Deep Learning: The Cerebellar Circuit

Introduction

Life is sure convenient, isn’t it? There are an innumerable number of different actions we perform every day without even giving it a second thought. Walking, talking, writing and so on are all examples of these actions, ones that seem mundane at the surface but are truly incredibly complex. For example, walking, requires the activation of numerous cortices and nuclei of the brain, such as the prefrontal cortex for containing the cognitive processes which initiate the action, the premotor cortex which acts as the bridge between these processes and the action they cause, and finally the motor cortex which converts all this neural activity to the format of a complex cognitive map of all your body’s muscles and which specific ones need to contract in order to perform the action (Lim). It is the final step, however, that is the most complex and undeniably important, coordination. This vast majority of this step takes place in the  relatively small and seldom known region of the brain, the pons, the middle third of the brainstem, or more specifically, its bizarre oblong outgrowth, the cerebellum.

Quick History: Evolution and Discovery

The cerebellum is evolutionarily among the oldest regions of the vertebrate brain. It evolved nearly half a billion years ago in the vaguely fish-like ancestor of vertebrates from the nuclei on the dorsal side of the pons, which are closely associated with regulating bodily functions and facial reflexes. All it took was a simple invagination of these nuclei into the surrounding tissue and the cerebellum came into being (Hodos). 550 million years and many evolutionary cerebellar enhancements later, we arrive at its initial discovery in the late 18th century by Luigi Ronaldo. He was the first person to connect motor impairment with lesions to the cerebellum, giving humanity its first key insight into its true function. It was not until nearly a century later that Jan Evangelista Purkinje, through experimental microscopy methods, first described that the brain and nervous system was composed of individual microscopic units known as neurons in 1839. The neurons he specifically discovered with purkinje cells, the giant cells which compose the bulk mass of the cerebellar cortex and will be thoroughly discussed in the following section. At last, we arrive at our final destination, for the purposes of this exploration at least, where Janos Szentagothai and John Eccles through morphological and electrophysical analysis, were able to outline the internal anatomy and circuitry of the cerebellum in the 1960s (Glickstein), which will be the subject of the explanatory section of this report.

The Science: Where Biology Ends and Physics Begins

The cerebellum consists of four main regions, the ancient vestibulocerebellum which fixes gaze, head position and posture, the slightly newer deep cerebellar nuclei which transmit signals between the other two cerebellar structures and the rest of the brain, the fairly new spinocerebellum which helps with basic motor coordination, and the new, prized cortex of the cerebellum, which stores the procedural memory for every single action in its sprawling, fleshy, folded labyrinth. Compared to its far better known counterpart, the cerebral cortex, the cerebellar cortex is far simpler in structure, possessing only three layers, the granular, purkinje and molecular layers, each dedicated to a different cell type. The granular cells which compose the bulk of the cells in the cerebellum, all reside in the granular layer and project their axons into the molecular layer in what are called parallel fibers, because they all run parallel to each other, never crossing or touching. Perpendicular to these fibers are the tree-like dendrites of the purkinje cells which synapse with them in the molecular layer, whilst having their bodies in the purkinje layer, and projecting down past the granular layer toward the deep cerebellar nuclei. Granular cells receive information from neural fibers known as mossy fibers, which consist of kinesthetic information from the spinal cord and motor signals from the motor cortex (remember this from earlier?). On the other hand, purkinje cells receive inputs from both granular cells’ parallel fibers, as mentioned, and climbing fibers, which originate from the inferior olivary nucleus, an important kinesthetic-processing center in the brain. All of these excitatory signals then lead to a single inhibitory response from a purkinje cell to the deep cerebellar nuclei. See the visual aid below for a better understanding (ignore the inhibitory cells, such as the basket and stellate cells as they only play a minor role in cerebellar processes):

Kano, Masanobu, “Simplified Scheme of Cerebellar Neural Circuitry.” Research Gate.

