Abstract
A recently reported form of visual masking named object substitution masking (OSM) occurs when a sparse mask such as a group of four dots is presented simultaneously with the target (Enns & DiLollo, 1997). When the mask lingers in the display following target offset, the masking can occur. Some studies reported that the masking could be observed even when the target and mask were presented at the separate position and time, and implied that the masking could be regarded as a token individuation failure between the target and the mask. We introduced a new technique to couple attentive tracking and OSM to investigate the effect of the individuation between the target and mask under the control of the locus of attention. Observers attentively tracked a disk in an ambiguous motion display of six disks. After several frames, the discs changed into Landolt-Cs. Observers were required to report the gap direction of the Landolt-C that was surrounded by the four dots mask. When the target and mask offset simultaneously, observers could responded correctly even if it was not the tracked item. Whereas, when the mask lingered at the target position after the target offset, observers responded correctly only when the target was the tracked item. That is, OSM was found. We assessed the minimum target-mask SOA that could produce the masking effect. Our initial hypothesis was that this SOA should be close to the minimum SOA between the frames in the ambiguous motion display with which the observer could attentively track one item, where we thought the visual system could individuate tokens of stimuli. Indeed the masking effect reduced around 140ms-SOA that was close to the minimum SOA for the tracking (7Hz or 143 ms; Verstraten et al., 2000). These results support the target-mask individuation failure view of the OSM and suggest that it is necessary for the effect that the target and the lingering mask are presented within the spatio-temporal attentional resolution.