Generalized common fate: Grouping by common luminance changes

A critical step toward object recognition is the segmentation of a scene in to relevant regions. One of the most important cues for segmentation is tha t of common fate: Elements that move together are grouped together. Here we describe a new instantiation of common fate, in which elements move togeth er not through physical space, but through luminance space. Experiment I sh ows that when elements of a scene become brighter together or darker togeth er, observers group those elements together Experiments 2 and 3 show that t his effect is not due to the availability of fixed luminance differences be tween target and background regions, but requires common changes within eac h region in the direction of luminance over time. The effect is differentia ted from the recently discovered grouping cue of temporal synchrony, and is considered instead to be an extension of Wertheimer's original grouping fa ctor of common fate. Common fate for luminance, or generalized common fate, is an extremely strong cue for the segmentation of a scene, yielding a tre mendous advantage over grouping by fixed luminance cues.