How does this all relate to physics? This entire neural circuit essentially equates to a NOT logic gate. Whatever excitatory signal enters the circuit from the mossy and climbing fibers is converted into an inhibitory by the purkinje cell to inhibit the activity of the deep cerebellar nuclei. To complete this circuit, the deep cerebellar nuclei (DCN) output back to the motor cortex, from where many of the mossy fibers originate in the first place. Every time a movement occurs, it stimulates a mossy fiber signal which is then inverted by this NOT gate into a signal which prevents the DCN from sending a signal back to the motor cortex to stimulate its activity.

Now, to fully understand this circuit, we must delve into how it can be altered. Whenever granule cells/mossy fibers stimulate a purkinje cell, it creates a single inhibitory response, known as a simple spike. However, when a climbing fiber stimulates a purkinje cell, due to how interconnected it is with its dendrites, it creates a powerful, long-lasting inhibitory response, known as a complex spike. After the spike, it enters a refractory period in which it cannot fire. This contrasts simple from complex spikes. Repeated stimulation by both parallel/mossy fibers and climbing fibers will cause a purkinje cell to neurochemically alter its dendritic structure to disconnect parallel fibers and solely receive inputs from climbing fibers, a process known as long term depression. A depressed purkinje cell, unlike its prior state, can only inhibit DCN activity through complex spikes. According to the Marr-Albus-Ito model of cerebellar function, the most popular and widely accepted one, purkinje cells and by extension, the cerebellum, receive body awareness, position and balance information primarily from the mossy/parallel fibers, causing them to constantly fire in simple spikes to inhibit DCN/motor activity. The DCN functions according to a similar but opposite basis, sending a signal to the motor cortex for every proprioceptive or motor signal it receives from the mossy fibers. Purkinje cells can essentially be thought of as filters, selectively inhibiting DCN activity depending on whether they are depressed or not.  If an error in movement occurs, such as a missed free throw in basketball, there will be a sudden burst of activity in the involuntary kinesthetic pathways leading from the spinal cord to the inferior olivary nucleus and into its climbing fibers, causing the corresponding purkinje cells (the ones linked to the motor neurons that caused the error as well as the proprioceptive pathway that sensed it) to output a complex burst to the DCN, inhibiting their firing in the short period that follows, thereby correcting the error temporarily. If this error repeats enough, then through long term depression, the purkinje cells causing it will lose their connection to the parallel fibers and no longer will be able to respond to kinesthetic or motor information through short spikes. The structure of the cerebellum will thus be permanently altered so as to never repeat those errors again. This is the crux of the phrase, “practice makes perfect”. Every time an action is repeated, this whole process repeats along with it, altering the cerebellum’s structure in such a way as to repeatedly refine and better the action by eliminating the firing capabilities of error-prone purkinje cells and their pathways (Purves). The final product is the reward reaped from dedicating the time to rewire one’s cerebellum, a new skill acquired.

Current Trends: Deep Learning

It was this complex yet simple self-altering circuit (as well as many others like it in the brain) that inspired machine learning in the first place. More specifically, one of the most prominent forms of machine learning, known as deep learning, was developed to replicate this process. Just as the cerebellum alters its inhibitive behaviors in response to depressive error signals, AI’s being trained through deep learning are exposed to the specific features of a dataset they are intended to identify. Both systems through repeated exposure to information, adapt to process it as intended by tweaking their parameters and internal configuration. The one substantial difference is that the cerebellum can rewrite its own parameters while the AI may need guidance from developers, but other than that, it is completely autonomous (What Is Machine Learning? | How It Works, Techniques & Applications).

The Future: Artificial Intelligence

How exactly is deep learning applicable to our modern world? It is completely transforming learning and our classrooms at this very moment. There is a relatively little known AI system, one that barely any of us have probably heard of, known as ChatGPT which uses deep learning as its foundation. The more data it is fed by developers, users and the worldwide web, the better it is able to adapt to the task it was given to fulfill. This is the main downside of deep learning, the immense amount of data that is required for it to function effectively. That is also, not coincidentally, the same downside to cerebellar motor learning, the immense amount of data that the system must be fed through repetition of desired movements or in other words practice. Moving on into the future, deep learning will continue to be employed with greater efficiency, requiring less data input and time, to contribute to the forefront of artificial intelligence (Singla).

