Abstract
Olzak, Gabree, Saarela, and Laurinen (under review) performed center-surround masking experiments with suprathreshold sinusoidal gratings, and observed asymmetries in masking. When making judgments on the center component they showed that adding an abutting annulus (surround) mask decreased performance relative to a control task. However, when judgments were based on the surround, an abutting center mask had no effect on performance. Eye tracking experiments, using identical stimuli, indicated different locations of fixation for the two different judgments (center or surround). When making judgments on the surround observers fixated around the outer perimeter of the annulus, where as when making judgments on the center observers fixated on locations much more central to the stimulus. This different pattern in the results may be caused by a difference in the organization of the mask and test components. When performing discrimination tasks on the center the test stimulus is masked on all sides. Conversely, when performing discrimination tasks on the surround, the test stimulus is masked on only one side. To test whether the organization of components affected performance we conducted a masking experiment with rectangular sinusoidal stimuli. The test stimulus was a rectangular sinusoid masked on one side (top, bottom, left or right) by a constant abutting grating of the same size. This organization closely replicates the possible conditions under which surround judgments were made. In all participants, masking still occurred irrespective of the location of the abutting grating. These results indicate that the asymmetries that have previously been reported are not due to mask-stimulus organization, but may be due to the segmentation of components into independent objects.
Funded by NIH grants to L.A.O. and by NSERC and SFI grants to D.E.