Abstract
Numerous studies have shown that visual short-term memory can store more task relevant features when they are integrated into a single object than when they appear as separate single-feature objects (Luck & Vogel, 1997; Olson & Jiang, 2002). For instance, four colors and four orientations can be encoded more efficiently when they appear as four colored and tilted bars than eight spatially separate single-feature objects. In contrast to these findings, mixed results have been found when features appear as different parts of an object, with very few experiments showing a clear performance benefit when features are organized as multi-part objects versus spatially dispersed single-feature objects (Davis & Holmes, 2005; Delvenne & Bruyer, 2004; Xu, 2002). Some researchers have suggested multi-part object integration is not mandatory due to the potential ambiguity of the display (Balaban & Luria, 2015; Luria and Vogel, 2014). For example, an oriented white bar across the middle of a red circle could be interpreted as two objects, a white bar occluding a red circle, or as a single two-colored object. We tested whether or not multi-part objects could be integrated in memory under less ambiguous conditions. Change detection performance for colored circles with an oriented white bar across the middle was compared to performance for colored spheres with an oriented white stripe across the middle. Luminance shading and linear perspective on the spheres indicated that the two features were part of one continuous surface on the object. These two conditions were also compared to a condition in which colored circles and orientated white bars were spatially separated and a condition in which the circles and bars partially overlapped. We found no performance differences between conditions suggesting multipart objects cannot be integrated in visual short-term memory even when they appear on a continuous surface and are unambiguously connected.
Meeting abstract presented at VSS 2016