Conclusion

To say the cerebellum is an amazing brain formation and adaptive learning system is an understatement. Not only is it responsible for the fine-tuning of each and every action we take, but it is in small thanks to attempts to imitate its unparalleled efficiency, that our lives and education are being forever changed by this unprecedented AI revolution we live in the midst of.

Works Cited

Lim, Shannon B., et al. “Brain Activity during Real-Time Walking and with Walking Interventions after Stroke: A Systematic Review.” Journal of NeuroEngineering and Rehabilitation, vol. 18, no. 1, 15 Jan. 2021, https://doi.org/10.1186/s12984-020-00797-w. Accessed 22 Apr. 2021.

Hodos, William. “Evolution of Cerebellum.” Encyclopedia of Neuroscience, pp. 1240–1243, https://doi.org/10.1007/978-3-540-29678-2_3124.

Glickstein, M., et al. “Cerebellum: History.” Neuroscience, vol. 162, no. 3, Sept. 2009, pp. 549–559, https://doi.org/10.1016/j.neuroscience.2009.02.054.

Purves, Dale, et al. “Modulation of Movement by the Cerebellum.” Neuroscience. 2nd Edition, 2001, www.ncbi.nlm.nih.gov/books/NBK11024/.

“What Is Machine Learning? | How It Works, Techniques & Applications.” Www.mathworks.com, www.mathworks.com/discovery/machine-learning.html#:~:text=Machine%20learning%20algorithms%20use%20computational.

Singla, Sonia. “Learning the Basics of Deep Learning, ChatGPT, and Bard AI.” Analytics Vidhya, 25 Feb. 2023, www.analyticsvidhya.com/blog/2023/02/learning-the-basics-of-deep-learning-chatgpt-and-bard-ai/#:~:text=How%20is%20ChatGPT%20programmed%3F. Accessed 29 Jan. 2024.

This is a map of Europe if the borders were defined by genetic similarity (shared haplogroups, SNPs and ethnic heritage) rather than cultural, linguistic and ethnic differences.  I created this map by finding similarities in ethnicity-specific y-chomrosome haplogroup distributions, so much so that the different groups could not be significantly distinguished from each other, and grouping their respective places of origin together.  The haplogroup data was then matched with the human migrations and historical linguistic/cultural groups that gave rise to the different genetic regions to synthesize the parenthetical descriptions.

Key (Gene Pools to Y-DNA Haplogroups)

• Uralic: N3
• Germanic: R1b, R1a and I1
• Celtic: R1b
• Italic: R1b and E3b
• Paleobalkan: R1b, R1a, E3b and J2
• Slavic: R1a and I2
• Aryan: R1a and R2
• Pre-Aryan: I2
• Caucasian: J2 and G2
• Crescentic: J2 and E3b
• Semitic: J1 and E3b
• Turkic: C3 and Q2

Key (Y-DNA Haplogroup Meanings)

• C3: Out-of-Africa Remnants (Usually found in West Neosiberians)
• E3b: Nilotes
• G2: Anatolian Hunter-Gatherers/Farmers
• I1: Northwest European Hunter-Gatherers
• I2: Southwest European Hunter-Gatherers
• J1: South Caucasian Hunter-Gatherers
• J2: North Caucasian Hunter-Gatherers
• N3: West Neosiberian Remnants (Usually found in Uralians)
• R1b: Northeast European Hunter-Gatherers/Pastoralists
• R1a: East European Hunter-Gatherers/Pastoralists
• R2: Southeast European Hunter-Gatherers/Pastoralists
• Q2: West Paleosiberians

Despite its very complex appearance, this is a grosely simplified model of the neural circuitry of the rhombencephalon (medulla oblongata, pons and cerebellum).  The largest simplification is that done to the neural circuitry surrounding the reticular formation which is not fully or even partially understood.  Some notable nerual circuits that can be seen on this graphic are that of the medullary vital nuclei, cerebellum and the pontine breathing centers.

