In 1949 the Canadian psychologist Donald O. Hebb published the book “The Organization of Behavior: A Neuropsychological Theory” (Wiley, New York). Only the year is documented as the publication date. The book set out a theory of how learning could be grounded in the physical changes between brain cells.
Hebb’s central proposal, now known as Hebbian learning, is often summarized as “neurons that fire together, wire together.” When one neuron repeatedly helps to fire another, the connection between them grows stronger. This gave researchers a concrete, biologically motivated rule for how a network of neurons could adapt and store information through experience.
The McCulloch and Pitts model showed that neural networks could compute; Hebb supplied the missing piece of how such networks might learn. His rule directly influenced the first trainable artificial neural networks, including Rosenblatt’s perceptron, and the principle remains central to both neuroscience and modern machine learning.