The SLEEP research symposium in January 2020 changed the mindset of how we look at the brain. Jenna Lendner, an anesthesiology resident at the University Medical Center in Tübingen, Germany, presented a different outlook on people's brain activity. And, as it turns out, there are fragile lines for signs between wakefulness and unconsciousness.
When physicians look at patients under comatose or anaesthesia, they need to distinguish between the two conditions correctly. But that can be a bit tricky as people in the dreaming state of rapid-eye-movement (REM) sleep also produce the same oscillating brain waves as they would when they are awake.
Quanta Magazine did the original research study with the aims of enhancing the understanding of science.
Jenna Lendner of the University Medical Center proposed that the answer isn’t in regular brain waves. It is actually in the neural activity that scientists often ignore: the brain’s background noise. Some were stunned by this argument. To sceptics, Lendner said – “someone’s noise could be another one’s signal”. Her approach was nevertheless fascinating to understand.
Lendner is one of the neuroscientists who came up with the idea that noise in the brain’s neural activity could help us understand its inner workings. This theory poses that what was once seen in television static could impact how scientists study the brain.
Voytek's Brain Algorithm Saves The Day!
Bradley Voytek, an associate professor of cognitive science and data science at the University of California, believed in studying brain activity’s noisy features. His study state that, based on electrical noise, brain activities change when people age. Statistical trends in irregular brain activity made him question that something was missing. He spent years trying to understand the data.
He collaborated with our neuroscientists and developed software that isolates regular oscillations, like the alpha waves that study both sleeping and waking subjects hidden in the brain activity’s aperiodic parts. This gave them a new tool to understand the regular waves and aperiodic activity in behaviour, cognition, and disease.
Neuroscientists quickly defined the phenomenon as the ‘scale-free activity’, but Voytek decided to name it ‘the aperiodic signal’ or ‘aperiodic activity.’ Lendner, Voytek, and others noticed that the pattern was similar to the complicated relationship throughout the natural world and technology.
Complex Brain Statistical Structure
The statistical structure is so mysterious that even scientists think it represents an undiscovered law of nature. None of them could establish what the arrhythmic brain activity could mean. They came up with better tools for isolating the aperiodic signal in new experiments.
Voytek’s algorithm has paved the way for numerous other studies that understand aperiodic activity, and could help study ageing, sleep, and development.
The main question arises- what is an aperiodic activity?
When our bodies groove to the rhythms of heartbeats and breaths- some cycles are essential for survival. They are equally important as the brain’s drumbeats that don’t have a pattern but better understand our behaviour and cognition.
Excitation arises when a neuron sends out a chemical called glutamate to another, and its recipient ends up making more ‘fire’. When a neuron spits out the neurotransmitter gamma-aminobutyric acid or GAB, the recipient neuron is unlikely to fire-up – this is called inhibition.
There are consequences to both. Too much excitation could lead to seizures, whereas inhibition could characterise more sleep or even coma.
The scientists tried to measure the brain’s electrical activity with electroencephalography, or EEG with cycles of waves between excitation and inhibition in different mental states to study the balance between both. If the brain emissions are around 8-12 hertz, the alpha wave pattern is associated with sleep.
Excitation V/S Inhibition - Different brain facets
The brain’s electrical output does not produce smooth curves; instead, there are slow upward and downward peaks. There is no regularity, and it looks more like electrical noise, which is fascinating. The ‘white noise’ is like static but has more of an impressive structure. The imperfections in smoothness and the noise are defined as ‘random’ by Voytek, making it even more enjoyable.
The scientists broke down the EEG data to quantify the aperiodic activity almost like a prism. They employed the technique called ‘Fourier Analysis’ and jotted down the trigonometric functions like sine waves, expressed as frequency and amplitude. They plotted the waves of amplitudes at different frequencies in a graph. It is known as a power spectrum.
They are plotted in logarithmic coordinates because of the wide range of values. This can help understand the random white nose- when it is flat and horizontal, a slope of zero is of the same frequencies. But when the neural data produces curves and negative slopes at lower frequencies, it has higher amplitudes.
This shape is called 1/f, which represents the inverse relationship between frequency and amplitude. Neuroscientists use this to understand the brain’s inner workings through the flatness or steepness of the slope.
The 1/F Phenomenon paved the way
According to Lawrence Ward, a cognitive neuroscientist at the University of British Columbia, EEG data can help understand the sound saves from an audio recording made on a bridge over a highway. The sound of the whistle or hum of tires produces aperiodic background features. It stimulates the spike in the sound wave adding to the 1/f slope.
The 1/f phenomenon dates back to a 1925 paper by JB Johnson of Bell Telephone Laboratories, which looked at noise in vacuum tubes. Hans Berger, a German scientist, published his study on EEG four years later. He found the 1/f fluctuations in electrical noise, stock market activity, biological rhythms, and even some music pieces.
In the 2010 paper in Neuron, the scientists found that EEG readouts, seismic waves, stock market fluctuations exhibit 1/f trends, but they exhibit higher-order statistical structures. A doubt arose in the law of nature, generating aperiodic signals in everything. However, in recent years, evidence suggests that scale-free brain activity contributes to brain functioning.
Voytek stumbled upon the idea of aperiodic signals. His motive was to remove white noise from EEG data, but he hacked a way to pull out noise entirely. In a 2015 study, Voytek and his advisor, Robert Knight, discovered that older adults’ brains have more aperiodic activity than young adults. As the brain ages, it is dominated by white noise, and the correlation between noise and age-related working memory declines.
Voytek posted his code to a website in April 2018 – it was downloaded 2000 times in a month. Because of its popularity, a dozen scientists were interested in knowing more. One of the collaborations was Lindner’s study about markers for arousal during sleep in July 2020. With Voytek’s software, neuroscientists found aperiodic noise for test subjects’ EEGs.
High-frequency activity dropped off faster during REM sleep than in the awake state. Lendner and other scientists believed that aperiodic signals are a unique signal to measure a person’s state of consciousness. This could even help better the practice of anaesthesia and coma patients.
The Mystery Behind what causes the Aperiodic Signals
Voytek’s code has led to a lot of research on aperiodic signals. There has been a lot of interpretation in neuroscience. The only limitation is that scientists haven’t discovered what causes the aperiodic features physiologically. Scientists have been working on sources to identify and accumulate knowledge.
Many theories explain aperiodic signals, such as where they could reflect the brain’s physical organisation. Neuroscience magazine suggests that the brain can make predictions about the sounds with 1/f properties, about the aperiodic activity involved in the processing and predicting the natural stimuli. Something that might not seem surprising is that music also has 1/f properties as music creates the mind.
There is yet a lot of testing to do to understand the neural circuitry. Researchers are trying to link brain-sites to understand the specific activity patterns better and predict the aperiodic and periodic signals in different brain signals. This is just one experiment. Others could apply the code as well.
Baillet explained that aperiodic signals are like dark matter. “It is the invisible scaffolding of the universe that interacts with ordinary matter through gravity. We are not sure of what it is made of, but it is out there holding the Milky Way together”.