Abstract
Crowding is under intense study due to its relation to object recognition, but little is known about its development during childhood. The commonly used technique for measuring critical spacing requires the report of a target's identity while it is surrounded by distracters. While this method has proven effective in adults, infants and other difficult-to-test populations are unable to report a target. Here we introduce change detection as a tool to overcome this obstacle in measuring critical spacing. We simultaneously present two targets, one on each side of fixation. Each target is flanked by two distracters. All the stimuli are flashed at 2 Hz for 3 s. During one of the many flashes, one of the two targets is changed to a different target. When distracters are placed at or beyond critical spacing, the change attracts the observer's attention and induces a measurable saccade. We vary the spacing between the target and the distracters to determine critical spacing with unconstrained viewing and eye tracking. In a separate test, observers fixate and press a key to report on which side of the screen the change occurred. To confirm that our method produces results comparable to the traditional method, the same observers participate in a standard crowding test. Our results with normal adults show that the eye tracking method accurately measures critical spacing (slightly less than one half eccentricity) without explicitly asking the observer to report.