[runeks]

> I wonder if 8h of sleep is the cause of the better performance or they both are a result of less stressful life, better health or something else.

This illustrates well the problem with these kinds of studies. How do we know that stress isn't causing both little/poor sleep and e.g. decreased cognitive performance (as opposed to little sleep being the cause of a decrease in cognitive performance)?

It's not like we can create a control group where stressed people are forced to get 8 hours of good sleep every night.

[blackbrokkoli]

We can actually. That is what sleep labs are for. I don't know if this exact setup has been done, but there is a lot of stuff like making people cram for an exam and then varying their sleep opportunity.

If you mean that you cannot literally make people sleep: No, but you can manipulate consistently by just setting sleep opportunity and then measure the actual sleep time on top of that to draw conclusions.