The History of AI

Hebb's rule

Era 2 · AI is born (1943–1979) › Neural networks and learning

Lineage
Neurons and networks
Kind
Publication

Who

McGill University

What happened

The Canadian psychologist Donald Hebb publishes The Organization of Behavior: A Neuropsychological Theory with Wiley.

Why it mattered

It proposes that when one neuron repeatedly helps make another fire, the connection between them grows stronger. It is the first learning rule for networks of neurons.

The full story

In the late 1940s, behaviorist psychology explained learning as associations between stimuli and responses without saying what happened in the brain. Hebb set out to bridge that gap: to explain perception, learning, and memory in terms of how neurons are connected.

The most quoted postulate reads: "When an axon of cell A is near enough to excite a cell B and repeatedly or persistently takes part in firing it, some growth process or metabolic change takes place in one or both cells such that A's efficiency, as one of the cells firing B, is increased." It is often summed up as "neurons that fire together wire together." Under that rule, groups of neurons that are active at the same time form "cell assemblies" that stand for a concept and can be reactivated, like a memory.

Hebb offered not a formula but a physiological hypothesis. Yet it supplied exactly what the McCulloch-Pitts model lacked: a way for connections to change with experience, with no one adjusting them from outside.

The rule inspired Minsky's SNARC in 1951, the first computer simulations of networks, and, decades later, unsupervised learning and Hopfield networks. In neuroscience, long-term potentiation, discovered in 1973, is regarded as a partial confirmation of the idea.

Leads to

  • The perceptron (1957)

    Documented direct influence

    The 1958 paper cites Hebb and takes from Hebb the idea that connections change with activity

  • SNARC (1951)

    Inherited, with no proven direct link

    SNARC strengthens the connections that have just fired, a mechanical version of Hebb's rule combined with reward

Threads that pass through here

Sources

How to cite

“Hebb's rule.” The History of AI, revised November 13, 2026. https://thehistoryofai.com/en/milestone/hebb-rule

This entry was last reviewed on November 13, 2026.

Open in the interactive timeline