Past research has identified visual objects as the units of information processing in visual short-term memory (VSTM) and has shown that two features from the same object can be remembered in VSTM as well (or almost as well) as one feature of that object and are much better remembered than the same two features from two spatially separated objects. It is not clear, however, what drives this object benefit in VSTM. Is it the shared spatial location (proximity), the connectedness among features of an object, or both? In six change detection experiments, both location/proximity and connectedness were found to be crucial in determining the magnitude of the object benefit in VSTM. Together, these results indicate that location/proximity and connectedness are essential elements in defining a coherent visual object representation in VSTM.
A typical visual scene we encounter usually contains a fair number of objects. Although we can extract the gist of a scene quite rapidly (Potter, 1976), as has been illustrated in numerous change detection studies, our detailed representations of the objects in a scene in visual short-term memory (VSTM) are quite poor (e.g., Rensink, O'Regan, & Clark, 1997; see Simons & Levin, 1997, for a review). With simple stimuli, such as letters and colored squares, studies have documented that we can retain a maximum of about four items at a time in our VSTM (e.g., Irwin, 1992; Luck & Vogel, 1997; Pashler, 1988; see also Vogel, Woodman, & Luck, 2001). With the aid of long-term memory, however, more items may be remembered (Hollingworth, 2004; see also Hollingworth & Henderson, 2002; Hollingworth, Williams, & Henderson, 2001). Recently, possible neural mechanisms behind VSTM have been documented (Todd & Marois, 2004; Vogel & Machizawa, 2004; Xu & Chun, 2006).
Despite the capacity limitation of VSTM, two features from the same object have been found to be remembered as well (or almost as well) as one feature of that object and are much better remembered than the same two features located on two spatially separated objects. These object effects were first discovered with the brief report paradigm, in which observers are asked to report visual features from briefly presented and then masked visual objects (Allport, 1971; Duncan, 1984, 1993; Duncan & Nimmo-Smith, 1996; Wing & Allport, 1972; see also Irwin & Andrews, 1996). More recently, these findings have been replicated with the change detection paradigm, in which observers are presented briefly with multiple visual objects and, after a short delay, detect a visual feature change to one of the objects (Luck & Vogel, 1997; Olson & Jiang, 2002; Wheeler & Treisman, 2002; Xu, 2002a, 2002b). This object benefit, or object-based feature integration, in VSTM has led to the notion that the units of information processing in VSTM are discrete objects, rather than individual features, and that an object's features may be integrated in VSTM with minimum cost (Luck & Vogel, 1997; Vogel et al., 2001). It is worth emphasizing that the object benefit, or object-based feature integration, in VSTM increases the total amount of information that can be retained in VSTM, so that with the same number of features distributed in a similar spatial envelope, more feature information can be retained in VSTM when the features are grouped into a few discrete objects than when they are distributed over many objects (Xu, 2002a).
It is not clear, however, what drives the object benefit in VSTM. Given that the presence of an object always corresponds to a spatial location and that features from an object are always located at or close to the same location, it could simply be that VSTM can retain features only from a limited number of distinct spatial locations and that all features present at the same location will be remembered together with minimum cost. This grouping of features by spatial location in VSTM corresponds to the Gestalt principle of grouping by proximity (Wertheimer, 1924/1950), which states that nearby items are more likely to be grouped together than are distant items. …