Abstract
Saccadic remapping may be an object-selective process (Wolfe and Whitney, VSS2014). Using the face aftereffect (FAE; Webster et al., 2004) in a saccadic paradigm, we showed that remapped information from a face that was never foveated could alter perception of a subsequently foveated face, suggesting that saccadic remapping may acquire object- rather than feature-level information. If remapping is object-selective, we reasoned that there should also be an inversion effect with our stimuli (Yin, 1969), and that our effect should be temporally tuned around the time of the saccade (Duhamel et al., 1992). In one experiment, we tested whether remapped face information exhibits an inversion effect. We presented an inverted emotional face (happy or sad) 15ยบ from fixation and asked subjects to saccade to it; it was removed from the display on saccade onset. Once the saccade terminated, an inverted morphed test face (ranging from happy to sad) was presented foveally and subjects judged whether the test face was happy or sad. Using the method of constant stimuli, we estimated the point of subjective equality as a measure of the FAE. Compared to the same experiment performed with upright pre-saccadic faces, we found a significantly reduced saccade-contingent FAE with inverted faces. We performed a separate experiment to determine the temporal tuning of saccadic remapping relative to saccade onset. Subjects were cued to saccade to a peripheral location, where an emotional face was presented for 50 ms at a random temporal offset after the cue and before the saccade. Subjects then judged a morphed test face at fixation after the saccade. We find that information is only remapped from the pre-saccadic face when it is presented 120 ms or less prior to the saccade. These new results support our account that saccadic remapping is object-selective and occurs immediately prior to the saccade.
Meeting abstract presented at VSS 2015