So you’re less likely to feel compelled to stay up past midnight when using Redshift, but does Redshift have an influence on how you fast you fall asleep (or some other sleep-quality-related metric)?
Right.
That would be covered by the other variables I looked at. For example, ‘Time to Z’ fell by 1.88 minutes with Redshift but my mean is ~24 minutes so it didn’t reach statistical-significance with the 500 or so days I randomized (p=0.109).
For the most part, the other sleep variables didn’t change much, hence I interpret Redshift (on top of melatonin) as changing timing rather than quality. I see Redshift as allowing the melatonin cycle to begin earlier at a more normal time, but once it’s started then in my particular environment, it has only small additional effects.