Variation of length of circadian rhythm in humans in absense of time reference

Question

Are there any studies which show the range of the length of circadian rhythms for people that are removed from actual time, (Eg no sunset/ sunrise, no clocks or other reference of time)?

For example, what proportion of people tend to have a natural 30 hour rhythm or greater?

I ask because I would like to know if it is a possible cause of insomnia, how likely insomnia is caused by this, and if this knowledge might in some way lead to solutions for insomnia

Answer 1

This might be the study to which you refer in your answer, the people tested averaged a daily cycle of 24 hours and 11 minutes with a surprisingly small variation in their rhythms:

The variation between our subjects, with a 95 percent level of confidence, was no more than plus or minus 16 minutes, a remarkably small range.

Worth noting they were all healthy and so it doesn't tell us much about insomnia.

This study about insomnia in adults with ADHD shows that circadian rhythms of insomnia sufferers aren't so much longer as delayed (for sleep-onset insomnia anyway). Figure 2 might be of interest, showing that insomnia sufferers have significantly delayed melatonin production.

Answer 2

Apparently, in one(?) study, it was shown that the natural period for humans is actually a bit greater than 24 hours.

I only know this by what Jessa Gamble said in her TED talk:

http://www.ted.com/talks/jessa_gamble_how_to_sleep?

She goes into a lot more detail, and apparently we wouldn't just have one continuous sleep each day, so that makes answering the part of the question relating to length of sleep more complicated.

But interestingly, she gets to the core of how sense of reference disturbs sleep. To quote:

The people in these studies report feeling so awake during the daytime that they realise they're experiencing true wakefulness for the first time in their lives.

Unfortunately it is sparse of detail, short, and lacks any references!