JD Moyer blogged about a simple sleep cycle experiment: no artificial light between sunset and sunrise. I read about this a couple weeks ago and immediately put it into action, albeit in an incomplete way (I would do a few hours of limited artificial light after sunset). It made me more energized and rested, and had me waking up much earlier, but I've scaled it back for two reasons. First, it's simply very different from the lifestyle I'm accustomed to; second, I have an acting class every Thursday night, which usually runs to 11pm or later, and a hacker meetup Tuesday nights which goes to 10pm. Both events had me experiencing artificial light until well into the night.
I think this produced a destabilizing effect. Certainly, my diurnal rhythms have varied in some fairly unprecedented ways over the past couple weeks since I first began the experiment. This morning, for example, I woke up at 3:15AM fully rested and energized; last night, I was completely exhausted at 9:30PM and fast asleep soon after. I went back to sleep again this morning, and stayed in bed til around 6:30 or 7, not because I was tired, but simply because without any artificial light, there just wasn't that much else to do. This happened to me at least a few times.
This experiment is a little too much for me as a permanent lifestyle choice, but it's proven incredibly effective, so I'll probably do another, similar short run with it at some later date. I suspect the highly artificial nature of "normal" Western sleep patterns degrades health, so my thinking is I'll use this as a kind of restorative technique to limit that damage.