Key:

Neurotransmitters: 

• G+ Glutamate
• G- GABA
• A+ Acetylcholine
• S- Serotonin
• N+ Noradrenaline
• D+ Dopamine
• E- Endorphins

Cranial Nerves:

• XII: Hypoglossal Motor Nerve
• XI: Accessory Motor Nerve
• X: Vagus Motor and Sensory Nerve
• IX: Glossopharyngeal Motor and Sensory Nerve
• VIII: Vestibulocochlear Sensory Nerve

Tracts: 

• SRT: Spinoreticular Tract
• CmF: Columnar Fasciculi
• RST: Reticulospinal Tract
• SOT: Spino-olivary Tract
• Sy: Sympathetic
• PS: Parasympathetic
• Dc: Decussation
• MMF: Medullary Mossy Fibers
• CbF: Climbing Fibers
• VCT: Vestibulocerebellar Tract
• CVT: Cerebellovestibular Tract
• PMF: Pontine Mossy Fibers
• VST: Vestibulospinal Tract

Nuclei

• CnN: Cuneate Nucleus
• GrN: Gracile Nucleus
• IRC: Inspiratory Respiratoty Center
• ERC: Expiratory Respiratory Center
• CdC: Cardiac Center
• VmC: Vomiting Center
• StN: Solitary Nucleus
• HgN: Hypoglossal Nucleus
• AbN: Ambiguus Nucleus
• KOv: Kinesthetic Olive
• MRN: Medulary Raphe Nuclei
• AnC: Apneustic Center
• PtC: Pneumotaxic Center
• ChN: Cochlear Nucleus
• VbN: Vestibular Nucleus
• Fcs: Flocculus
• Vms: Vermis
• PVm: Paravermis
• CCx: Cerebellar Cortex
• IpN: Interposed Nucleus
• DtN: Dentate Nucleus
• FsN: Fastigial Nucleus
• PnN: Pontine Nucleus

Colors:

• Dark Green: Gross Touch
• Green: Fine Touch
• Blue: Kinesthesia
• Dark Orange: Vestibular Sense
• Orange: Audition
• Red: Autonomic Motor
• Purple: Somatic Motor
• Yellow: Neurotransmitter-Specific Signals

Abstract: These problem sets and hypothetical solutions were created in preparation for crafting a more effective hypothesis for the abiogenic origin of life on Earth than the more popular and deeply flawed hypotheses, such as the RNA World and Metabolism-First hypotheses.

Disclaimer: None of these solutions are confirmed or even have sufficient evidence to be considered beyond the realm of hypothesis.  They are merely possible solutions, speculation essentially, and nothing more.

Problem Set 1: Abiogenic Paradoxes

• Chemical Ratios
  • Solution: In abiogenic experiments such as that of Miller-Urey, incorrect, abnormally high concentrations of pure substances were used, which did not mimic true early Earth, panspermic early Mars or another reducing panspermic environment; however, that does not mean it is impossible, rather it is far more unlikely for abiogenesis.
• Water Paradox
  • Solution: Polymers do not emerge until later in evolutionary history until an established metabolic system can create the necessary mechanisms to prevent degradation.
• Asphalt Paradox
  • Solution: Most of these problems have a hypothetical solution, however, the asphalt paradox is the most difficult of these to solve: the very processes which create the chemicals necessary for life also create contaminants which destroy or cause these organic chemicals to be ineffective, therefore, the best solution to this problem boils down to near-impossible events or strange metabolic pathways which may be able to avoid creating tar and can be formed without the creation of excess tar but do not have much resemblance to modern-day metabolic pathways.
• Single Biopolymer Paradox
  • Solution: A metabolic system can create the diverse environments needed to form diverse polymers; not even one polymer is needed to establish a metabolism.
• Probability Paradox
  • Solution: RNA is a bad candidate for the genetic storage of early life, and therefore, should not be used as a benchmark, especially when arguing for the capacity of early nucleic acids to spur positive reactions when an XNA could be a far better, simpler and more versatile candidate for such.

