Abstract
Symmetry is a salient characteristic of visual stimuli that may be used to detect and recognize many natural and manufactured objects. Psychophysical studies have demonstrated that we are extremely fast and efficient at extracting symmetry, but the neural mechanisms underlying symmetry processing remain largely unknown. In one of very few ERP studies, Norcia et al. (JOV 2002) examined the time course of symmetry processing. In their experiment, observers were presented with symmetric and random dot patterns alternating every 500 ms. ERPs measured with symmetric-random sequences diverged from control random-random sequences between 130 and 220 ms after stimulus onset. However, it is not clear whether this response difference was due to symmetry per se or to the presence of structure in the symmetric dot patterns. We tested this possible confound by adding textured dot patterns (Glass patterns), which were asymmetric yet clearly distinguishable from random dot patterns, to the same experimental design. EEG was recorded from 256 electrodes while observers passively viewed alternating random-texture, random-symmetric, and random-random stimuli. Our preliminary results replicate and extend the findings of Norcia et al.: ERPs to structured dot textures were very similar to those evoked by random dots, and both begin to differ from ERPs to symmetric dots at about 160–170 ms after stimulus onset. We note that the onset of the symmetry effect is relatively late compared to the time course of object processing as described in the ERP literature. To further investigate this issue we will conduct another experiment in which observers actively discriminate between symmetric and non-symmetric patterns. This will allow us to relate the behavioral RTs with the onset of the symmetry ERP effects. Source analyses on individual subjects will also be conducted.
This work was supported by NSERC Discovery Grants 42133 and 105494 and the Canada Research Chair program.