Abstract
The near exclusive use of forced choice tasks in object-substitution masking (OSM) paradigms has restricted their use to identifying coarse changes in target processing. To resolve this limitation the present study combines OSM paradigms with a delayed report paradigm from the working memory literature (Wilken & Ma, 2004; Zhang & Luck, 2007). This paradigm relies on a mixture model approach that allows for an estimate of the separable contributions of two components to task performance—one component being the precision of target representation, and the second being a guessing component. Participants estimated the gap location (0 to 360 degrees) of oriented Landolts identified by a four dot mask. In Experiment 1, using a standard OSM paradigm, both set size (2, 4, 8) and mask duration (0, 150, 300 ms) were varied in order to investigate their effects on estimates of the precision and guessing components. Estimates of guessing increased as a function of increases in both set size and mask durations. In contrast, estimates of precision were reduced only for increases in mask duration. In Experiment 2, we controlled for mask energy using a novel OSM paradigm described by Jannati, Spalek and Di Lollo (2013). This paradigm manipulates the interstimulus interval (ISI) between the initial display (both target and mask presented) and a following mask only display. Using this paradigm, effective masking has been shown for ISIs of less than 80 ms but not for larger ISIs. Results using this novel OSM paradigm converged with those from the standard OSM paradigm in showing that masking decreases precision and increases guessing. Using this mixture model approach which provides the ability to obtain a richer measure of object representation we found that OSM can produce not just perceptual effacement but also perceptual degradation of a target's representation.
Meeting abstract presented at VSS 2014