Problem Set 2: Energy Dilemmas

• Prebiotic Soup
  • Solution: Deep-sea hydrothermal vents known as black smokers despite providing ample organic molecules needed for abiogenesis, create ample waste products and thermal energy, on the other hand, white smokers, despite their low energy content and alkalinity, could provide adequate energy for low-energy life forms with genetic information that is resistant to alkalinity such as an XNA.  Unfortunately cellular life forms could not originate from them and would likely need to originate from an unrelated managed metabolism.
• Energy Absorption
  • Solution: Despite seemingly irreducibly complex systems for generating ATP from raw energy sources, an earlier and ancient metabolic component of cellular respiration such as a variant of glycolysis could provide energy without the need for complex protein systems and gradients such as those involved in ATP synthase.
• Fermentation
  • Fermentation, while a possible candidate for ancient metabolisms due to its not not relying on ATP synthase, is highly polluting.  There are 3 possible solutions to this:
    • Solution 1: While rather unlikely, natural processes in a certain environment could clear harmful byproducts
    • Solution 2: Fermenters could coevolve with another organism with a contrasting metabolism which clears their waste
    • Solution 3: Other metabolisms similar to fermentation, however less polluting, represented ancient metabolisms.

Problem Set 3: Polymerization Obstacles

• Dilution
  • Solution: Time increases chance of monomer bonding in a solvent.
• Hydrolysis
  • Solution 1: Proto-metabolic world constitutes early life until it provides a proper environment for polymer formation, especially polysaccharides.
  • Solution 2: Most polymers are more resilient to water degradation then are polysaccharides.
• Homochirality
  • Solution 1: Certain environments, as demonstrated by meteorites, favor certain chiralities over others.
  • Solution 2: Early life could exist with multiple chiralities and evolve to be homochiral.
• Intramolecular Cyclisation
  • Solution 1: Chance prevents many instances of cyclisation.
  • Solution 2: Proto-metabolic environment discourages cyclisation.
• Impurities
  • Solution: Certain possible proto-metabolisms eliminate and, in so doing, discourage organic chemistry.
• Premature Truncation
  • Solution 1: Chance prevents many instances of truncation.
  • Solution 2: Proto-metabolic environment discourages truncation.
• Radiation
  • Solution: Thick reducing atmosphere absorbs UV rays and ionizing radiation.
• Free Radicals
  • Solution: Managed metabolism prevents free radicals and free-riding side reactions.  Coevolution between a proto-metabolism and nucleic acids could lead to the evolution of a managed metabolism.

