Neurons are the building blocks of the brain. They take information they receive from upstream neurons and summarize it as a spike that is sent to downstream neurons.
(image courtesy of wikimedia commons)
Let’s talk about the anatomy of neurons:
- Nucleus: Like any cell in your body, each neuron has a nucleus, where its DNA is stored
- Dendrite: The dendrites of a neuron are its receivers. They collect information from upstream neurons. This information is passed in the form of neurotransmitters, which are chemicals that bind to the dendrites.
- Cell body: The cell body of a neuron houses the nucleus, and also stores electrical energy (voltage). When the voltage of the cell body is above a certain threshold, the neuron will spike.
- Axon: The axon is the part of the neuron that carries the spike downstream. It is covered by the myelin sheath, a substance produced by Schwann cells, which insulates and keeps the electrical energy inside. The energy is renewed at the Nodes of Ranvier.
- Axon terminal: The downstream end of the neuron. When the neuron spikes, the neuron releases neurotransmitter of its own at the axon terminal.
The connection between two neurons (where the dendrite of one meets the axon terminal of the other) is called a synapse.
In its most basic form, information flows through a neuron like so (here is a good video for more details):
- Neurotransmitters attach to receptors on the neuron’s dendrites and cause ion channels and gates to open.
- The activation of ion channels and gates cause the voltage in the cell body of the neuron to change. Excitatory signals cause a voltage increase, and inhibitory signals cause a voltage decrease.
- When the voltage in the cell body passes a threshold, it initiates an action potential, sometimes referred to as a spike.
- The action potential travels down the axon. You can think of the axon (insulated by the myelin sheath) as a large wire that conducts the electrical spike.
- The nodes of ranvier act like amplifiers of the signal, giving it a boost (this is helpful because the conductance in the axon is not perfect).
- When the action potential reaches the axon terminal, it triggers the release of neurotransmitters, which then pass the signal on to the other neurons downstream.
While there are many types of neurons, they all generally follow the above procedure, and they also follow two rules:
- Neurotransmitters are released in discrete amounts, as in either 3 molecules of neurotransmitter are released or 4, never 3.5.
- Neurons either excite or inhibit (very rarely both) downstream neurons
How does learning happen in the brain?
Let’s first talk about the strength of a connection between two neurons. Two neurons can be more tightly coupled, i.e. have greater synaptic strength if either the upstream neuron has more quanta of neurotransmitter to release, or if the downstream neuron has more receptors for that neurotransmitter. Note that these are both discrete concepts.
The guiding theory of how synaptic strength can be changed is called Hebbian Learning. The basic principle is that “neurons who fire together, wire together.” When the upstream neuron fires and then the downstream neuron fires soon after, the synaptic strength between them increases. When the opposite happens (downstream fires before the upstream), the synapse weakens.
A synapse can be strengthened by increasing the amount of neurotransmitter released by the upstream neuron, or by having the downstream neuron grow more receptors to that neurotransmitter. It can be weakened through the selective pruning of dendrites on the downstream neuron, or by a decrease in upstream neurotransmitter.
This is very simple and it’s hard to see how learning complex concepts or tasks could possibly arise using this simplistic learning rule. However,this is currently the most consistent theory we have for learning in the brain.
This is a fairly abstract overview of what neurons are and how they function in the brain. There are many types of neurons that exhibit heterogeneity in structure and behavior. How neurons function, even in small circuits, is still an active area of research in neuroscience.