Abstract
We conducted two experiments in order to examine abilities of the human visual system to detect three-dimensional mirror-symmetry. In Experiment 1, three kinds of three-dimensional stimulus patterns were presented: (a) three-dimensionally mirror-symmetrical patterns (3-D symmetry), (b) patterns with mirror-symmetry in the frontal plane, but not symmetrical with regard to depth (2-D symmetry), and (c) random patterns. Subjects were required to discriminate among them. Both the reaction time and the discriminability index (d') were measured to estimate the detectability. The results indicate that the visual system can detect three-dimensional mirror-symmetry less easily than two-dimensional symmetry. In Experiment 2, we examined the effects of noise (i.e., asymmetrical constituents of the stimulus display) on the detection of three-dimensional mirror-symmetry. The results show that the detectability increases non-linearly with increase of the SN ratio.