Problem Set 4: Genetic Hurdles

• Replicase Necessity
  • Solution: All known nucleic acids require a replicase or polymerase as a “clutch” to facilitate their propagation, however, much simpler nucleic acids such as XNAs may have been able to replicate with a far lesser clutch, possibly related to a proto-metabolism. The fact that RNA can undergo independent replication with a replicase or similar enzymes, as indicated by many lab results, including those that created Spiegelman’s Monster, gives significant credence to an earlier acellular ribonucleoprotein world. A world from which selfish genetic elements, viruses and cells arise, which was not created abiogenically, having evolved from simpler managed metabolic systems, but instead represents the LUCA for both cellular and acellular life forms.
• Spiegelman’s Monster
  • In a prebiotic environment, Spiegelman’s Monster would not be an evolutionary trend due to its lack of fitness when it comes to 2 major factors:
    • Solution 1: Early nucleic acids from the RNA World and those that existed before it would partner with and derive energy from a pre-metabolic system which would spur the evolution toward greater complexity, and certainly not simplicity.
    • Solution 2: The lab setup of Spiegelman did not represent the prebiotic world, being devoid of waste and limited resources which would help spur the advancement of metabolism and the nucleic acids that manage it, to clear wastes and derive organic molecules from inconvenient resources.
• Regiospecificity
  • Solution 1: Deformations in the formation and replication of early nucleic acids would be tolerated due to the simplicity of the acids and would thereby not completely hinder their replicating abilities, later being selected for to be removed entirely as both the nucleic acids and the enzymes they created become more efficient.
  • Solution 2: A proto-metabolic system could create a more suitable environment and even be selected to promote the formation of non-malformed nucleic acids which would finish the feedback loop by providing better access to catalyzing enzymes.
• Annealing
  • Solution: A pre-metabolism or ribozymal RNA could provide the necessary catalytic function to separate replicated RNA strands without the need for high temperatures or excess energy, which would destroy RNA.
• Error Catastrophe
  • Solution: Early life could contain slow-replicating nucleic acids that could exist in a state far larger than classically known error-catastrophe-sized threshold, granting them the ability to develop repair enzymes.
• RNA Degradation
  • Solution: Hyper-stable yet impractical XNA which are resilient for diverse PH environments could predate fast-spoiling RNA.
• Lifespan
  • Solution 1: Early nucleic acids replicate rapidly enough to avoid decay.
  • Solution 2: Once nucleic acids exceed a certain length threshold, rapid selection for preservation enzymes may occur.
  • Solution 3: Highly stable XNA molecules could have catalyzed the RNA World.
• Reactivation
  • Solution: A pre-metabolism could create, use and extend the life of nucleic acid constituents in order to allow them to form nucleic acid polymers.
• Divalent Metal Ions
  • Solution: While indeed useful for the creation of simple organic molecules such as simple sugars, they do promote RNA degradation; however, it is likely that some forms of XNA may be immune to their effects.
• Primers
  • Solution: Self-replicating nucleic acids may not need to reside within a membrane, granting them easy access to primers, especially in a metabolic environment.
• Strand Reannealing
  • Solution: The same mechanisms which were mentioned earlier to prevent annealing, could also prevent strand reannealing.
• Spaghetti Conundrum
  • Solution: Encapsulation occurred much later in the evolutionary history of life, possibly after the evolution of selfish genetic elements, thus allowing proper precautions such as RNA replication regulation to evolve and prevent this conundrum.
• RNA Peptide World
  • Solution: Firstly, other XNA peptide worlds may predate the RNA peptide world and so too would metabolic systems, allowing better coevolution of nucleic acids and peptides that would not hinder replication.
• RNA Folding
  • Solution: RNA/XNA communalism may explain the existence of replicating unfolded nucleic acids which normally would be useless and confer no evolutionary advantage; however, in a community with folded ribozymes which cannot replicate, but have functionality, unfolded acids would have a greater evolutionary purpose.  Proto-metabolism and ribozymes could prevent hydrogen bonding of unpaired bases to water and promote hydrogen bonding to other bases.

Problem Set 5: Irreducible Membranes

• Semipermeable Membrane
  • Solution: Managed metabolic world could exist without a membrane and later evolve partial membranes which confer it with small evolutionary energy advantages, thereby spurring the development of seemingly irreducibly complex membranes as it becomes more dependent on these membranes.
• Environment Dependency
  • Solution: An organism with far less genes could be autonomous if those genes correspond to a far simpler metabolism catered to more opportunistic environments.  In a prebiotic environment with a high concentration of prebiotic molecules, an early proto-metabolism could increase its fitness by evolving to convert prebiotic molecules into organic monomers and eliminate wastes.

I created essentially a “family tree” displaying the relations between different tissues in the body through stem cell differentiation and specialization.  Bodily tissues are not completely homogeneous in their cellular compositions, but I treat them as such as most are mainly (at least 80%) composed of a single cell type or two closely related cell types.  The neural crest and its derviatives, unfortunately, could not be fully accounted for due to many of its stem cells being assimilated into various mesodermal tissues such as bone, cartilage and muscle before and during the embryonic period of development.

This is a phylogenetic tree I created which includes all major taxons of biological life and their approximate dates of divergence based on molecular and morphological evidence.  It also includes the complex, hypothetical (I arranged them based on the most prevalent scientific evidence available and a few deductive leaps on my part) relationships between different viral realms, mobile genetic elements and cellular life.  The prebiotic aspects on the tree are also hypothetical and have even less evidence to support them (they are not based on any current abiogenic hypotheses, but rather were tailored to avoid any abiogenic inconsistencies and contrivances, also synthesized by my reasoning).