Set up some experiments using http://www.quantified-mind.com/ to see whether there's any difference in your cognitive performance when you've been sleeping for 6 hours vs. for 8. Practice the tests a few times, then start a custom experiment with your sleep variable in there. Alternate back and forth between each, maybe a week at a time, for a few weeks–test at the same time of day each day. Hit the Data Minding and see whether there's any perceptible difference in performance.

(Disclaimer: I wanted this to exist, so I helped Yoni build it.)