Brain scientists wonder less about twinkles in the sky

Brain scientists wonder less about twinkles in the sky
Neuroscientists have offered an explanation for an optical illusion that stumped Galileo Galilei, shown in this photograph of a painting attributed to Filippo di Nicola Furini. ("The Illustrated Longitude" by Dava Sobel and William J.H. Andrewes)

Neuroscientists may have figured out what's behind a visual trick that puzzled Galileo Galilei and stumped many others for centuries.

The answer to this trompe l'oeil also could explain why Mom and Dad always warned that it's bad to read in dim light.


Galileo noticed that bright planets looked far different through a telescope than with the naked eye, which gave them a blurry "crown" that made Venus seem far larger than Jupiter. The Italian astronomer brushed it off as a result of refraction, perhaps due to eye moisture.

Three centuries later, German polymath Hermann von Helmholtz took up where Galileo left off, mathematically chronicling an imbalance in how we perceive changes of luminance at the edge of a light object on a dark background versus changes far from the edges. But he remained puzzled at the physical cause of this disparity, which he dubbed the "irradiation illusion."

As telescopes became more accurate and moved beyond the visible light spectrum, the illusion didn't matter as much -- to astronomers, at least.

But the conundrum mattered to optometry and neuroscience. Researchers from State University of New York's College of Optometry have been delving into how we perceive light and dark objects for years. By 2008, their results were baffling them.

The team was examining how the brain's "on" neurons, which respond to light objects against dark backgrounds, differ from the "off" neurons that respond to dark objects on lighter backgrounds. They assumed that the "on" ones would predominate, since we are creatures of the lighted world who mostly close our eyes at night.

They were wrong. The "off" neurons, specialized for darker objects, were over-represented in the visual cortex, and their signaling path was faster. Were humans tuned for darkness?

“It was paradoxical to us,” said neuroscientist Jose-Manuel Alonso, principal author of a study published online Monday in Proceedings of the National Academy of Sciences. “Now it’s beginning to make sense. We live in a bright world. Most of our lives are in a brightly illuminated world. We are exposed to images that have a bright background.”

Objects important to our daily survival are dimmer than the average luminance of the scene -- much in the way these words contrast with their background, a preference that persists in the age when computers allow us to change that contrast.

By studying brain responses of house cats, monkeys and humans, the researchers found that "on" neurons have a more extreme reaction to small changes in illumination against dark backgrounds -- a biological marker for what Von Helmholtz described.

"When we talk about blur, we usually are talking about optics," said Alonso. "But there is another type of blur that is not in the optics but is in the response of the neurons themselves."

The "off" neurons respond to illumination changes in a steady, linear fashion, the study found. The responses of "on" neurons, however, are greater at the extremes, tracing the non-linear, curved line that von Helmholtz first described.

The linear response to darker shades on a light background makes spatial resolution finer and more reliable for objects in daylight -- an advantage for diurnal predatory animals such as humans, Alonso noted. A disproportionate response to a small twinkle in the dark instead offers a hair-trigger detection system for nocturnal stimuli that may be a threat, he noted. In other words, we don't need to know much about a faint flash in the dark to run away from it, but we need a lot of detail to hunt or gather.

Notably, there was no appreciable difference in this asymmetrical brain response among cats, which tend to be active in the dimmer backdrop of dawn and dusk, and the diurnal monkeys and humans, Alonso noted.
Reindeer, by the way, seem to have the best of both worlds. Their eyes adapt to the winter season of perpetual twilight in the Arctic through a change in the hue of their tapetum lucidum, a reflective surface behind the central retina that gives many animals better optical processing of light. That change, however, leaves them virtually glaucomic, with precious little resolution. No matter: They know enough to bolt from predators cloaked in the night.

Researchers suspect the source of our signaling imbalance could start with the first cells that convert light into biochemical electrical impulses. These photoreceptors respond to reductions in light, Alonso noted.

"Most of the receptors that you have for any other sense, for hearing or for touch, respond to an increase in stimulus," he said.

This brings us back to why you shouldn't read in dim light, a warning generally proffered by elders and based on vague anecdotal evidence. Alonso didn't fare much better when a student asked why this admonition endures among professionals.

"I didn't have a good answer," he said. "I asked my colleagues; no one had a good answer. This could be an answer: When you reduce the light, you are going to blur your visual representation, and blur, it's been shown, could drive eye growth."

If the eyes attempt to correct optically for what ultimately is a brain signaling imbalance, they could change in ways that are bad for vision in general, Alonso said.

"For us, where it starts is not as important as what the consequences are," Alonso said.