Theories of neural networks have been developed over many years. Since the late 19th century, there have been attempts to create mathematical models that mimic the functioning of the human nervous system. The discovery by Cajal in 1892  that the nervous system is comprised of neurons communicating with each other by sending electrical signals down their axons, was a
breakthrough. A neuron receives various signals from other linked neurons, processes them, and then decides to inhibit or send a signal based on some internal logic, thus acting as an electrical gate. This distributed neuronal process is the mechanism the brain uses to perform its various tasks. This finding led to an increase in research on mathematical models that attempt to simulate such behavior, with the hope, yet to be substantiated, of achieving human-like classification performance.