Breaching the Black Box: A Neuroscience Primer for Neurotechnology Enthusiasts
Elusive & obscure, the human brain is the epitome of evolutionary engineering. For thousands of years, the three pounds of tissue inhabiting your inner skull has captivated scientists and philosophers alike. But while much progress has been made, the brain is still commonly referred to as a ‘black box’ — meaning we’re still clueless as to much of its innerworkings.
In this neuroscience primer, we will explore the basic anatomy of a neuron, neuronal communication, and how brain-computer interfaces might leverage this complex information processing machinery to enhance human potential.
Perched at the intersection of biology and technology, the field of brain-computer interfacing is unique in its incredible interdisciplinarity. BCI development requires a basic background in many fields including neuroscience, signal processing, machine learning, information technology, and engineering. The advancement of high-bandwidth BCIs therefore is a testament not only to human innovation, but to scientific communication and collaboration. This primer is designed as a brief intro to those who may be interested in the development and application of neurotechnology but are unfamiliar with the underlying biology.
The Neuron
The brain owes its computational prowess to the unique manner in which it processes information. This processing is primarily carried out via electrochemical units called neurons, the chief cell-type of the nervous system. In the crudest of descriptions, the neuron can be thought of as a “leaky bag” of charged liquids. The membrane of a neuron is made up of a lipid bilayer with little openings, known as ‘ion channels’ that selectively allow the entrance of charged ions. The levels of charged liquids on the inside vs. outside play a pivotal role in information transfer.
While neurons in different brain regions vary anatomically, a typical neuron is made of a cell body, or soma, with many tree-like branches called dendrites and one long branch called an axon. The axon is responsible for carrying signals to other neurons, while dendrites primarily receive signals from other neurons. Axons are often sheathed in myelin which acts as an electrical insulator boosting the speed of action potentials through the axon thus increasing communication efficiency between neurons.
Membrane Potential
Neurons reside in liquid full of charged particles called ions. Ions are particles which carry a net charge due to the loss or gain of electrons. Anions carry a negative charge, while cations carry a positive one. The difference in the number of charged ions inside versus outside of a neuron is responsible for carrying a signal through a neuron. This difference in charge across a neuron’s membrane is known as its membrane potential.
There is a larger concentration of sodium (Na+), chloride (Cl-), and calcium (Ca2+) ions outside the neuron, while there are more potassium (K+) and organic anions (A-) inside. This ionic imbalance results in a potential difference of about -70 mV across the neuron’s membrane at rest. This is known as a neuron’s resting potential. Ion pumps actively maintain this potential difference through energy expenditure.
To remember the most important ionic differences accounting for membrane potential, you can think of the neuron as a ‘salty banana,’ with lots of sodium on the outside and potassium on the inside.
Action Potentials / Spikes
An action potential occurs when a neuron receives strong enough signals from other neurons to initiate an all-or-nothing rapid rise and fall of membrane potential which propagates down the length of an axon. The rise in membrane potential is caused by the opening of Na+ ion channels allowing Na+ to flow into the cell. Once the membrane potential gets high enough, K+ channels open allowing K+ to flow back out returning the neuron to resting potential. Because this event is all or nothing, information is carried not only through the event itself, but through the number of action potentials, or spikes, occurring over time, also known as the firing rate.
Brain-computer interfaces often leverage these electrical signals to decode the activity of the brain and encode our own information from external stimuli.
The Synapse
Neurons communicate with one another via chemical junctions called synapses. Essentially, an action potential will propagate down the axon of a neuron until it reaches the axon terminal. Then it will induce the release of chemicals called neurotransmitters into the small gap between the terminal of the presynaptic neuron and the dendrite of the postsynaptic terminal. These neurotransmitters bind to receptors on the postsynaptic neuron which generally result in the opening of ion channels and changes to the membrane potential of the postsynaptic neuron.
Synapses can be excitatory, meaning postsynaptic membrane potential increases as a result of receiving an action potential, or inhibitory, meaning postsynaptic membrane potential decreases. Whether or not a synapse is excitatory or inhibitory depends on which kind of neurotransmitters are released. The most common excitatory neurotransmitter is glutamate, whereas the chief inhibitory neurotransmitter is gamma-Aminobutyric acid (GABA).
Synaptic Plasticity
Above all, the brain is more remarkable in its incredible flexibility, constantly evolving in response to environmental and behavioral stimuli. This adaptability, known as synaptic plasticity, is primarily carried out through changes in neuronal connection strength. The two most studied forms of synaptic plasticity are called long-term potentiation (LTP) and long-term depression (LTD).
LTP is the increased strength of a synaptic connection as a result of correlated firing. As neuroscientists often say, “Neurons that fire together, wire together.” This notion is traditionally known as Hebb’s postulate. LTD is simply the opposite of LTP — a decrease in synaptic strength as a result of uncorrelated firing. The changes in synaptic strength stemming from LTP or LTD can last for hours, days, or sometimes longer.
The ability to modify synaptic strengths is the basis for learning and memory in the brain.
CNS vs . PNS
The nervous system can be divided into the central nervous system (CNS) and peripheral nervous system (PNS). The CNS is composed of the brain and spinal cord, which is the primary pathway the brain uses to send motor signals to muscles and receive sensory feedback information from the skin and muscles. Neurons in the spinal cord also function in maintaining reflexes, such as the knee-jerk reflex your doctor tests when they hit your knee with a mallet.
