Chi T. K. Dao^1,4 (), Nihan Alp², Anthony M. Norcia³, Peter J. Kohler^1,4; ¹Department of Psychology, York University, ²Psychology Program, Sabanci University, Istanbul, ³Department of Psychology, Stanford University, ⁴Centre for Vision Research, York University

Symmetry and familiarity share important roles in visual perception. Both are detected rapidly and help the visual system resolve the structure of complex scenes. Yet it remains unclear how familiarity interacts with symmetry. Addressing this gap is essential for understanding how symmetry and familiarity jointly shape perceptual organization. Here, we investigate how familiarity influences symmetry responses, using high-density EEG (Study 1) and a psychophysical symmetry-detection task (Study 2). Object recognition depends strongly on configural shape—relationships between local curvatures around the contour. We used three classes of shape silhouettes that vary in configural shape and recognizability, but are well-matched in terms of local features: Intact animal silhouettes, silhouettes with disrupted configural shape (Frankensteins; Baker & Elder, 2022), and curvature-matched controls (CMCs; Elder et al., 2018), which share local curvature statistics with animal silhouettes but lack familiar configural shape. We tiled the shapes to create patterns with and without reflection symmetry. In Study 1 (N = 25), we recorded Steady-Visual Evoked Potentials (SSVEPs) while participants viewed a series of exemplars in which the two pattern types alternated, with each shape class shown on separate trials. The amplitude of symmetry-specific brain responses decreased systematically with familiarity: Lowest for intact, a bit higher for Frankensteins, and highest for CMCs. In Study 2 (N = 14), participants completed a 2IFC task with an adaptive staircase to measure symmetry-detection duration thresholds. Based on the SSVEP results, we predicted longer duration thresholds for more familiar shapes. Frankensteins and CMCs did show the expected pattern, with CMCs having significantly shorter durations (p=0.029), but thresholds for intact stimuli were intermediate, not significantly different from the other two conditions. Taken together, our results show that symmetry processing can be influenced by familiarity, but at a suprathreshold level that is not captured by our psychophysical measurements.

Acknowledgements: This research was undertaken thanks in part to funding from the Connected Minds Program and from the VISTA Program, both supported by the Canada First Research Excellence Fund. Additional support was provided by a Natural Sciences and Engineering Research Council of Canada Discovery Grant awarded to PJK.