The Effect of Familiarity and Novelty on Visual Preference Across Different Object and Scene Categories
L. Sha,M. Meng
DOI: https://doi.org/10.1167/11.11.840
2011-01-01
Journal of Vision
Abstract:With repetition, the visual preference for a synthetic human face increases, indicating a familiarity-driven effect; in contrast, the visual preference for a natural scene picture decreases, indicating a novelty-driven effect (Park, Shimojo, & Shimojo, 2010). However, the reason for this dichotomy of preference for faces and natural scenes remains unknown. We addressed this question by first testing whether the familiarity-driven preference is specific to faces, and whether the novelty-driven preference is specific to natural scenes. We examined the visual preference across nine different categories: natural scene pictures, Chinese landscape paintings, face pictures, face drawings, Beijing opera facemasks, pairings of human ears, houses, apples, and 1/f noise patterns. Pairings of human ears, initially devised by P. Sinha, are images of an individual's left and right ears pasted together (Gandhi et al., 2009). They share similar skin tone, skin textures, and biological symmetry with human faces. Our results suggest a significant novelty-driven preference for natural scene pictures, replicating Park et al.'s findings. Chinese landscape paintings were unaffected by repetitions, and the difference between natural scene pictures and Chinese landscape paintings was significant. Trends for a familiarity-driven preference were found for facemasks and face drawings, similar to Park et al.'s findings with synthetic faces. However, our face pictures failed to produce the familiarity-driven preference observed by Park et al. with synthetic faces. Interestingly, a significant familiarity-driven preference was found for ear pairings, with a significant difference between ear pairings and face pictures. Of the remaining categories, houses showed trends of a novelty-driven preference, while apples and 1/f noise patterns were unaffected by repetitions. Overall, our results suggest that the familiarity-driven visual preference is not face specific. The underlying mechanism of this preference awaits further investigations.