Abstract
Crowding refers to the difficulty in identifying an object when other objects are presented nearby. The phenomenon of "holistic crowding" (in which the recognition of an upright target face is more impaired by upright flanker faces than inverted ones), suggests that crowding may occur at the level of holistic face representations of upright faces rather than the low-level features in the images (Farzin, Rivera, & Whitney, 2009). Presently, we examined whether the flanker-inversion effect on crowding in face recognition can be observed with non-face flankers that are processed holistically. Visual expertise leads to increased holistic processing of non-face objects, so we opted to use expertise with Chinese characters as a means of studying the generality of holistic crowding. We hypothesized Chinese characters would induce holistic crowding of faces only in individuals with sufficient expertise. In Experiment 1, a target face was briefly presented in the fovea or the periphery. The target was either presented alone or surrounded by faces or Chinese characters that were presented upright or inverted. Non-Chinese speakers (N = 20) and native Chinese speakers (N = 19) indicated the sex of the target face. Our results demonstrated that categorization performance was worse when the target was surrounded by faces compared to Chinese characters in both groups. Also, native Chinese speakers showed a stronger crowding effect when the target was surrounded by upright compared to inverted Chinese characters (p = .035). However, the orientation of face flankers did not modulate the crowding effect in either group. In Experiment 2, we used the same experimental design with two-tone Mooney faces and obtained the same results. Our data suggest that visual expertise (and possibly holistic processing) affects crowding even when flankers and targets belong to separate categories; therefore, the high-level components of crowding may depend less on stimulus appearance and more on processing strategy.
Meeting abstract presented at VSS 2012