Sleep, vol. 1: intro, EEG and fMRI BOLD
Sleep: EEG and fMRI BOLD
Since everyone with a high-speed internet connection is ‘doing research’, let me disclose that I am not a sleep medicine professional.
However, I have been very fortunate to learn from people (actual researchers), who are very well versed in this domain. It is their work that inspired me to write a piece on sleep, with something that hopefully approximates a degree of scientific accuracy, relevant to my understanding of human physiology
The ‘godfather’ of EEG research, Hans Berger, was the first one to report the discovery of brain electrical activity in humans as measured by the electroencephalogram in 1924 (Berger, 1929). We have since enjoyed an unprecedented uptick in scientific and clinical applications of EEG. Courtesy of ongoing research and several milestones which allowed to determine the safety of this non-invasive technique, the use of EEG has proliferated and spilt-over into other areas outside of the medical realm. In fact, wearable EEG devices are now beginning to become commercially available for users worldwide (De Zambotti et al, 2019).
EEG, or electroencephalography, is a non-invasive technique allowing to study brain function in the form of electrical brain activity. Scalp electrodes are placed on the subjects forehead in order to record electrical events. In contrast to intracranial EEG, commercially available wearable EEG devices do not necessitate the use of electrodes to be placed directly on the exposed surface of the brain. Instead, the oscillation of cortical currents is measured several centimeters below the scalp. High quality non-invasive EEG devices are able to record brain activity while mitigating the interference of multiple resistive layers (most notably, the skull) which could otherwise have a blurring effect (Burle et al. 2015)
Brain waves, or more accurately, neural oscillations can be conceived of as synchronous, repetitive patterns of neural activity. Raw data from EEG allows you analyze how those patterns (measured in units of Hertz) occur across time. The oscillations are characterized by frequency range and are grouped into corresponding bands, which allow the researcher to determine relevant brain state with a high degree of temporal accuracy (measured in milliseconds). When segmenting the ever-changing electrical brain signals into relevant categories as a way of characterizing brains states, one need to be mindful not to neglect their graded dynamics (Libourel et al. 2015), however this is beyond the scope of this article.
Sample EEG data showing various brain waves and corresponding brain activity. Image courtesy of Enchanted Wave
EEG: TALK TO ME AS IF WAS A 5 YEAR OLD.
No problem: grow up.
(okay, I am kidding!)
In his book, ‘Why we sleep’, Dr Matthew Walker, painted an interesting analogy depicting EEG as a microphone dangling in the middle of a stadium; what follows is my own take and extension of this concept.
Due to its central location, such microphone (EEG device) can register the cheering and booing of the crowd (brain activity) and tell you exactly at what timestamps (temporal resolution) noise levels went up (excitory postsynaptic potentials) or down (inhibitory postsynaptic potentials). In the course of any game, there are easily distinguishable patterns (neural oscillations). For example, as the crowds gather before the first whistle (sleep onset), the characteristic chatter of incoming fans becomes detectable.
No game is ever the same (nor is your recorded sleep brain activity). However, there are several ‘set pieces’ which reliably mark every game (sleep architecture). In fact, when looking at the audio recording (EEG data), you can determine how things unfolded on the athletic field that night across time (sleep staging). This is due to the occurrence of noise patterns (sleep cycles). Rather predictably (and owing to cyclic nature of sleep stages), you will able to identify the key events in relation to noise level : team warm up, last-second touchdown and the final whistle will all generate different levels of rowdiness (or lack thereof) from the fans. After the referee blows the final whistle and the spectacle (sleep) comes to an end, you might wish to check how long you have been in the stadium for (total sleep time).
However, even the best quality microphones may struggle to identify where exactly the noise is coming from (spatial resolution), and where the fans are located in the stadium (brain regions). This makes perfect sense- the avid stadium goer that you are, you can personally attest to the fluid dynamic of the crowd. Many fans hardly ever sit still- in fact, many of them like to move around (cerebral blood flow), or perhaps even migrate to a different zone. As the fans spontaneously shift from one place to another, so will the noise levels (that’s because cerebral blood flow and neural activity are coupled). How can you account for any such excursions?
If you were to be tasked with investigating (or mapping out) the exact make-up of migrating crowd based on an audio recording alone, you would understandably struggle. As you start tearing your hair out, your inner Sherlock tells you that therein lies the solution: you might indeed need to go BOLD! After all, the crowds need oxygen to live, be active and move around (and as such, are Blood Oxygen Level Dependent- BOLD).
Functional Magnetic Resistance Imaging works by detecting changes in blood flow and oxygenation. Increase in neural activity is accompanied by an increase in oxygen consumption in the relevant brain region(s). In order to meet this demand, blood flow increases to the active area(s). Hemoglobin delivers oxygen to neurons via capillary red blood cells. fMRI Blood Oxygenation Level Dependent imaging technique accounts for fluctuations in blood oxygenation as they pertain to neural activity. As such, it allows to indirectly measure brain activity
The only catch? You need to be comfortable sleeping inside a large, cylinder shaped tube surrounded by a circular magnet. The cost of an fMRI machine is considerably higher as compared to EEG devices. Unfortunately, it’s not exactly portable, either.
Stay tuned for Part 2!
Burle B, Spieser L, Roger C, Casini L, Hasbroucq T, Vidal F. Spatial and temporal resolutions of EEG: Is it really black and white? A scalp current density view. Int J Psychophysiol. 2015;97(3):210‐220. doi:10.1016/j.ijpsycho.2015.05.004
Cebolla AM and Cheron G (2019) Understanding Neural Oscillations in the Human Brain: From Movement to Consciousness and Vice Versa. Front. Psychol. 10:1930. doi: 10.3389/fpsyg.2019.0193
De Zambotti M, Cellini N, Goldstone A, Colrain IM, Baker FC. Wearable Sleep Technology in Clinical and Research Settings. Med Sci Sports Exerc. 2019;51(7):1538‐1557. doi:10.1249/MSS.0000000000001947
Hillman, E. (2014). Coupling Mechanism and Significance of the BOLD Signal: A Status Report. Annual Review Of Neuroscience, 37(1), 161-181. doi: 10.1146/annurev-neuro-071013-014111
Kamarajan C, Pandey AK, Chorlian DB, Porjesz B. The use of current source density as electrophysiological correlates in neuropsychiatric disorders: A review of human studies. Int J Psychophysiol. 2015;97(3):310‐322. doi:10.1016/j.ijpsycho.2014.10.013
Libourel PA, Corneyllie A, Luppi PH, Chouvet G, Gervasoni D. Unsupervised online classifier in sleep scoring for sleep deprivation studies. Sleep. 2015;38(5):815‐828. Published 2015 May 1. doi:10.5665/sleep.4682
Shambroom JR, Fábregas SE, Johnstone J. Validation of an automated wireless system to monitor sleep in healthy adults. J Sleep