What's the Benefit of So Many Sensors?

 

First, I just wanted to share with all of my backers, that the short article I wrote, for Huffington Post, is now live. Learn How To Sleep Like A Baby.  It's a short and fun read. I encourage comments on the article page, too!

In my last lab note, I promised I would explain why these new sensor caps, with 124 sensors rather than just 1-30, are so important for research.  I touch on that briefly in the article above. For a long time, we thought sleep was just a state in which the brain was turned off.  In the 1950's, rapid-eye-movement sleep was discovered, by a graduate student studying infant sleep - how encouraging!  At that point researchers began to focus on two different states of sleep, which I call macrostates, rapid-eye-movement and non-rapid-eye-movement. You can actually break sleep down into these two states pretty well with only one sensor, and less accurately using a consumer actigraph, like a Fitbit. A few years ago, NSF funded my colleagues and me in the Gluck Lab, using this technique with a commercial one-channel electroencephalography (EEG) device called Zeo, which allowed us to measure sleep brain activity over two consecutive weeks, at home, in the same individuals. We had some exciting findings involving emotion,  which are soon to be published, in the journal, Neurobiology of Learning and Memory.  

Researchers have learned a lot from separating sleep into these two macrostates, and this can be really useful both at a clinical level and in consumer devices. However, it only gives us a piece of the picture, as with so many things in science, more data measurements over time or space can give a different point of view. More recently, sleep, and non-REM sleep in particular, has been found to have a localized structure. Meaning, certain parts of the brain can have different activity than other parts, simultaneously. This had been found in animals (notably certain fish like dolphins, and birds, where one half of the brain is asleep while the other half is awake allowing them to swim or fly while sleeping), but this localization could not be confirmed in humans until these dense EEG (many sensors, typically from 128-256) caps were used. By examining brain activity at a finer detail, using many sensors across the whole head, we can differentiate different types of activity in different brain regions and consider how they relate to the development of behaviors like cognition, language and motor skills.  In order to explore brain activity as a predictive biomarker of specific types of behavioral development, we need this kind of high resolution technology.

In my next lab note, I will explain more about the E4 autonomic sensors that this campaign will help to fund.