Skip to content

In 1958 a cell in a cat's visual cortex went off like a machine gun as Hubel and Wiesel slid a glass slide into the ophthalmoscope. The dot was irrelevant. The cell wanted the faint straight shadow of the glass edge, at one narrow angle.

By Justin Brown

Torsten Wiesel photographed in Stockholm in 2011
Torsten Wiesel in Stockholm in 2011. Photo: Markus Marcetic for Young Academy of Sweden via Wikimedia Commons, CC BY-SA 4.0

In the spring of 1958, in a windowless basement room at the Wilmer Institute of Ophthalmology at the Johns Hopkins Hospital, a cell in an anaesthetized cat's visual cortex started firing hard at something nobody had meant to show it. The account comes from David Hubel's Nobel lecture, delivered on 8 December 1981 under the title "Evolution of Ideas on the Primary Visual Cortex, 1955-1978: A Biased Historical Account."

Here is the passage, in his words.

"We were inserting the glass slide with its black spot into the slot of the ophthalmoscope when suddenly over the audiomonitor the cell went off like a machine gun. After some fussing and fiddling we found out what was happening. The response had nothing to do with the black dot. As the glass slide was inserted its edge was casting onto the retina a faint but sharp shadow, a straight dark line on a light background. That was what the cell wanted, and it wanted it, moreover, in just one narrow range of orientations."

The dot was the experiment. The edge of the glass carrying it was the discovery.

What cortical cells actually did in an animal's daily life had no settled answer at the time, and that is the gap the accident walked into.

Hubel is direct about it in the same lecture. "This was unheard of. It is hard, now, to think back and realize just how free we were from any idea of what cortical cells might be doing in an animal's daily life." He goes on to list what was in circulation then and now looks strange to him, including suppressor strips, reverberating circuits and electrical field effects, the last of which Sperry had had to put to rest in 1955 by dicing up the cortex with mica plates and skewering it with tantalum wire.

The plan had been settled not long before, in a hospital cafeteria. "In the early spring of 1958 I drove over to Baltimore from Washington, D.C., and in a cafeteria at Johns Hopkins Hospital met Stephen Kuffler and Torsten Wiesel, for a discussion that was more momentous for Torsten's and my future than either of us could have possibly imagined," the lecture says.

Hubel had spent three years at the Walter Reed Army Institute of Research, where he developed a tungsten microelectrode and a technique for recording from chronically implanted cats, comparing the firing of cells in the visual pathways of sleeping and waking animals. A postdoctoral fellowship with Mountcastle at Hopkins was awkward to start, because Mountcastle was remodelling his laboratories.

Kuffler offered an interim place to work. "One day Kuffler called and asked if I would like to work in his laboratory at the Wilmer Institute of Ophthalmology at the Johns Hopkins Hospital with Torsten Wiesel, until the remodeling was completed," according to the lecture, which describes the room they were given as a dark and dingy inner windowless room of the Wilmer Institute basement, deemed ideal for visual studies. They ran their first experiment within a week of Hubel's arrival.

The plan was simple. They would extend Kuffler's retinal work into the brain and, in Hubel's phrasing, "map receptive fields with small spots, and look for any further processing of the visual information."

That strategy came out of what Kuffler had already found in the retina. He had described two types of retinal ganglion cell, "on-center" and "off-center", each with a receptive field made of two mutually antagonistic regions, a centre and a surround, one excitatory and the other inhibitory. The strategy of mapping with small spots seemed obvious, Hubel wrote.

The stimuli themselves were pieces of hardware. "Small spots of light were produced by sliding 2 cm X 5 cm metal rectangles containing various sizes of holes into a slot in the apparatus, just as one puts a slide into a slide projector," Hubel wrote. For the opposite contrast, the lecture says, "To obtain a black spot on a light background one used a piece of glass like a microscope slide, onto which a black dot had been glued."

So the glass slide in the famous story is a real object with a job. It was how they made a dark spot.

For roughly a month, the approach failed. They were still using the ophthalmoscope designed by Talbot and Kuffler, which projected a background light and a stimulus spot directly onto the retina while the experimenter looked in. "[T]he cells simply would not respond to our spots and annuli," the lecture records.

Then came an unusually stable recording, helped by a closed chamber over the cortex, built on Davies' idea and carried over from Hubel's chronic recording setup, which the 1959 paper says was used to minimize respiratory and vascular movements of the cortex. "For 3 or 4 hours we got absolutely nowhere," the lecture says, before vague and inconsistent responses began appearing from the midperiphery of the retina. Then the machine gun.

