Twenty-four visits to Stockholm: a concise history of the Rockefeller Nobel Prizes.

Part XVII: Torsten Wiesel, 1981 Prize in Physiology or Medicine.

By Joseph Luna

In the late 1950’s, two scientists sat with a cat in a darkened room and flicked on a projector screen. For this particular movie night with kitty, the scientists showed a series of simple images to the cat, and between each one they waited for the cat to respond. Nearly all cat owners, myself included, have probably performed a variant of this basic experiment, whether with a treat or a feathery toy, to get hold of a cat’s finicky attention, or to divert it from a precarious vase or an exposed ankle. But the two scientists, David Hubel and Torsten Wiesel, first at Johns Hopkins and then at Harvard, were after something much deeper. They wanted the cat to tell them what it saw. And magically enough, they had surgically created a talking cat: an electrode was inserted into the visual cortex of the anesthetized cat’s head and set up to record from a tiny patch of the brain (rest assured the cat was fine after the experiment). By showing different images to this conked-out kitty, Hubel and Weisel aimed to find the specific stimulus that excited the area they were recording from, be it a picture of a stationary dot or a simple line moving across the screen. If they succeeded at finding the right stimulus, they would hear the characteristic rat-tat-tat of a neuron firing. In other words, a tiny and specific part of the cat’s brain would seem to be saying “yup, that’s a line right there.”

How we perceive the outside world has been a central human question for millennia, underwriting large swathes of philosophy, and later, psychology and neuroscience. In the first half of the 20th century, technological developments aimed at measuring the electrical activity of a stimulated neuron in the brain yielded a concrete path to explore how organisms perceive their surroundings. Of the five most obvious senses, studying vision seemed particularly attractive since the input was physically always the same: photons. And yet photons could be arranged in wildly complex patterns to signal, in the case of a cat, the difference between a mouse and a shampoo bottle. How did light get transformed when it hit the eye into something “recognizable”?

This was a motivating question for a generation of scientists in the Department of Physiology at Johns Hopkins Medical School in the middle of the 20th century. And one such scientist was a young faculty member named Stephen Kuffler, who, in 1948, recorded from single cells in the cat retina and found that these cells did not signal absolute levels of light to the brain, but rather they transmitted the contrast information between light and dark. Small spots of light could activate retinal neurons, whereas flooding the eye with light didn’t do so. This finding largely confirmed in a mammal what a fellow soon-to-be Hopkins faculty member (and subject of this series) H. Keffer Hartline had seen while measuring the eye of the horseshoe crab over a decade earlier. Like Hartline, Kuffler could conclude that the “raw data” from light was passed to the brain as a code that essentially said, “this part is dark and this part is light”, but what happened after the retina was a mystery.

This is where Hubel and Weisel jumped in. Both had joined Kuffler’s lab at Hopkins to explore vision in the cat, picking up where Kuffler left off and using a technique that enabled recording from single cells in the brain. They started by repeating Kuffler’s observation, but this time measuring from the next stop in the brain after the retina, a relay station known as the lateral geniculate nucleus (LGN). Here they found a similar principle at work: the same small spots of light that activated the retina could be measured in the LGN. But when they moved onto the next and most complex stop in the cat’s visual cortex, small spots of light did not cause the neurons there to fire. Instead, the neurons in one spot would only fire if the cat saw a very particular abstraction of light, such as a line moving left or a box moving right. They discovered that cortical neurons weren’t really responding to light anymore but an abstract feature of that light. Moreover, a cortical neuron could be immensely precise in the type of stimulus that would activate it, such as a line at a 30-degree angle, but not 90 degrees. By taking measurements at different points of the cat’s visual cortex and looking for a stimulus that corresponded to each particular position, Hubel and Wiesel were able to map the pattern selectivity of the cells in this region. It was a first and stirring demonstration of what visual perception really looks like: after light enters the eyes, the signal that makes it to the visual cortex has been broken down into a constellation of lines, constituent shapes, and other features each computed at distinct locations, to yield a full picture.

It was a truly profound discovery, for it implied that visual perception is quite literally computed and created in the brain. In other words, it is an illusion. An important demonstration of this idea came from Hubel and Wiesel a few years later. By depriving one eye of a newborn kitten of light, while leaving the other alone, they observed that the kitten could become blind in the deprived eye, but not because there was something wrong with the actual eye. Instead they found that the region in the visual cortex responsible for that eye had failed to develop. Depriving an adult eye of light for the same amount of time never caused blindness. Hubel and Wiesel thus defined a “critical window” in brain development during which neural connections present at birth could be modified or even lost if deprived of their essential stimulus. This work directly influenced the improved treatment of children born with cataracts and other correctable eye conditions by highlighting the sense of urgency defined by the critical window.

It is somewhat hard to overstate how Hubel and Wiesel’s’ findings have shaped our perception of the brain as an exquisitely complex computer that creates the world around us. But along with a bit of feline help, this is exactly what they did.