A running controversy among scientists studying visual perception centers on the neural basis for our powers to pick out figure from background. Look, for instance, at Edgar Rubin's famous image at the right. Do you see two faces or a vase? When you switch your attention, one or the other becomes clear. New research reported in today's Nature reveals that this kind of attention can actually allow people to distinguish between two objects at the same spot, even when they assume the same appearance at different times.
Erik Blaser and Zenon W. Pylyshyn of Rutgers University and Alex O. Holcombe of Harvard University superimposed two disk-shaped striped objects of the same size and asked observers to focus their attention on one of them. Each object kept changing constantly in three ways: its stripes turned, either clockwise or counterclockwise; its stripes increased or decreased in width; and the color of the stripes became more or less intense. They found that even when the objects didn't move relative to each other, volunteers were able to reliably tell them apart. In a second experiment, the changes of orientation, width and color of the two objects showed occasional simultaneous "jumps," which the observers were asked to report. The researchers found that whereas subjects were good at monitoring the nature of two jumps occurring in the same object, they couldn't keep track of one jump in each object.
The scientists conclude that rather than being able to focus on any two features ("feature-based attention"), the brain prefers to pay attention to several features of the same object at once ("object-based attention"). Thus, an object is defined not only by its movement in space but also by how its features change over time. "It remains to be seen exactly how attention can distinguish between objects represented by populations of neurons that are so intimately entwined," writes Jochen Braun of the University of Plymouth in an accompanying article. "But at the very least, the striking capabilities of visual attention revealed by Blaser et al. give us new reasons to think hard about how objects are represented in the visual cortex."