liveBook · Manning

Neuron

As explained in Machines that Think

A nervous-system cell that processes and transmits signals through electrical and biochemical activity.

A biological neuron is a nervous-system cell. In artificial intelligence, “neuron” names a deliberate abstraction: mathematical operations over weighted inputs replace biological complexity, while learning adjusts connection weights. The shared term invites comparison, but does not make the mechanisms equivalent. section 5.5 section 5.1.2

Biological neurons can adapt their activity in response to electrical feedback. That observable adaptation does not by itself establish understanding or intention, and neural wiring alone does not determine function apart from the developmental and experiential conditions in which it forms. section 5.3.3

Neurons responding to a game

Human neuron clusters connected to Doom received patterns of electrical stimulation derived from game visuals; their responses were interpreted as actions. The resulting behavior resembled gameplay, but did not mean the neurons were playing with human understanding or intent. section 5.3.3

How Machines that Think covers it

How deep it goes
Contrasts biological complexity with weighted-input abstractions, but does not detail neuron anatomy or biochemical mechanisms.
Examples
Describes human neuron clusters responding to electrical stimulation linked to the game Doom.
What you can do after
Evaluate neuron-based claims by separating observed adaptation from understanding or intention.
How it connects
Links neurons to artificial neurons, neural networks, distributed representations, and connectomics.

Also explained in

Compare how 3 books explain neuron →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machines that Think. The section links point to the parts of the book it draws on.