Steven Pinker -- How the Mind Works - Hampshire High Italian ...

The Mind's Eye | 237pen vertically in front of you, with the clip facing away at eleven o'clock.When you close each eye in turn, you will notice that only the left eyecan see the clip; it is hidden from the right eye by the rest of the pen.Was natural selection as astute as Leonardo when it designed the brain,letting it use this valuable clue to an object's boundary? Or does thebrain ignore the clue, grudgingly chalking up each mismatch as anexception to the cohesive-matter assumption? The psychologists KenNakayama and Shinsuke Shimojo have shown that natural selection didnot ignore the clue. They created a random-dot stereogram whose depthinformation lay not in shifted dots but in dots that were visible in oneeye's view and absent in the other's. Those dots lay at the corners of animaginary square, with dots at the top and bottom right comers only inthe right eye's picture, and dots in the top and bottom left corners onlyin the left eye's picture. When people view the stereogram, they see afloating square defined by the four points, showing that the brain indeedinterprets features visible to only one eye as coming from an edge inspace. Nakayama and the psychologist Barton Anderson suggest thatthere arc neurons that detect these occlusions; they would respond to apair of marks in one eye, one of which can be matched with a mark inthe other eye and the other of which cannot be matched. These 3-Dboundary detectors would help a stereo network home in on the outlinesof the floating patches.Stereo vision does not come free with the two eyes; the circuitry has tobe wired into the brain. We know this because about two percent of thepopulation can see perfectly well out of each eyeball but not with thecyclopean eye; random-dot stereograms remain flat. Another four percentcan see stereo only poorly An even larger minority has more selectivedeficits. Some can't see stereo depth behind the point of fixation;others can't see it in front. Whitman Richards, who discovered theseforms of stereoblindness, hypothesized that the brain has three pools ofneurons that detect differences in the position of a spot in the two eyes.One pool is for pairs of spots that coincide exactly or almost exactly, forfine-grained depth perception at the point of focus. Another is for pairsof spots Hanking the nose, for farther objects. A third is for pairs of spotsapproaching the temples, for nearer objects. Neurons with all these