This is a guest post by Jason Buberel, an engineer, mountain biker, and Zeo customer who keeps a blog of tips, tricks and tools that he has found interesting or helpful. When we came across this post on Jason’s blog, we knew that it would be great for sharing and Jason agreed. Find the original blog post and other notes on how Jason has been using Zeo at blog.buberel.org.
—
With a solid month of sleep data now captured (using my Zeo Sleep Coach), I’ve been able to tease out an important contributor to how rested I feel when I wake up in the morning.
I had initially assumed that my body would allow me to sleep enough hours each night in order for it to fully recuperate. In other words, if I go to bed late I figured that my body would just sleep in a bit longer and wake me up a bit later. But based on the recordings from my Zeo, this turned out to not be the case at all. In fact, no matter how late I went to bed, my body would just ‘wake up’ at right around 6am the next morning.
So even if I went to bed very late – 2am for example – I would still wake up around 6am. Of course, I didn’t feel well rested, but I was never able to go back to sleep. Once the eyes were open, they stayed that way.
That told me that if I wanted to feel more rested in the mornings, that my only realistic option was to go to bed earlier. So that is what I did – adjusted by bedtime to 10pm, no exceptions. Sure enough, I still woke up just about 6am, but I felt much better rested.
The results of adjusting my bed time to 10pm are shown in the chart below. I made the shift on Thursday, 8/27. Notice how my total sleep time jumped by over an hour? I could definitely feel the difference:
Screenshot of Jason's Trends Tool from myZeo
A bit of research into why this might be the case lead me to this article: Differential rescue of light- and food-entrainable circadian rhythms. In the study, the scientists show how circadian rhythms are strongly influenced by feeding/meal times. As it turns out, I eat dinner very consistently at 6pm each night. So my speculation is that my circadian clock says “Jason – you will wake up 12 hours after your last meal, no matter how much time you spent sleeping.”
Although I’m not sure meal time dictates my wakeup time completely, it sure does seem to be a strongly contributing factor.
{ 1 trackback }
{ 5 comments… read them below or add one }
Wow — great insight!!! Thanks for sharing!
This again highlights the usefulness of a couple of things. First, users need to be able to enter fairly precisely their own variables. I usually eat at the same time, but not always. How does that affect my sleep? If I had a way to enter my dinner time, I’d be able to find out. Secondly, it points out the importance of statistical testing. There is variability in my dinner time; there is variability in my Z score. Relating these two factors requires the use of statistical testing. I don’t see that it needs to be that complicated or intimidating. Where you have a 2 X 2 chart showing, say, weekend vs. weekdays and Z-score, and you see a difference, you can just have an indication “This is a statistically significant difference” or “This is not a significantly different difference.” There could be more info if people wanted (degree of confidence, etc.), but I think most Zeo users are aware of these concepts.
The ability to enter a few variables such as eating times could add more finesse to our sleep patterns. This would also add value to the Zeo system since Users would receive more robust graphs showing relationships or not.
I would add my voice to the above in wishing for a way to enter different information on a nightly basis that has a far better chance of explaining variability in my sleep patterns than the standard questions we are expected to answer. It is wearisome to repeatedly tell myZeo that I don’t drink coffee, much less 3 hours before bed!
What Jason’s “discovery” and his sharing of it illustrate is the potential here for a small community of us to share research about our own sleep patterns. Are there others out there with hypotheses they are looking into, statistically or otherwise? Let’s make a list of topics we have at least thought of looking at with our own data. Perhaps a few will attract the attention of others. No actual sharing of data mind you, just hunches to patterns to open questions to ‘findings’..
I would like to be able to log the point in my female cycle. I am curious if other women find that they are extremely tired during the beginning of menstruation. My hypothesis is that sleep quality is worse around that time, possibly because (I think?) body temperature is higher.
Also, I would love to know statistical significance and/or confidence intervals around the 2-variable charts, per Mr. Feldman. Additionally, it would be nice to know the median/mean and st dev of the averages on the zq wheel, as well as the gender/age specific averages accessible through the personalized data.
Awesome product for data nerds!