Usually, in our lab and researching in free living conditions human circadian rhythms, we work on a complete week because we think that it is representative.

So, my response would be as many cycles as you need for a representative sample.

Normally, in our case, we have to choice between reliability and subjects' comfort.

Goog luck

Antonio

Thank you

I think this depends on the sample size which can be either single time points per cycle or tested individuals per time point. However, if you use only one individual and measure e.g. 1000 data points of only one cycle, you will get an exact interpretation of this particular cycle but it will not give you a reliable conclusion or evidence that all other cycles of this individual or even of other individuals look the same.

thank you; this is what I though but I am not a specialist at all. Your answer and Antonio's answer confirm my thoughts

Here is a paper by Cornelissen. I hope it helps!

Thank you for the paper