- News Home
5 December 2013 11:26 am ,
Vol. 342 ,
An animal rights group known as the Nonhuman Rights Project filed lawsuits in three New York courts this week in an...
Researchers have been hot on the trail of the elusive Denisovans, a type of ancient human known only by their DNA and...
Thousands of scientists in the Russian Academy of Sciences (RAS) are about to lose their jobs as a result of the...
Dyslexia, a learning disability that hinders reading, hasn't been associated with deficits in vision, hearing, or...
Exotic, elusive, and dangerous, snakes have fascinated humankind for millennia. They can be hard to find, yet their...
Researchers have sequenced and analyzed the first two snake genomes, which represent two evolutionary extremes. The...
Snake venoms are remarkably complex mixtures that can stun or kill prey within minutes. But more and more researchers...
At age 30, Dutch biologist Freek Vonk has built up a respectable career as a snake scientist. But in his home country,...
- 5 December 2013 11:26 am , Vol. 342 , #6163
- About Us
Educated Guesses Gone Wrong
28 May 2002 (All day)
Tennis and golf players, at least the good ones, experience it with every shot: Their brains compute the velocity and direction of a flying ball with high precision in real-time. But under dim light, when the contrast is low, even the pros misjudge. Now theoretical neuroscientists show that such errors are not due to sloppy neuronal computation, but the result of educated--but erroneous--guesses as the brain struggles under difficult conditions.
Dim light can lead to some strange errors. In one classic example, a rhombus moving horizontally with the four corners covered appears to move vertically at low contrast (see link below with interactive animation). Similarly, car drivers underestimate their speed in foggy weather and tend to drive faster than on a sunny day. Cognitive neuroscientists have tried to exploit such observations to figure out how the brain perceives motion.
Theoretical neuroscientist Yair Weiss of the Hebrew University of Jerusalem now presents a simple theoretical model that explains most of the known illusions and misperceptions. The model, described in a paper published online 21 April by Nature Neuroscience, is based on a statistical theory that takes into account two basic assumptions about the real world: First, most objects in the world don't move at all, and the rest tend to move slowly. Second, it's hard to pin down the exact location of objects under low contrast. When Weiss fed the model numbers mimicking motion signals detected by the retina, he found that the model behaved surprisingly like a human: It came up with precise estimates of objects' motion in good "lighting" conditions and made the same mistakes when conditions were less optimal. Weiss says this suggests that the human brain uses the same assumptions to compute the motion of objects.
Cognitive neuroscientist Wilson Geisler of the University of Texas, Austin, is intrigued that illusions are not the result of "irrational" processes in the brain as some researchers had assumed. "Now it can simply be explained by a system that is making educated guesses," Geisler says. "That's what is so nice about it."