The PNS is constituted by the autonomic nervous system (neurons controlling visceral functions like breathing and heart pumping) somatic nervous system (neurons connected to the skin, sensory organs, and skeletal muscles, skin).
Basic Brain Anatomy
The discipline of neuroanatomy is extensive, intricate, and complex. However, it is advantageous to have at least a basal understanding of the structure of the brain in the context of interpreting signals collected from disparate brain areas because distinctive brain regions are associated with different aspects of cognition and behavior. As an example, if you’re working to improve visual system function you would do better to look at signals emanating from the back of the brain than the front. For the sake of this primer, we will simply detail the location and function of several prominent brain regions.
The brain stem is composed of the medulla, pons, and midbrain. This region is responsible for conveying information from the brain to the rest of the body. The medulla and pons are additionally accountable for basic regulatory functions such as breathing, sleeping, and arousal.
The cerebellum (“little brain”) is the little creased structure sitting at the back of the brain. It is highly involved with voluntary movement coordination, balance, and posture.
The diencephalon is made up of the thalamus and hypothalamus. The thalamus is the brain’s relay station — channeling information from sensory organs to the rest of the brain and vice versa. The hypothalamus is involved in regulation of hunger, thirst, and other homeostatic systems.
Moving away from the base of the brain are the two cerebral hemispheres, containing the neocortex, basal ganglia, amygdala and hippocampus.
The neocortex is the folded surface covering the surface of the hemispheres. It contains 6 layers with a total of around 30 billion neurons. These neurons have about 10,000 synapses with other neurons meaning there are a whopping 300 trillion connections in this region. The neocortex can be divided into different functional lobes: the frontal, parietal, temporal and occipital lobes. Right behind your forehead sits the frontal lobe, which is renowned for its functionality in planning higher level cognitive functions. Meanwhile towards the back of your head lies the occipital lobe, which specializes in visual processing. Between the frontal and occipital lobes are the parietal lobe, which works in sensory integration, and the temporal lobe which functions in auditory processing.
The basal ganglia are a group of structures located deep within the two hemispheres which are primarily involved in motor learning and control. While the amygdala is canonically recognized as the emotional regulator in the brain, the hippocampus is renowned for its function in learning and memory.
How do BCIs monitor brain activity?
Now that we have a basic understanding of neuroscience, let’s look to see how we can use what we’ve learned in the context of BCIs.
BCIs record, analyze, and translate brain activity to output devices designed to carry out desired actions. There is even a growing focus on transcription of neural data to actionable insights devices can leverage to improve user experience. For example, a BCI could monitor focus levels and respond in a way which refocuses you when it deems you distracted. This illustrates an innovative way to engage technology in enhancing the human experience. But while ideas like this are incredibly exciting, the most immediate goal in BCI development is the restoration of motor and communicative function to paralyzed patients. (Nicolas-Alonso LF, 2012).
There are various avenues one can take to detect brain signals, but the principal method being used in current BCI research and development is electrode collection of brain electrical activity. Let’s look a little more closely how this is accomplished and the advantages / shortcomings of different approaches.
Invasive vs. Noninvasive BCIs
Non-invasive BCIs often rely on electroencephalogram (EEG) measurements of electrical activity in the brain via macro-electrodes placed on the scalp. These electrons record electrical brain wave patterns. These patterns are a result of the summation of the very small electrical signals produced by individual neurons due to the movement of ions. This approach is advantageous in allowing cost-effective and monitoring of large-scale neuronal activity with minimal risk. However, the tradeoff to recording from the scalp as opposed to in the cortex is lower spatial resolution and the inability to detect higher frequency signals. (Waldert S., 2016).
On the other side of the spectrum, invasive BCIs rely on the use of surgically inserted intracortical microelectrodes into the gray matter of the brain. These electrodes record the electrical activity of neurons at a much higher temporal and spatial resolution. These electrodes are optimized to record neuronal action potentials (APs) and low field potentials (LFPs), the localized electrical activity occurring in the extracellular surrounding neurons. But although this approach allows for greater regional specificity, the dangerous and costly nature of neurosurgery poses a critical issue to invasive BCI therapeutic treatment (Steyrl, David, 2016).
Much of this piece has focused on the physiology of neurons and the basis of neuronal connection. This is largely because brain-computer interfacing often relies on the electrical nature of nervous system communication through use of electrodes which monitor this activity invasively or noninvasively.
Despite the brain being the most complex piece of machinery known to man, hopefully this primer has given you some context as you enter into the exciting field of neurotechnology.
Resources
Purves, Dale. Neuroscience. 6th Edition. Sinauer Association; 2017.
Rao P. N., Rajesh. Brain-Computer Interfacing: An Introduction. Cambridge University Press; 2013.
Nicolas-Alonso LF, Gomez-Gil J. Brain computer interfaces, a review. Sensors (Basel). 2012;12(2):1211–1279. doi:10.3390/s120201211
Steyrl, David & Kobler, Reinmar & Müller-Putz, Gernot. (2016). On Similarities and Differences of Invasive and Non-Invasive Electrical Brain Signals in Brain-Computer Interfacing. Journal of biomedical science and engineering. 9. 393–398. 10.4236/jbise.2016.98034.
Waldert S. Invasive vs. Non-Invasive Neuronal Signals for Brain-Machine Interfaces: Will One Prevail? Front Neurosci. 2016 Jun 27;10:295. doi: 10.3389/fnins.2016.00295. PMID: 27445666; PMCID: PMC4921501
Images created using Canva and BioRender.com