"The cell in question lasted 9 hours, and by the end we had a very different feeling about what the cortex might be doing," Hubel wrote.

They did not trust it at first. "It took us months to convince ourselves that we weren't at the mercy of some optical artifact, such as anyone can produce by squinting one's eyes and making vertical rays emanate from street lights," the lecture says. Recording several cells in sequence in the same penetration, each with a different optimal orientation, would, he wrote, have convinced anyone. By January they were ready to take the cells they thought they could understand and write them up, adding in the lecture that these were later called simple cells. The paper went through eleven complete reworkings.

It appeared as "Receptive fields of single neurones in the cat's striate cortex," received by the Journal of Physiology on 22 April 1959 and published in volume 148, pages 574 to 591.

The methods are stated plainly. Twenty-four cats were used, anaesthetized with intraperitoneal thiopental sodium at 40 mg/kg, with light anaesthesia maintained by further injections. The eyes were immobilized with intravenous succinylcholine, which made artificial respiration necessary, and the left retina was stimulated through the same Talbot and Kuffler multibeam ophthalmoscope. For binocular work the animal faced a screen. When the two methods were checked against each other on the same unit, the paper says, they gave similar results.

The sample matters, because the anecdote centres on one neuron and the published claim does not. "Several hundred units were recorded in the cat's striate cortex," the paper reports. "The findings to be described are based on thorough studies of forty-five of these, each of which was observed for a period of from 2 to 9 hr."

Most of those receptive fields split into excitatory and inhibitory regions arranged side by side, with a central area flanked by antagonistic areas, rather than in the concentric rings seen in the retina. In most fields, the paper notes, a stationary spot large enough to include the whole receptive field was entirely without effect.

One unit shows the orientation effect cleanly. There was no response to an 8 degree spot covering the whole receptive field. The same cell was strongly activated by a narrow slit measuring 1 degree by 8 degrees, oriented vertically over the excitatory region, while a horizontal slit of light was completely ineffective. As the optimum vertical orientation of the slit was approached, the paper says, responses appeared and rapidly increased to a maximum.

The paper makes the same point about an edge rather than a slit, for a different unit, in one sentence: "Such a boundary between light and darkness, when properly positioned and oriented, was often an effective type of stimulus."

The summary states the requirement plainly. Effective driving of a unit, the authors write, required a stimulus specific in form, size, position and orientation.

The caveats are in the paper itself. The authors say outright that "the sampling (45 units) was small, and may well be biased by the micro-electrode techniques," and that they may have failed to record from smaller cells or from units that lacked maintained activity. The cats were anaesthetized, and deepening the anaesthesia tended to suppress maintained activity, though the paper also points to similar findings in unanaesthetized, freely moving cats and says the poor response to large spots is probably not a result of anaesthesia.

Two things are not in this paper. The division into simple and complex cells belongs to the 1962 paper the lecture cites on receptive fields, binocular interaction and functional architecture, and the work establishing the shape and arrangement of orientation columns in the cat is a 1963 paper. The 1959 paper uses neither label as a category, though it notes that units with more complex properties were under study.

Hubel also cautioned against one reading of the result. Orientation-specific cells are specific for the direction of a short line segment, he wrote, adding that the cells are thus best not thought of as "line detectors" and are no more line detectors than they are curve detectors.

The date is loose. The lecture gives early spring 1958 for the cafeteria meeting that began the collaboration and places the discovery about a month in, with no exact day recorded, and it was delivered in 1981, twenty-three years after the day it describes. Retellings of the story sometimes put the slide in a projector rather than in the slot of the ophthalmoscope. The lecture is the first-hand account, and it is the version followed here.

What does not depend on anyone's memory sits in the 1959 figures. A spot filling a cell's whole receptive field often did nothing at all, and rotating a slit away from the cell's preferred angle reduced or abolished the response.

The wiring behind that preference was still open when Hubel collected the prize. In the same lecture that tells the machine gun story, he wrote that the actual circuit by which orientation specificity is derived from centre-surround cells is not known, and that the techniques necessary for solving it may still not be available.

Citation: D. H. Hubel and T. N. Wiesel, "Receptive fields of single neurones in the cat's striate cortex," Journal of Physiology 148, 574-591 (received 22 April 1959). First-person account: David H. Hubel, "Evolution of Ideas on the Primary Visual Cortex, 1955-1978: A Biased Historical Account," Nobel lecture, 8 December 1981.

Sources:

From around the web

Keep exploring