Now I recently came across this interesting paper
- Circadian Timing of REM Sleep Is Coupled to an Oscillator within the Dorsomedial Suprachiasmatic Nucleus
By Michael L. Lee, Beryl E. Swanson and Horacio O. de la Iglesia
Current Biology, 19, 1-5 (2009)
presenting the results of experiments on rat that were subject to an increased daylight period, similar in duration to that on a flight from Central Europe to the East Coast. The rats had previously been on an artificial day-night cycle with 12 hours light and 12 hours darkness. Then their daylight period was increased by 6 hours, followed by a 12 hour night, another 12 hour day and then complete darkness. Among other things, the rats were implanted with EEG electrodes recording their brain activity.
The authors study two different sleep phases in this experiment: the so-called Slow-Wave-Sleep and the REM (rapid eye movement sleep). They can be distinguished by brain activity. The sleeping pattern is regulated by a brain region called the suprachiasmatic nucleus responsible for the roughly 24 hour rhythm of living organisms. It consists of two areas, the ventrolateral and dorsomedial part. The ventrolateral part obtains light information from the eye, the dorsomedial part doesn't. Exposing the rats to a longer period of daylight leads to an "enforced desynchronization" of these two regions.
The measurements of brain activity show that while the the slow-wave-sleep immediately adjusts after the delay of the next night phase, the REM sleep needed several days to come back in synch with the slow-wave-sleep, indicating that the former is regulated by the ventrolateral, the latter by the dorsomedical suprachiasmic nucleus. In the meantime, the sleep architecture of the rats was messed up.
Reading the paper I had been wondering why, after increasing the light-period, the day-night cycle was not continued but instead the rats were put into total darkness while measurements were continued. Horacio de la Iglesia, one of the authors of the paper, was so kind to answer my question and explained that previous research had shown that in such a way the dysynchronization continues to 6-7 days, whereas with a continued day-night cycle the readjustment might happen faster.
I find this study really interesting. Of course there are many reasons why this experiment on rats doesn't too accurately describe what happens in the human brain when jetlagged, but I think chances are good the basic insights about the messed-up sleep patterns are transferable. There are many further questions that spring into mind, for example one would like to know if the desynchronization lasts longer if instead of increasing the daylight period, the nighttime period is decreased. That's because I find it typically much harder to adjust after an eastward than after a westward flight. After reading the paper I believe it might be because in this case I tend to sleep into the day too long, missing many hours of sunlight which delays adjustment further. One would also like to know how the time needed for adjustment relates to the phase shift in the day-night cycle.
Something else I learned from this paper is that "sacrificing" (in this case the rat) is an euphemism for "chopping off their head and removing their brains".
As the DPG spring meetings are traditionally well attended by young students, dominantly those carrying a Y-chromosome, the lecture hall was completely crowded when 
