Object recognition and sensitivity
“Explain, explain,” grumbled Étienne. “If you people can't name something you're incapable of seeing it.”— Cortázar, 1966, Hopscotch
How people recognize an object might seem trivial, because we do it so easily, but it has resisted all attempts to understand and explain it. Our past work has shown that letter identification begins with independent detection of features, and then integrates those features. We can say quite a bit about the feature detectors, and rather little about the feature integrator. Very briefly, the feature detectors are local in space and and spatial frequency, and have a tuning that scales in a nonlinear way with letter size. These results generalize to faces and line drawings of familiar objects. In a recent advance, “crowding,” a kind of lateral masking, turns out to be feature integration over an inappropriately large area, and applies to all tasks in the peripheral visual field that require feature integration. This result has led to several quick successes in addressing old problems and seems to offer a window into the workings of the feature integration process.