Tip:
Highlight text to annotate it
X
This whole process we’ve learned a lot about the complications of measuring sleep and reporting
it to people. It’s a complex thing; you know, when you think about sleep, the subjective
experience is that I close my eyes, I become completely unconscious, and at some point
my eyes open and I become conscious again.
One: we have no idea what happened there, and Two: if someone said that they could measure
it, you’d think it would be pretty straightforward.
Turns out when you record one night of sleep--one recording--you have two people look at that
same recording, they can come up with very different answers--even with the gold standard
rules that have been around for forty years.
And so, an average agreement between two people is about 85%. I’m throwing this up there
to give us a little understanding so when we talk about how we validated Zeo, and what
those numbers looked like, how did they compare to this 85%.
Another thing to look at is the subjective reporting. When people talk about how deep
their sleep was or how restorative their sleep was, it has basically zero correlation with
what was actually happening in terms of sleep quality. There might be a little bit around
how much deep sleep you actually got--the more deep sleep you get can actually make
your subjective report a little bit worse--but that’s all very tenuous; it’s not a very
strong connection.
This is actually true in very Light sleep and REM sleep where people don’t have the
subjective experience of “Am I asleep or not.”
So, another thing: So we know that sleep is a subjective measure. Another thing we have
to keep in mind is when we’re talking about PSG, the gold standard, we’re recording
sleep from the scalp, mainly from the face, from the head. There are some additional signals
around the body, but mostly it’s from up here.
But when you start thinking about it, sleep is really a whole body phenomenon. So when
you put your definition all up here, you’re going to be missing some of the data that’s
elsewhere.
This is especially true when you’re thinking about circadian rhythms. There’s more and
more data coming out now that your brain has one circadian pacemaker--and it’s called
the superchiasmatic nucleus which is in the pineal gland, which is dictating your body’s
day and night rhythm.
Turns out that all our organs might have internal circadian clocks that could be running on
different times, especially if you’re jet lagged, or if you’re one of those people
who like to sleep in on weekend to catch up on lost sleep during the week. You could be
putting you’re brain on a different schedule than your heart, which is on a different schedule
than your liver, which is on a different schedule than your kidneys, and it can really start
to screw up your ability to sleep and to sleep well.
The third thing to talk about when thinking about the gold standard: When you’re picking
up all these electrical activities from the brain, it’s aggregate data. You have billions
of brain cells but the one signal that you’re going to be getting is “what are all these
brain cells doing in concert?”
When you’re in Deep sleep, they’re all kind of firing at the same time; that’s
where these big, slow waves where everything is happening all at once [are].
When you’re awake, lots of neurons can be firing in different places at different rates
so you end up with really noisy, choppy, low-amplitude signal. This is especially true in terms of
the differences of what’s going on in your brain at different points.
Sleep onset--Sleep offset; you’d think “I fall asleep, that means I’m unconscious,
it happens like this”. It doesn’t.
Parts of your brain--and I can’t remember which direction it is--but your hippocampus
is falling asleep at a slightly different time than your cortex and it kind of propagates
back and forth. That’s at sleep onset and sleep offset.
So there’s a couple of minutes, perhaps, where you’re falling asleep and when you’re
waking up, when you’re actually sort of in-between the two, and it’s really difficult
to define whether you were awake or asleep, objectively and subjectively.
Again, with the aggregate sleep data, sleep spindles, [those things that I mentioned before,
those just don’t happen brainwide--it’s not like your whole brain just “freaks out”
and creates this spindle feature. What happens is that is generally happens--it occurs--it
starts in one place and then propagates to the rest of your brain.
It’s kind of a really cool phenomenon if you look at some of the MRI data on how sleep
spindles propagate; it’s actually very fascinating data.
A third think to think about is that some of these features that we talk about sleep--I’ve
mentioned these big slow waves that we get during deep sleep--those occur more upfront.
You’re going to get more concerted activity up here.
A feature of being awake and of a quiet awake--when you close your eyes and relax--most people--something
like 90-99% of the population--create what’s called alpha waves. It’s a very distinct
feature, but that happens all back here. So if you’re recording something up here you’re
not going to pick up that key indicator of wakefulness coming from the back of the head.