The Motor Cortex and the Homunculus

From Fritsch and Hitzig's dogs to Penfield's maps and modern brain-computer interfaces: how the motor cortex represents the body, and why the neat 'little man' is only part of the story.

Discovery

In 1870 the German physicians Gustav Fritsch and Eduard Hitzig passed weak electric currents through the exposed cortex of dogs and found that stimulating particular spots made specific limbs on the opposite side of the body twitch. They had discovered the motor cortex — strong evidence that brain functions are localised. The Scottish neurologist David Ferrier extended the work to monkeys, and Charles Sherrington mapped apes.

Penfield's homunculus

In the 1930s to 1950s, Wilder Penfield stimulated the cortex of awake human patients during epilepsy surgery. The primary motor cortex, on the precentral gyrus just in front of the central sulcus, contained an orderly map: the toes and legs at the top, on the inner surface; then the trunk, arm and hand; then the face, lips and tongue near the bottom. Body parts needing fine control — fingers, lips, tongue — had disproportionately large areas. Drawn as a figure, this became the motor homunculus.

Large neurons, long reach

Layer 5 of the motor cortex contains some of the largest neurons in the brain, the Betz cells, whose axons run all the way down the spinal cord. Some connect directly onto motor neurons controlling the hand — direct connections especially developed in primates and thought to underlie skilled finger movements.

More than a keyboard

The neat map is a simplification. Neighbouring areas overlap, a single muscle is represented in many places, and the map changes with learning and after injury. In 2002, Michael Graziano showed that longer stimulation of monkey motor cortex evoked complex, purposeful actions — bringing the hand to the mouth, defensive gestures — organised more by behaviour than by body part. A 2023 study by Evan Gordon and colleagues even reported regions interleaved within the classic map that seem to coordinate whole-body action and connect to planning networks.

Reading the motor cortex

Because motor cortex activity encodes intended movements, it is the main target for brain-computer interfaces: arrays of electrodes there have allowed paralysed people to move robotic arms and computer cursors by thinking about moving.