Abstract
When a briefly flashed target object is surrounded by four dots that all disappear at the same time, the target object can be easily discriminated. However, if the four dots remain visible after the target object disappears, discrimination is severely impaired. The object-substitution masking (OSM) account proposes that the representation of the target object is replaced by the representation of the four-dot mask (Di Lollo et al., 2000). Jiang and Chun (2001) have shown that a four-dot mask need not surround a target object to impair its discrimination but that OSM occurs even when the four-dot mask is displaced away from the target. The present study extends past work by testing whether the four-dot stimulus may mask targets in an object-based manner such that neighboring items that are perceptually grouped with a target may suffer from greater OSM. This hypothesis predicts that the four-dot masks will not only replace the target they surrounded but also neighboring items that are grouped with it via connectedness or color. However, across a series of experiments using color and connectedness as grouping cues the flanking objects that were grouped with the masked object were always reported more accurately than flanking items that were not grouped with the masked item. These findings indicate that attention spreads to grouped flankers but not ungrouped flankers prior to OSM, making the grouped object less vulnerable to OSM. However, the substitution masking of a target stimulus does not spread in an object-based manner.