To build an artificial brain, it helps to first look at the real thing. A biological neuron has three main parts. Dendrites act like antennas, collecting signals from other neurons. The cell body, or soma, processes these incoming signals. If the combined signal is strong enough, the soma generates its own signal, which travels down the axon to be passed on to other neurons.

In 1943, neurophysiologist Warren McCulloch and logician Walter Pitts proposed a simplified mathematical model based on this process. Their goal wasn't to perfectly replicate the biological neuron, but to create a basic computational unit that could perform logical operations. This model, the McCulloch-Pitts neuron (MCP), became the ancestor of every artificial neuron used in AI today.