The History of AI

Hubel, Wiesel, and the visual cortex

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

Lineage
Neurons and networks
Kind
Publication

Who

Johns Hopkins University, Harvard University

What happened

The neurophysiologists David Hubel and Torsten Wiesel record the activity of neurons in the cat's visual cortex and publish their receptive fields in The Journal of Physiology, in 1959 and, with the full architecture, in 1962.

Why it mattered

They show that the visual cortex analyzes an image in stages: some neurons respond to an edge with a certain orientation at a specific spot, and others to the same feature over a wider area. That hierarchy inspires the neocognitron and convolutional networks.

The full story

By the late 1950s it was known that neurons in the retina respond to spots of light, but not what the cerebral cortex did with those signals. At Johns Hopkins University, Hubel and Wiesel used microelectrodes to record the activity of single neurons in the visual cortex of cats while projecting stimuli in front of them.

The neurons barely responded to spots of light. They did respond to edges and bars with a particular orientation: each had a preferred angle and stopped responding if the bar was rotated. In the 1962 paper, by then at Harvard, they distinguished simple cells, which respond to an edge at a precise position, from complex cells, which respond to the same orientation over a wider area. They also found that neurons with the same preference are grouped in columns.

The underlying idea is a hierarchy: the image is broken down into simple features, which other neurons combine into more general features that depend less on position. In 1981 this work earned them the Nobel Prize in Physiology or Medicine, shared with Roger Sperry.

Kunihiko Fukushima turned that hierarchy into an artificial network, the neocognitron, with layers of S-cells and C-cells modeled on the simple and complex cells. The convolutional networks that came to dominate computer vision descend from it.

Videos

Leads to

  • The neocognitron (October 1979)

    Documented direct influence

    The S-cell and C-cell layers of the neocognitron take their names and their roles from Hubel and Wiesel's simple and complex cells

Threads that pass through here

Sources

Scholarship and context

How to cite

“Hubel, Wiesel, and the visual cortex.” The History of AI, revised November 13, 2026. https://thehistoryofai.com/en/milestone/hubel-wiesel-visual-cortex

This entry was last reviewed on November 13, 2026.

Open in the interactive timeline