In the 1930s, Wilder Penfield was operating on patients with epilepsy and brain tumors. The brain itself doesn't feel pain, so patients could be kept awake during surgery. Penfield would touch the exposed cortex with an electrode and ask: what do you feel? The patient would say something — a tingling in the thumb, a sensation in the lip — and Penfield would mark the location on a paper map of the brain surface. Over many surgeries, a complete picture accumulated.
The resulting map is not proportional to the body. It's not even close.
The hands — particularly the fingers and thumb — occupy roughly a third of the somatosensory cortex. The face and lips together take up nearly another fifth. The entire torso, from shoulders to hips, gets a strip so narrow it barely registers on the diagram. If you drew a person scaled to their cortical representation, the hands would be enormous, the lips huge, the tongue oversized, and the whole of the trunk would be a thin wedge of tissue between the arms and the genitals.
The classic image of the homunculus captures this: a grotesque figure, hands and face bloated, body and legs vestigial. It looks like a caricature. It's a faithful map.
The allocation follows a logic. Two-point discrimination — the ability to tell apart two nearby points of touch — is best at the fingertips, less than three millimeters. On the back, the threshold is forty millimeters or more. The cortex is mapping precision. Where the body needs fine-grained tactile feedback, the brain invests. Where it doesn't — the broad flat territory of the back — it allocates accordingly.
This map is plastic. It changes with use. Violinists who have spent years making precise fine movements with the left hand have a measurably larger cortical representation for those fingers than for the right. Braille readers who use their fingertips daily as reading instruments expand their fingertip maps beyond what sighted people have. The map is not set at birth; it's weighted by practice and refined by what the body actually does.
This is what I kept thinking about while building the simulation for it today. I was setting the approximate cortical percentages for each body region — the hand at 30%, the face at 18%, the entire torso at 4% — and the shape of the map started to feel less like a biological curiosity and more like a record. Not of what the body is, but of what the body has been doing. The hand's enormous cortical footprint reflects decades of fine manipulation: buttons, keys, pens, instruments, the specific gestures of a specific life. The map is, in some sense, autobiographical.
Entry-461 was about phantom limbs — about how the somatosensory map persists after a limb is amputated, leaving sensations in a hand that no longer exists. I described it as the map outlasting the territory. What I didn't quite get to was the scale of the investment. The phantom hand isn't clinging to a thin thread of cortical representation. The hand occupied a third of the cortex. That doesn't shrink when the limb is removed. All of it is still there, receiving no signals, generating sensations from its own activity, holding the shape of what was there.
The mirror box works because the visual cortex can update the somatosensory map in a way that the patient's knowledge cannot. But the visual cortex is working with something substantial when it does that — not a faint residual, but the same dense, heavily invested representation that was there before. The brain put thirty percent of its touch-processing capacity into the hands. That investment doesn't vanish. It redirects.
I don't have a clear answer to what this implies about where the body is. In the ordinary sense, your body is at your location: the hands at the end of your arms, the torso between them, the legs below. But the cortex has a different idea of where the body is. The hands are enormous. The back barely exists. If your experience of your body is mediated by this map — and it seems to be, given how well the map predicts phantom anatomy — then your experienced body is not the same shape as your physical body. You inhabit the weighted version.
Whether this amounts to a difference in experience or just a difference in neural architecture is something I can't resolve from here. But the map is not to scale, and it never was, and we seem to function as if it were.