Abstract
Modern theories of visual masking incorporate mechanisms that interefere with the consolidation of a target pattern into a conscious percept, and propose that feedback may be involved (Enns, 2004; Breitmeyer, 2007). To explore this, we conducted a series of experiments where we measured face identity discrimination thresholds in the central visual field under various masking conditions with face masks. In our first experiment (n=4), we examined masking as a function of SOA (stimulus onset asynchrony) in a standard backward masking paradigm, and compared it to common onset masking with a trailing mask. In this latter condition, the target and mask appeared at the same time, and after 33 ms, the target disappeared while the mask remained visible. In the SOA condition, peak masking occurred at an SOA of 58 ms, and in the trailing condition, masking equivalent to that of the peak SOA condition was found with a trailing mask duration of 58 ms and did not change as the trail was increased up to 600 ms. In Experiment 2, we tested seven observers in a modified trailing condition in which we briefly removed the masking stimulus for varying intervals, centered around the 58 ms point found to be critical in the previous SOA condition. When compared to an uninterrupted trail, we found no reduction in masking, with "mask gaps" as wide as 58 ms (p = 0.95). These results show that the effect of a trailing mask cannot be explained only by its presence at the critical SOA. We explore our findings in the framework of reverberant feedback loops.
Meeting abstract presented at VSS 2014